Add StatsAPI to README

* Change ccMetric->Influx functions

* Use a meta_as_tags string list in config but create a lookup map afterwards

* Add meta as tag logic to sampleSink

.github/workflows/AlmaLinux.yml

@@ -1,64 +0,0 @@
-# See: https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions
-
-# Workflow name
-name: AlmaLinux 8.5 RPM build
-
-# Run on tag push
-on:
- push:
-  tags:
-  - '**'
-
-jobs:
-
-  #
-  # Build on AlmaLinux 8.5 using go-toolset
-  #
-  AlmaLinux-RPM-build:
-    runs-on: ubuntu-latest
-    # See: https://hub.docker.com/_/almalinux
-    container: almalinux:8.5
-    steps:
-
-    # Use dnf to install development packages
-    - name: Install development packages
-      run: dnf --assumeyes group install "Development Tools" "RPM Development Tools"
-
-    # Checkout git repository and submodules
-    # fetch-depth must be 0 to use git describe
-    # See: https://github.com/marketplace/actions/checkout
-    - name: Checkout
-      uses: actions/checkout@v2
-      with:
-        submodules: recursive
-        fetch-depth: 0
-
-    # Use dnf to install build dependencies
-    - name: Install build dependencies
-      run: dnf --assumeyes builddep scripts/cc-metric-collector.spec
-
-    - name: RPM build MetricCollector
-      id: rpmbuild
-      run: make RPM
-
-    # See: https://github.com/actions/upload-artifact
-    - name: Save RPM as artifact
-      uses: actions/upload-artifact@v2
-      with:
-        name: cc-metric-collector RPM for AlmaLinux 8.5
-        path: ${{ steps.rpmbuild.outputs.RPM }}
-    - name: Save SRPM as artifact
-      uses: actions/upload-artifact@v2
-      with:
-        name: cc-metric-collector SRPM for AlmaLinux 8.5
-        path: ${{ steps.rpmbuild.outputs.SRPM }}
-
-    # See: https://github.com/softprops/action-gh-release
-    - name: Release
-      uses: softprops/action-gh-release@v1
-      if: startsWith(github.ref, 'refs/tags/')
-      with:
-        name: cc-metric-collector-${{github.ref_name}}
-        files: |
-         ${{ steps.rpmbuild.outputs.RPM }}
-         ${{ steps.rpmbuild.outputs.SRPM }}

.github/workflows/RedHatUniversalBaseImage.yml

@@ -1,64 +0,0 @@
-# See: https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions
-
-# Workflow name
-name: Red Hat Universal Base Image 8 RPM build
-
-# Run on tag push
-on:
- push:
-  tags:
-  - '**'
-
-jobs:
-
-  #
-  # Build on UBI 8 using go-toolset
-  #
-  UBI-8-RPM-build:
-    runs-on: ubuntu-latest
-    # See: https://catalog.redhat.com/software/containers/ubi8/ubi/5c359854d70cc534b3a3784e?container-tabs=gti
-    container: registry.access.redhat.com/ubi8/ubi:8.5-226.1645809065
-    steps:
-
-    # Use dnf to install development packages
-    - name: Install development packages
-      run: dnf --assumeyes --disableplugin=subscription-manager install rpm-build go-srpm-macros rpm-build-libs rpm-libs gcc make python38 git
-
-    # Checkout git repository and submodules
-    # fetch-depth must be 0 to use git describe
-    # See: https://github.com/marketplace/actions/checkout
-    - name: Checkout
-      uses: actions/checkout@v2
-      with:
-        submodules: recursive
-        fetch-depth: 0
-
-    # Use dnf to install build dependencies
-    - name: Install build dependencies
-      run: dnf --assumeyes --disableplugin=subscription-manager builddep scripts/cc-metric-collector.spec
-
-    - name: RPM build MetricCollector
-      id: rpmbuild
-      run: make RPM
-
-    # See: https://github.com/actions/upload-artifact
-    - name: Save RPM as artifact
-      uses: actions/upload-artifact@v2
-      with:
-        name: cc-metric-collector RPM for UBI 8
-        path: ${{ steps.rpmbuild.outputs.RPM }}
-    - name: Save SRPM as artifact
-      uses: actions/upload-artifact@v2
-      with:
-        name: cc-metric-collector SRPM for UBI 8
-        path: ${{ steps.rpmbuild.outputs.SRPM }}
-
-    # See: https://github.com/softprops/action-gh-release
-    - name: Release
-      uses: softprops/action-gh-release@v1
-      if: startsWith(github.ref, 'refs/tags/')
-      with:
-        name: cc-metric-collector-${{github.ref_name}}
-        files: |
-         ${{ steps.rpmbuild.outputs.RPM }}
-         ${{ steps.rpmbuild.outputs.SRPM }}

.github/workflows/Release.yml

@@ -0,0 +1,184 @@
+# See: https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions
+
+# Workflow name
+name: Release
+
+# Run on tag push
+on:
+ push:
+  tags:
+  - '**'
+
+jobs:
+
+  #
+  # Build on AlmaLinux 8.5 using go-toolset
+  #
+  AlmaLinux-RPM-build:
+    runs-on: ubuntu-latest
+    # See: https://hub.docker.com/_/almalinux
+    container: almalinux:8.5
+    # The job outputs link to the outputs of the 'rpmrename' step
+    # Only job outputs can be used in child jobs
+    outputs:
+      rpm : ${{steps.rpmrename.outputs.RPM}}
+      srpm : ${{steps.rpmrename.outputs.SRPM}}
+    steps:
+
+    # Use dnf to install development packages
+    - name: Install development packages
+      run: dnf --assumeyes group install "Development Tools" "RPM Development Tools"
+
+    # Checkout git repository and submodules
+    # fetch-depth must be 0 to use git describe
+    # See: https://github.com/marketplace/actions/checkout
+    - name: Checkout
+      uses: actions/checkout@v2
+      with:
+        submodules: recursive
+        fetch-depth: 0
+
+    # Use dnf to install build dependencies
+    - name: Install build dependencies
+      run: dnf --assumeyes builddep scripts/cc-metric-collector.spec
+
+    - name: RPM build MetricCollector
+      id: rpmbuild
+      run: make RPM
+    
+    # AlmaLinux 8.5 is a derivate of RedHat Enterprise Linux 8 (UBI8),
+    # so the created RPM both contain the substring 'el8' in the RPM file names
+    # This step replaces the substring 'el8' to 'alma85'. It uses the move operation
+    # because it is unclear whether the default AlmaLinux 8.5 container contains the 
+    # 'rename' command. This way we also get the new names for output.
+    - name: Rename RPMs (s/el8/alma85/)
+      id: rpmrename
+      run: |
+        OLD_RPM="${{steps.rpmbuild.outputs.RPM}}"
+        OLD_SRPM="${{steps.rpmbuild.outputs.SRPM}}"
+        NEW_RPM="${OLD_RPM/el8/alma85}"
+        NEW_SRPM=${OLD_SRPM/el8/alma85}
+        mv "${OLD_RPM}" "${NEW_RPM}"
+        mv "${OLD_SRPM}" "${NEW_SRPM}"
+        echo "::set-output name=SRPM::${NEW_SRPM}"
+        echo "::set-output name=RPM::${NEW_RPM}"
+
+    # See: https://github.com/actions/upload-artifact
+    - name: Save RPM as artifact
+      uses: actions/upload-artifact@v2
+      with:
+        name: cc-metric-collector RPM for AlmaLinux 8.5
+        path: ${{ steps.rpmrename.outputs.RPM }}
+    - name: Save SRPM as artifact
+      uses: actions/upload-artifact@v2
+      with:
+        name: cc-metric-collector SRPM for AlmaLinux 8.5
+        path: ${{ steps.rpmrename.outputs.SRPM }}
+
+  #
+  # Build on UBI 8 using go-toolset
+  #
+  UBI-8-RPM-build:
+    runs-on: ubuntu-latest
+    # See: https://catalog.redhat.com/software/containers/ubi8/ubi/5c359854d70cc534b3a3784e?container-tabs=gti
+    container: registry.access.redhat.com/ubi8/ubi:8.5-226.1645809065
+    # The job outputs link to the outputs of the 'rpmbuild' step
+    outputs:
+      rpm : ${{steps.rpmbuild.outputs.RPM}}
+      srpm : ${{steps.rpmbuild.outputs.SRPM}}
+    steps:
+
+    # Use dnf to install development packages
+    - name: Install development packages
+      run: dnf --assumeyes --disableplugin=subscription-manager install rpm-build go-srpm-macros rpm-build-libs rpm-libs gcc make python38 git
+
+    # Checkout git repository and submodules
+    # fetch-depth must be 0 to use git describe
+    # See: https://github.com/marketplace/actions/checkout
+    - name: Checkout
+      uses: actions/checkout@v2
+      with:
+        submodules: recursive
+        fetch-depth: 0
+
+    # Use dnf to install build dependencies
+    - name: Install build dependencies
+      run: dnf --assumeyes --disableplugin=subscription-manager builddep scripts/cc-metric-collector.spec
+
+    - name: RPM build MetricCollector
+      id: rpmbuild
+      run: make RPM
+
+    # See: https://github.com/actions/upload-artifact
+    - name: Save RPM as artifact
+      uses: actions/upload-artifact@v2
+      with:
+        name: cc-metric-collector RPM for UBI 8
+        path: ${{ steps.rpmbuild.outputs.RPM }}
+    - name: Save SRPM as artifact
+      uses: actions/upload-artifact@v2
+      with:
+        name: cc-metric-collector SRPM for UBI 8
+        path: ${{ steps.rpmbuild.outputs.SRPM }}
+
+  #
+  # Create release with fresh RPMs
+  #
+  Release:
+    runs-on: ubuntu-latest
+    # We need the RPMs, so add dependency
+    needs: [AlmaLinux-RPM-build, UBI-8-RPM-build]
+
+    steps:
+    # See: https://github.com/actions/download-artifact
+    - name: Download AlmaLinux 8.5 RPM
+      uses: actions/download-artifact@v2
+      with:
+        name: cc-metric-collector RPM for AlmaLinux 8.5
+    - name: Download AlmaLinux 8.5 SRPM
+      uses: actions/download-artifact@v2
+      with:
+        name: cc-metric-collector SRPM for AlmaLinux 8.5
+    
+    - name: Download UBI 8 RPM
+      uses: actions/download-artifact@v2
+      with:
+        name: cc-metric-collector RPM for UBI 8
+    - name: Download UBI 8 SRPM
+      uses: actions/download-artifact@v2
+      with:
+        name: cc-metric-collector SRPM for UBI 8
+
+    # The download actions do not publish the name of the downloaded file,
+    # so we re-use the job outputs of the parent jobs. The files are all
+    # downloaded to the current folder.
+    # The gh-release action afterwards does not accept file lists but all
+    # files have to be listed at 'files'. The step creates one output per
+    # RPM package (2 per distro)
+    - name: Set RPM variables
+      id: files
+      run: |
+        ALMA_85_RPM=$(basename "${{ needs.AlmaLinux-RPM-build.outputs.rpm}}")
+        ALMA_85_SRPM=$(basename "${{ needs.AlmaLinux-RPM-build.outputs.srpm}}")
+        UBI_8_RPM=$(basename "${{ needs.UBI-8-RPM-build.outputs.rpm}}")
+        UBI_8_SRPM=$(basename "${{ needs.UBI-8-RPM-build.outputs.srpm}}")
+        echo "ALMA_85_RPM::${ALMA_85_RPM}"
+        echo "ALMA_85_SRPM::${ALMA_85_SRPM}"
+        echo "UBI_8_RPM::${UBI_8_RPM}"
+        echo "UBI_8_SRPM::${UBI_8_SRPM}"
+        echo "::set-output name=ALMA_85_RPM::${ALMA_85_RPM}"
+        echo "::set-output name=ALMA_85_SRPM::${ALMA_85_SRPM}"
+        echo "::set-output name=UBI_8_RPM::${UBI_8_RPM}"
+        echo "::set-output name=UBI_8_SRPM::${UBI_8_SRPM}"
+
+    # See: https://github.com/softprops/action-gh-release
+    - name: Release
+      uses: softprops/action-gh-release@v1
+      if: startsWith(github.ref, 'refs/tags/')
+      with:
+        name: cc-metric-collector-${{github.ref_name}}
+        files: |
+         ${{ steps.files.outputs.ALMA_85_RPM }}
+         ${{ steps.files.outputs.ALMA_85_SRPM }}
+         ${{ steps.files.outputs.UBI_8_RPM }}
+         ${{ steps.files.outputs.UBI_8_SRPM }}

Makefile

@@ -15,6 +15,8 @@ COMPONENT_DIRS   := collectors \
 			internal/ccTopology \
 			internal/multiChanTicker
 
+BINDIR = bin
+
 
 .PHONY: all
 all: $(APP)
@@ -24,15 +26,27 @@ $(APP): $(GOSRC)
 	go get
 	go build -o $(APP) $(GOSRC_APP)
 
+install: $(APP)
+	@WORKSPACE=$(PREFIX)
+	@if [ -z "$${WORKSPACE}" ]; then exit 1; fi
+	@mkdir --parents --verbose $${WORKSPACE}/usr/$(BINDIR)
+	@install -Dpm 755 $(APP) $${WORKSPACE}/usr/$(BINDIR)/$(APP)
+	@mkdir --parents --verbose $${WORKSPACE}/etc/cc-metric-collector $${WORKSPACE}/etc/default $${WORKSPACE}/etc/systemd/system $${WORKSPACE}/etc/init.d
+	@install -Dpm 600 config.json $${WORKSPACE}/etc/cc-metric-collector/cc-metric-collector.json
+	@sed -i -e s+"\"./"+"\"/etc/cc-metric-collector/"+g $${WORKSPACE}/etc/cc-metric-collector/cc-metric-collector.json
+	@install -Dpm 600 sinks.json $${WORKSPACE}/etc/cc-metric-collector/sinks.json
+	@install -Dpm 600 collectors.json $${WORKSPACE}/etc/cc-metric-collector/collectors.json
+	@install -Dpm 600 router.json $${WORKSPACE}/etc/cc-metric-collector/router.json
+	@install -Dpm 600 receivers.json $${WORKSPACE}/etc/cc-metric-collector/receivers.json
+	@install -Dpm 600 scripts/cc-metric-collector.config $${WORKSPACE}/etc/default/cc-metric-collector
+	@install -Dpm 644 scripts/cc-metric-collector.service $${WORKSPACE}/etc/systemd/system/cc-metric-collector.service
+	@install -Dpm 644 scripts/cc-metric-collector.init $${WORKSPACE}/etc/init.d/cc-metric-collector
+
+
 .PHONY: clean
 .ONESHELL:
 clean:
-	@for COMP in $(COMPONENT_DIRS)
-	do
-	    if [[ -e $$COMP/Makefile ]]; then
-	        make -C $$COMP clean
-	    fi
-	done
+	@for COMP in $(COMPONENT_DIRS); do if [ -e $$COMP/Makefile ]; then make -C $$COMP clean;  fi; done
 	rm -f $(APP)
 
 .PHONY: fmt
@@ -69,7 +83,7 @@ RPM: scripts/cc-metric-collector.spec
 	@COMMITISH="HEAD"
 	@VERS=$$(git describe --tags $${COMMITISH})
 	@VERS=$${VERS#v}
-	@VERS=$${VERS//-/_}
+	@VERS=$$(echo $$VERS | sed -e s+'-'+'_'+g)
 	@eval $$(rpmspec --query --queryformat "NAME='%{name}' VERSION='%{version}' RELEASE='%{release}' NVR='%{NVR}' NVRA='%{NVRA}'" --define="VERS $${VERS}" "$${SPECFILE}")
 	@PREFIX="$${NAME}-$${VERSION}"
 	@FORMAT="tar.gz"
@@ -86,3 +100,28 @@ RPM: scripts/cc-metric-collector.spec
 	@     echo "::set-output name=SRPM::$${SRPMFILE}"
 	@     echo "::set-output name=RPM::$${RPMFILE}"
 	@fi
+
+.PHONY: DEB
+DEB: scripts/cc-metric-collector.deb.control $(APP)
+	@BASEDIR=$${PWD}
+	@WORKSPACE=$${PWD}/.dpkgbuild
+	@DEBIANDIR=$${WORKSPACE}/debian
+	@DEBIANBINDIR=$${WORKSPACE}/DEBIAN
+	@mkdir --parents --verbose $$WORKSPACE $$DEBIANBINDIR
+	#@mkdir --parents --verbose $$DEBIANDIR
+	@CONTROLFILE="$${BASEDIR}/scripts/cc-metric-collector.deb.control"
+	@COMMITISH="HEAD"
+	@VERS=$$(git describe --tags --abbrev=0 $${COMMITISH})
+	@VERS=$${VERS#v}
+	@VERS=$$(echo $$VERS | sed -e s+'-'+'_'+g)
+	@ARCH=$$(uname -m)
+	@ARCH=$$(echo $$ARCH | sed -e s+'_'+'-'+g)
+	@PREFIX="$${NAME}-$${VERSION}_$${ARCH}"
+	@SIZE_BYTES=$$(du -bcs --exclude=.dpkgbuild "$$WORKSPACE"/ | awk '{print $$1}' | head -1 | sed -e 's/^0\+//')
+	@SIZE="$$(awk -v size="$$SIZE_BYTES" 'BEGIN {print (size/1024)+1}' | awk '{print int($$0)}')"
+	#@sed -e s+"{VERSION}"+"$$VERS"+g -e s+"{INSTALLED_SIZE}"+"$$SIZE"+g -e s+"{ARCH}"+"$$ARCH"+g $$CONTROLFILE > $${DEBIANDIR}/control
+	@sed -e s+"{VERSION}"+"$$VERS"+g -e s+"{INSTALLED_SIZE}"+"$$SIZE"+g -e s+"{ARCH}"+"$$ARCH"+g $$CONTROLFILE > $${DEBIANBINDIR}/control
+	@make PREFIX=$${WORKSPACE} install
+	@DEB_FILE="cc-metric-collector_$${VERS}_$${ARCH}.deb"
+	@dpkg-deb -b $${WORKSPACE} "$$DEB_FILE"
+	@rm -r "$${WORKSPACE}"

README.md

@@ -20,6 +20,7 @@ There is a main configuration file with basic settings that point to the other c
   "collectors" : "collectors.json",
   "receivers" : "receivers.json",
   "router" : "router.json",
+  "stats_api" : "api.json",
   "interval": 10,
   "duration": 1
 }
@@ -32,6 +33,7 @@ See the component READMEs for their configuration:
 * [`sinks`](./sinks/README.md)
 * [`receivers`](./receivers/README.md)
 * [`router`](./internal/metricRouter/README.md)
+* [`stats_api`](./internal/metricRouter/StatsApi.md)
 
 
 # Installation

cc-metric-collector.go

@@ -28,6 +28,7 @@ type CentralConfigFile struct {
 	RouterConfigFile    string `json:"router"`
 	SinkConfigFile      string `json:"sinks"`
 	ReceiverConfigFile  string `json:"receivers,omitempty"`
+	StatsApiConfigFile  string `json:"stats_api,omitempty"`
 }
 
 func LoadCentralConfiguration(file string, config *CentralConfigFile) error {
@@ -52,6 +53,7 @@ type RuntimeConfig struct {
 	CollectManager  collectors.CollectorManager
 	SinkManager     sinks.SinkManager
 	ReceiveManager  receivers.ReceiveManager
+	StatsApi        mr.StatsApi
 	MultiChanTicker mct.MultiChanTicker
 
 	Channels []chan lp.CCMetric
@@ -152,11 +154,16 @@ func shutdownHandler(config *RuntimeConfig, shutdownSignal chan os.Signal) {
 		cclog.Debug("Shutdown SinkManager...")
 		config.SinkManager.Close()
 	}
+	if config.StatsApi != nil {
+		cclog.Debug("Shutdown StatsApi...")
+		config.StatsApi.Close()
+	}
 }
 
 func mainFunc() int {
 	var err error
 	use_recv := false
+	use_api := false
 
 	// Initialize runtime configuration
 	rcfg := RuntimeConfig{
@@ -164,6 +171,7 @@ func mainFunc() int {
 		CollectManager: nil,
 		SinkManager:    nil,
 		ReceiveManager: nil,
+		StatsApi:       nil,
 		CliArgs:        ReadCli(),
 	}
 
@@ -253,6 +261,16 @@ func mainFunc() int {
 		use_recv = true
 	}
 
+	// Create new statistics API manager
+	if len(rcfg.ConfigFile.StatsApiConfigFile) > 0 {
+		rcfg.StatsApi, err = mr.NewStatsApi(rcfg.MultiChanTicker, &rcfg.Sync, rcfg.ConfigFile.StatsApiConfigFile)
+		if err != nil {
+			cclog.Error(err.Error())
+			return 1
+		}
+		use_api = true
+	}
+
 	// Create shutdown handler
 	shutdownSignal := make(chan os.Signal, 1)
 	signal.Notify(shutdownSignal, os.Interrupt)
@@ -260,6 +278,11 @@ func mainFunc() int {
 	rcfg.Sync.Add(1)
 	go shutdownHandler(&rcfg, shutdownSignal)
 
+	// Start the stats api early to be prepared for init settings
+	if use_api {
+		rcfg.StatsApi.Start()
+	}
+
 	// Start the managers
 	rcfg.MetricRouter.Start()
 	rcfg.SinkManager.Start()

collectors.json

@@ -12,6 +12,12 @@
             "proc_total"
         ]
     },
+    "netstat": {
+        "include_devices": [
+            "enp5s0"
+        ],
+        "send_derived_values": true
+    },
     "numastats": {},
     "nvidia": {},
     "tempstat": {

collectors/Makefile

@@ -1,79 +1,25 @@
-# Use central installation
-CENTRAL_INSTALL = false
-# How to access hardware performance counters through LIKWID.
-# Recommended is 'direct' mode
-ACCESSMODE = direct
 
-#######################################################################
-# if CENTRAL_INSTALL == true
-#######################################################################
-# Path to central installation (if CENTRAL_INSTALL=true)
-LIKWID_BASE=/apps/likwid/5.2.1
-# LIKWID version (should be same major version as central installation, 5.2.x)
+all: likwid
+
+
+# LIKWID version
 LIKWID_VERSION = 5.2.1
 
-#######################################################################
-# if CENTRAL_INSTALL == false and ACCESSMODE == accessdaemon
-#######################################################################
-# Where to install the accessdaemon
-DAEMON_INSTALLDIR = /usr/local
-# Which user to use for the accessdaemon
-DAEMON_USER = root
-# Which group to use for the accessdaemon
-DAEMON_GROUP = root
+.ONESHELL:
+.PHONY: likwid
+likwid:
+	INSTALL_FOLDER="$${PWD}/likwid"
+	BUILD_FOLDER="$${PWD}/likwidbuild"
+	if [ -d $${INSTALL_FOLDER} ]; then rm -r $${INSTALL_FOLDER}; fi
+	mkdir --parents --verbose  $${INSTALL_FOLDER} $${BUILD_FOLDER}
+	wget -P "$${BUILD_FOLDER}" ftp://ftp.rrze.uni-erlangen.de/mirrors/likwid/likwid-$(LIKWID_VERSION).tar.gz
+	tar -C $${BUILD_FOLDER} -xf $${BUILD_FOLDER}/likwid-$(LIKWID_VERSION).tar.gz
+	install -Dpm 0644 $${BUILD_FOLDER}/likwid-$(LIKWID_VERSION)/src/includes/likwid*.h $${INSTALL_FOLDER}/
+	install -Dpm 0644 $${BUILD_FOLDER}/likwid-$(LIKWID_VERSION)/src/includes/bstrlib.h $${INSTALL_FOLDER}/
+	rm -r $${BUILD_FOLDER}
 
 
-
-#################################################
-# No need to change anything below this line
-#################################################
-INSTALL_FOLDER = ./likwid
-BUILD_FOLDER = ./likwid/build
-
-ifneq ($(strip $(CENTRAL_INSTALL)),true)
-LIKWID_BASE := $(shell pwd)/$(INSTALL_FOLDER)
-DAEMON_BASE := $(LIKWID_BASE)
-GROUPS_BASE := $(LIKWID_BASE)/groups
-all: $(INSTALL_FOLDER)/liblikwid.a cleanup
-else
-DAEMON_BASE= $(LIKWID_BASE)/sbin
-all: $(INSTALL_FOLDER)/liblikwid.a cleanup
-endif
-
-
-
-$(BUILD_FOLDER)/likwid-$(LIKWID_VERSION).tar.gz: $(BUILD_FOLDER)
-	wget -P $(BUILD_FOLDER) ftp://ftp.rrze.uni-erlangen.de/mirrors/likwid/likwid-$(LIKWID_VERSION).tar.gz
-
-$(BUILD_FOLDER):
-	mkdir -p $(BUILD_FOLDER)
-
-$(INSTALL_FOLDER):
-	mkdir -p $(INSTALL_FOLDER)
-
-$(BUILD_FOLDER)/likwid-$(LIKWID_VERSION): $(BUILD_FOLDER)/likwid-$(LIKWID_VERSION).tar.gz
-	tar -C $(BUILD_FOLDER) -xf $(BUILD_FOLDER)/likwid-$(LIKWID_VERSION).tar.gz
-
-$(INSTALL_FOLDER)/liblikwid.a: $(BUILD_FOLDER)/likwid-$(LIKWID_VERSION) $(INSTALL_FOLDER)
-	cd "$(BUILD_FOLDER)/likwid-$(LIKWID_VERSION)" && make "PREFIX=$(LIKWID_BASE)" "SHARED_LIBRARY=false" "ACCESSMODE=$(ACCESSMODE)" "INSTALLED_ACCESSDAEMON=$(DAEMON_INSTALLDIR)/likwid-accessD"
-	cp \
-	    $(BUILD_FOLDER)/likwid-$(LIKWID_VERSION)/liblikwid.a \
-	    $(BUILD_FOLDER)/likwid-$(LIKWID_VERSION)/ext/hwloc/liblikwid-hwloc.a \
-	    $(BUILD_FOLDER)/likwid-$(LIKWID_VERSION)/src/includes/likwid*.h \
-	    $(BUILD_FOLDER)/likwid-$(LIKWID_VERSION)/src/includes/bstrlib.h \
-	    $(INSTALL_FOLDER)
-
-$(DAEMON_INSTALLDIR)/likwid-accessD: $(BUILD_FOLDER)/likwid-$(LIKWID_VERSION)/likwid-accessD
-	sudo -u $(DAEMON_USER) -g $(DAEMON_GROUP) install -m 4775 $(BUILD_FOLDER)/likwid-$(LIKWID_VERSION)/likwid-accessD $(DAEMON_INSTALLDIR)/likwid-accessD
-
-prepare_collector: likwidMetric.go
-	cp likwidMetric.go likwidMetric.go.orig
-	sed -i -e s+"const GROUPPATH =.*"+"const GROUPPATH = \`$(GROUPS_BASE)\`"+g likwidMetric.go
-
-cleanup:
-	rm -rf $(BUILD_FOLDER)
-
-clean: cleanup
+clean:
 	rm -rf likwid
 
 .PHONY: clean

collectors/README.md

@@ -37,6 +37,8 @@ In contrast to the configuration files for sinks and receivers, the collectors c
 * [`cpufreq_cpuinfo`](./cpufreqCpuinfoMetric.md)
 * [`numastat`](./numastatMetric.md)
 * [`gpfs`](./gpfsMetric.md)
+* [`beegfs_meta`](./beegfsmetaMetric.md)
+* [`beegfs_storage`](./beegfsstorageMetric.md)
 
 ## Todos
 

collectors/beegfsmetaMetric.go

@@ -0,0 +1,234 @@
+package collectors
+
+import (
+	"bufio"
+	"bytes"
+	"encoding/json"
+	"fmt"
+	"io/ioutil"
+	"os"
+	"os/exec"
+	"os/user"
+	"regexp"
+	"strconv"
+	"strings"
+	"time"
+
+	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
+	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
+)
+
+const DEFAULT_BEEGFS_CMD = "beegfs-ctl"
+
+// Struct for the collector-specific JSON config
+type BeegfsMetaCollectorConfig struct {
+	Beegfs            string   `json:"beegfs_path"`
+	ExcludeMetrics    []string `json:"exclude_metrics,omitempty"`
+	ExcludeFilesystem []string `json:"exclude_filesystem"`
+}
+
+type BeegfsMetaCollector struct {
+	metricCollector
+	tags                  map[string]string
+	matches               map[string]string
+	config                BeegfsMetaCollectorConfig
+	skipFS                map[string]struct{}
+	statsProcessedMetrics int64
+}
+
+func (m *BeegfsMetaCollector) Init(config json.RawMessage) error {
+	// Check if already initialized
+	if m.init {
+		return nil
+	}
+	// Metrics
+	var nodeMdstat_array = [39]string{
+		"sum", "ack", "close", "entInf",
+		"fndOwn", "mkdir", "create", "rddir",
+		"refrEn", "mdsInf", "rmdir", "rmLnk",
+		"mvDirIns", "mvFiIns", "open", "ren",
+		"sChDrct", "sAttr", "sDirPat", "stat",
+		"statfs", "trunc", "symlnk", "unlnk",
+		"lookLI", "statLI", "revalLI", "openLI",
+		"createLI", "hardlnk", "flckAp", "flckEn",
+		"flckRg", "dirparent", "listXA", "getXA",
+		"rmXA", "setXA", "mirror"}
+
+	m.name = "BeegfsMetaCollector"
+	m.setup()
+	// Set default beegfs-ctl binary
+
+	m.config.Beegfs = DEFAULT_BEEGFS_CMD
+
+	// Read JSON configuration
+	if len(config) > 0 {
+		err := json.Unmarshal(config, &m.config)
+		if err != nil {
+			return err
+		}
+	}
+
+	//create map with possible variables
+	m.matches = make(map[string]string)
+	for _, value := range nodeMdstat_array {
+		_, skip := stringArrayContains(m.config.ExcludeMetrics, value)
+		if skip {
+			m.matches["other"] = "0"
+		} else {
+			m.matches["beegfs_cmeta_"+value] = "0"
+		}
+	}
+
+	m.meta = map[string]string{
+		"source": m.name,
+		"group":  "BeegfsMeta",
+	}
+	m.tags = map[string]string{
+		"type":       "node",
+		"filesystem": "",
+	}
+	m.skipFS = make(map[string]struct{})
+	for _, fs := range m.config.ExcludeFilesystem {
+		m.skipFS[fs] = struct{}{}
+	}
+
+	// Beegfs file system statistics can only be queried by user root
+	user, err := user.Current()
+	if err != nil {
+		return fmt.Errorf("BeegfsMetaCollector.Init(): Failed to get current user: %v", err)
+	}
+	if user.Uid != "0" {
+		return fmt.Errorf("BeegfsMetaCollector.Init(): BeeGFS file system statistics can only be queried by user root")
+	}
+
+	// Check if beegfs-ctl is in executable search path
+	_, err = exec.LookPath(m.config.Beegfs)
+	if err != nil {
+		return fmt.Errorf("BeegfsMetaCollector.Init(): Failed to find beegfs-ctl binary '%s': %v", m.config.Beegfs, err)
+	}
+	m.statsProcessedMetrics = 0
+	m.init = true
+	return nil
+}
+
+func (m *BeegfsMetaCollector) Read(interval time.Duration, output chan lp.CCMetric) {
+	if !m.init {
+		return
+	}
+	//get mounpoint
+	buffer, _ := ioutil.ReadFile(string("/proc/mounts"))
+	mounts := strings.Split(string(buffer), "\n")
+	var mountpoints []string
+	for _, line := range mounts {
+		if len(line) == 0 {
+			continue
+		}
+		f := strings.Fields(line)
+		if strings.Contains(f[0], "beegfs_ondemand") {
+			// Skip excluded filesystems
+			if _, skip := m.skipFS[f[1]]; skip {
+				continue
+			}
+			mountpoints = append(mountpoints, f[1])
+		}
+	}
+
+	if len(mountpoints) == 0 {
+		return
+	}
+
+	for _, mountpoint := range mountpoints {
+		m.tags["filesystem"] = mountpoint
+
+		// bwwgfs-ctl:
+		// --clientstats: Show client IO statistics.
+		// --nodetype=meta: The node type to query (meta, storage).
+		// --interval:
+		// --mount=/mnt/beeond/: Which mount point
+		//cmd := exec.Command(m.config.Beegfs, "/root/mc/test.txt")
+		mountoption := "--mount=" + mountpoint
+		cmd := exec.Command(m.config.Beegfs, "--clientstats",
+			"--nodetype=meta", mountoption, "--allstats")
+		cmd.Stdin = strings.NewReader("\n")
+		cmdStdout := new(bytes.Buffer)
+		cmdStderr := new(bytes.Buffer)
+		cmd.Stdout = cmdStdout
+		cmd.Stderr = cmdStderr
+		err := cmd.Run()
+		if err != nil {
+			fmt.Fprintf(os.Stderr, "BeegfsMetaCollector.Read(): Failed to execute command \"%s\": %s\n", cmd.String(), err.Error())
+			fmt.Fprintf(os.Stderr, "BeegfsMetaCollector.Read(): command exit code: \"%d\"\n", cmd.ProcessState.ExitCode())
+			data, _ := ioutil.ReadAll(cmdStderr)
+			fmt.Fprintf(os.Stderr, "BeegfsMetaCollector.Read(): command stderr: \"%s\"\n", string(data))
+			data, _ = ioutil.ReadAll(cmdStdout)
+			fmt.Fprintf(os.Stderr, "BeegfsMetaCollector.Read(): command stdout: \"%s\"\n", string(data))
+			return
+		}
+		// Read I/O statistics
+		scanner := bufio.NewScanner(cmdStdout)
+
+		sumLine := regexp.MustCompile(`^Sum:\s+\d+\s+\[[a-zA-Z]+\]+`)
+		//Line := regexp.MustCompile(`^(.*)\s+(\d)+\s+\[([a-zA-Z]+)\]+`)
+		statsLine := regexp.MustCompile(`^(.*?)\s+?(\d.*?)$`)
+		singleSpacePattern := regexp.MustCompile(`\s+`)
+		removePattern := regexp.MustCompile(`[\[|\]]`)
+
+		for scanner.Scan() {
+			readLine := scanner.Text()
+			//fmt.Println(readLine)
+			// Jump few lines, we only want the I/O stats from nodes
+			if !sumLine.MatchString(readLine) {
+				continue
+			}
+
+			match := statsLine.FindStringSubmatch(readLine)
+			// nodeName = "Sum:" or would be nodes
+			// nodeName := match[1]
+			//Remove multiple whitespaces
+			dummy := removePattern.ReplaceAllString(match[2], " ")
+			metaStats := strings.TrimSpace(singleSpacePattern.ReplaceAllString(dummy, " "))
+			split := strings.Split(metaStats, " ")
+
+			// fill map with values
+			// split[i+1] = mdname
+			// split[i] = amount of md operations
+			for i := 0; i <= len(split)-1; i += 2 {
+				if _, ok := m.matches[split[i+1]]; ok {
+					m.matches["beegfs_cmeta_"+split[i+1]] = split[i]
+				} else {
+					f1, err := strconv.ParseFloat(m.matches["other"], 32)
+					if err != nil {
+						cclog.ComponentError(
+							m.name,
+							fmt.Sprintf("Metric (other): Failed to convert str written '%s' to float: %v", m.matches["other"], err))
+						continue
+					}
+					f2, err := strconv.ParseFloat(split[i], 32)
+					if err != nil {
+						cclog.ComponentError(
+							m.name,
+							fmt.Sprintf("Metric (other): Failed to convert str written '%s' to float: %v", m.matches["other"], err))
+						continue
+					}
+					//mdStat["other"] = fmt.Sprintf("%f", f1+f2)
+					m.matches["beegfs_cstorage_other"] = fmt.Sprintf("%f", f1+f2)
+				}
+			}
+
+			for key, data := range m.matches {
+				value, _ := strconv.ParseFloat(data, 32)
+				y, err := lp.New(key, m.tags, m.meta, map[string]interface{}{"value": value}, time.Now())
+				if err == nil {
+					output <- y
+					m.statsProcessedMetrics++
+				}
+			}
+		}
+	}
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
+}
+
+func (m *BeegfsMetaCollector) Close() {
+	m.init = false
+}

collectors/beegfsmetaMetric.md

@@ -0,0 +1,75 @@
+## `BeeGFS on Demand` collector
+This Collector is to collect BeeGFS on Demand (BeeOND) metadata clientstats.
+
+```json
+  "beegfs_meta": {
+	"beegfs_path": "/usr/bin/beegfs-ctl",
+    "exclude_filesystem": [
+      "/mnt/ignore_me"
+    ],
+    "exclude_metrics": [     
+          "ack",
+          "entInf",
+          "fndOwn"
+    ]
+  }
+```
+
+The `BeeGFS On Demand (BeeOND)` collector uses the `beegfs-ctl` command to read performance metrics for
+BeeGFS filesystems.
+
+The reported filesystems can be filtered with the `exclude_filesystem` option
+in the configuration.
+
+The path to the `beegfs-ctl` command can be configured with the `beegfs_path` option
+in the configuration.
+
+When using the `exclude_metrics` option, the excluded metrics are summed as `other`.
+
+Important: The metrics listed below, are similar to the naming of BeeGFS. The Collector prefixes these with `beegfs_cstorage`(beegfs client storage).
+
+For example beegfs metric `open`-> `beegfs_cstorage_open`
+
+Available Metrics:
+
+* sum
+* ack
+* close
+* entInf
+* fndOwn
+* mkdir
+* create
+* rddir
+* refrEnt
+* mdsInf
+* rmdir
+* rmLnk
+* mvDirIns
+* mvFiIns
+* open
+* ren
+* sChDrct
+* sAttr
+* sDirPat
+* stat
+* statfs
+* trunc
+* symlnk
+* unlnk
+* lookLI
+* statLI
+* revalLI
+* openLI
+* createLI
+* hardlnk
+* flckAp
+* flckEn
+* flckRg
+* dirparent
+* listXA
+* getXA
+* rmXA
+* setXA
+* mirror
+
+The collector adds a `filesystem` tag to all metrics

collectors/beegfsstorageMetric.go

@@ -0,0 +1,226 @@
+package collectors
+
+import (
+	"bufio"
+	"bytes"
+	"encoding/json"
+	"fmt"
+	"io/ioutil"
+	"os"
+	"os/exec"
+	"os/user"
+	"regexp"
+	"strconv"
+	"strings"
+	"time"
+
+	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
+	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
+)
+
+// Struct for the collector-specific JSON config
+type BeegfsStorageCollectorConfig struct {
+	Beegfs            string   `json:"beegfs_path"`
+	ExcludeMetrics    []string `json:"exclude_metrics,omitempty"`
+	ExcludeFilesystem []string `json:"exclude_filesystem"`
+}
+
+type BeegfsStorageCollector struct {
+	metricCollector
+	tags                  map[string]string
+	matches               map[string]string
+	config                BeegfsStorageCollectorConfig
+	skipFS                map[string]struct{}
+	statsProcessedMetrics int64
+}
+
+func (m *BeegfsStorageCollector) Init(config json.RawMessage) error {
+	// Check if already initialized
+	if m.init {
+		return nil
+	}
+	// Metrics
+	var storageStat_array = [18]string{
+		"sum", "ack", "sChDrct", "getFSize",
+		"sAttr", "statfs", "trunc", "close",
+		"fsync", "ops-rd", "MiB-rd/s", "ops-wr",
+		"MiB-wr/s", "gendbg", "hrtbeat", "remNode",
+		"storInf", "unlnk"}
+
+	m.name = "BeegfsStorageCollector"
+	m.setup()
+	// Set default beegfs-ctl binary
+
+	m.config.Beegfs = DEFAULT_BEEGFS_CMD
+
+	// Read JSON configuration
+	if len(config) > 0 {
+		err := json.Unmarshal(config, &m.config)
+		if err != nil {
+			return err
+		}
+	}
+	println(m.config.Beegfs)
+	//create map with possible variables
+	m.matches = make(map[string]string)
+	for _, value := range storageStat_array {
+		_, skip := stringArrayContains(m.config.ExcludeMetrics, value)
+		if skip {
+			m.matches["other"] = "0"
+		} else {
+			m.matches["beegfs_cstorage_"+value] = "0"
+		}
+	}
+
+	m.meta = map[string]string{
+		"source": m.name,
+		"group":  "BeegfsStorage",
+	}
+	m.tags = map[string]string{
+		"type":       "node",
+		"filesystem": "",
+	}
+	m.skipFS = make(map[string]struct{})
+	for _, fs := range m.config.ExcludeFilesystem {
+		m.skipFS[fs] = struct{}{}
+	}
+
+	// Beegfs file system statistics can only be queried by user root
+	user, err := user.Current()
+	if err != nil {
+		return fmt.Errorf("BeegfsStorageCollector.Init(): Failed to get current user: %v", err)
+	}
+	if user.Uid != "0" {
+		return fmt.Errorf("BeegfsStorageCollector.Init(): BeeGFS file system statistics can only be queried by user root")
+	}
+
+	// Check if beegfs-ctl is in executable search path
+	_, err = exec.LookPath(m.config.Beegfs)
+	if err != nil {
+		return fmt.Errorf("BeegfsStorageCollector.Init(): Failed to find beegfs-ctl binary '%s': %v", m.config.Beegfs, err)
+	}
+	m.statsProcessedMetrics = 0
+	m.init = true
+	return nil
+}
+
+func (m *BeegfsStorageCollector) Read(interval time.Duration, output chan lp.CCMetric) {
+	if !m.init {
+		return
+	}
+	//get mounpoint
+	buffer, _ := ioutil.ReadFile(string("/proc/mounts"))
+	mounts := strings.Split(string(buffer), "\n")
+	var mountpoints []string
+	for _, line := range mounts {
+		if len(line) == 0 {
+			continue
+		}
+		f := strings.Fields(line)
+		if strings.Contains(f[0], "beegfs_ondemand") {
+			// Skip excluded filesystems
+			if _, skip := m.skipFS[f[1]]; skip {
+				continue
+			}
+			mountpoints = append(mountpoints, f[1])
+		}
+	}
+	if len(mountpoints) == 0 {
+		return
+	}
+	// collects stats for each BeeGFS on Demand FS
+	for _, mountpoint := range mountpoints {
+		m.tags["filesystem"] = mountpoint
+
+		// bwwgfs-ctl:
+		// --clientstats: Show client IO statistics.
+		// --nodetype=meta: The node type to query (meta, storage).
+		// --interval:
+		// --mount=/mnt/beeond/: Which mount point
+		//cmd := exec.Command(m.config.Beegfs, "/root/mc/test.txt")
+		mountoption := "--mount=" + mountpoint
+		cmd := exec.Command(m.config.Beegfs, "--clientstats",
+			"--nodetype=storage", mountoption, "--allstats")
+		cmd.Stdin = strings.NewReader("\n")
+		cmdStdout := new(bytes.Buffer)
+		cmdStderr := new(bytes.Buffer)
+		cmd.Stdout = cmdStdout
+		cmd.Stderr = cmdStderr
+		err := cmd.Run()
+		if err != nil {
+			fmt.Fprintf(os.Stderr, "BeegfsStorageCollector.Read(): Failed to execute command \"%s\": %s\n", cmd.String(), err.Error())
+			fmt.Fprintf(os.Stderr, "BeegfsStorageCollector.Read(): command exit code: \"%d\"\n", cmd.ProcessState.ExitCode())
+			data, _ := ioutil.ReadAll(cmdStderr)
+			fmt.Fprintf(os.Stderr, "BeegfsStorageCollector.Read(): command stderr: \"%s\"\n", string(data))
+			data, _ = ioutil.ReadAll(cmdStdout)
+			fmt.Fprintf(os.Stderr, "BeegfsStorageCollector.Read(): command stdout: \"%s\"\n", string(data))
+			return
+		}
+		// Read I/O statistics
+		scanner := bufio.NewScanner(cmdStdout)
+
+		sumLine := regexp.MustCompile(`^Sum:\s+\d+\s+\[[a-zA-Z]+\]+`)
+		//Line := regexp.MustCompile(`^(.*)\s+(\d)+\s+\[([a-zA-Z]+)\]+`)
+		statsLine := regexp.MustCompile(`^(.*?)\s+?(\d.*?)$`)
+		singleSpacePattern := regexp.MustCompile(`\s+`)
+		removePattern := regexp.MustCompile(`[\[|\]]`)
+
+		for scanner.Scan() {
+			readLine := scanner.Text()
+			//fmt.Println(readLine)
+			// Jump few lines, we only want the I/O stats from nodes
+			if !sumLine.MatchString(readLine) {
+				continue
+			}
+
+			match := statsLine.FindStringSubmatch(readLine)
+			// nodeName = "Sum:" or would be nodes
+			// nodeName := match[1]
+			//Remove multiple whitespaces
+			dummy := removePattern.ReplaceAllString(match[2], " ")
+			metaStats := strings.TrimSpace(singleSpacePattern.ReplaceAllString(dummy, " "))
+			split := strings.Split(metaStats, " ")
+
+			// fill map with values
+			// split[i+1] = mdname
+			// split[i] = amount of operations
+			for i := 0; i <= len(split)-1; i += 2 {
+				if _, ok := m.matches[split[i+1]]; ok {
+					m.matches["beegfs_cstorage_"+split[i+1]] = split[i]
+					//m.matches[split[i+1]] = split[i]
+				} else {
+					f1, err := strconv.ParseFloat(m.matches["other"], 32)
+					if err != nil {
+						cclog.ComponentError(
+							m.name,
+							fmt.Sprintf("Metric (other): Failed to convert str written '%s' to float: %v", m.matches["other"], err))
+						continue
+					}
+					f2, err := strconv.ParseFloat(split[i], 32)
+					if err != nil {
+						cclog.ComponentError(
+							m.name,
+							fmt.Sprintf("Metric (other): Failed to convert str written '%s' to float: %v", m.matches["other"], err))
+						continue
+					}
+					m.matches["beegfs_cstorage_other"] = fmt.Sprintf("%f", f1+f2)
+				}
+			}
+
+			for key, data := range m.matches {
+				value, _ := strconv.ParseFloat(data, 32)
+				y, err := lp.New(key, m.tags, m.meta, map[string]interface{}{"value": value}, time.Now())
+				if err == nil {
+					output <- y
+					m.statsProcessedMetrics++
+				}
+			}
+		}
+	}
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
+}
+
+func (m *BeegfsStorageCollector) Close() {
+	m.init = false
+}

collectors/beegfsstorageMetric.md

@@ -0,0 +1,55 @@
+## `BeeGFS on Demand` collector
+This Collector is to collect BeeGFS on Demand (BeeOND) storage stats.
+
+```json
+  "beegfs_storage": {
+	"beegfs_path": "/usr/bin/beegfs-ctl",
+    "exclude_filesystem": [
+      "/mnt/ignore_me"
+    ],
+    "exclude_metrics": [     
+          "ack",
+		  "storInf",
+		  "unlnk"
+    ]
+  }
+```
+
+The `BeeGFS On Demand (BeeOND)` collector uses the `beegfs-ctl` command to read performance metrics for BeeGFS filesystems.
+
+The reported filesystems can be filtered with the `exclude_filesystem` option
+in the configuration.
+
+The path to the `beegfs-ctl` command can be configured with the `beegfs_path` option
+in the configuration.
+
+When using the `exclude_metrics` option, the excluded metrics are summed as `other`.
+
+Important: The metrics listed below, are similar to the naming of BeeGFS. The Collector prefixes these with `beegfs_cstorage_`(beegfs client meta).
+For example beegfs metric `open`-> `beegfs_cstorage_`
+
+Note: BeeGFS FS offers many Metadata Information. Probably it makes sense to exlcude most of them. Nevertheless, these excluded metrics will be summed as `beegfs_cstorage_other`. 
+
+Available Metrics:
+
+* "sum"
+* "ack"
+* "sChDrct" 
+* "getFSize"
+* "sAttr"
+* "statfs"
+* "trunc"
+* "close"
+* "fsync"
+* "ops-rd"
+* "MiB-rd/s" 
+* "ops-wr"
+* "MiB-wr/s" 
+* "endbg" 
+* "hrtbeat"
+* "remNode"
+* "storInf"
+* "unlnk"
+
+
+The collector adds a `filesystem` tag to all metrics

collectors/collectorManager.go

@@ -19,7 +19,6 @@ var AvailableCollectors = map[string]MetricCollector{
 	"memstat":          new(MemstatCollector),
 	"netstat":          new(NetstatCollector),
 	"ibstat":           new(InfinibandCollector),
-	"ibstat_perfquery": new(InfinibandPerfQueryCollector),
 	"lustrestat":       new(LustreCollector),
 	"cpustat":          new(CpustatCollector),
 	"topprocs":         new(TopProcsCollector),
@@ -35,6 +34,8 @@ var AvailableCollectors = map[string]MetricCollector{
 	"nfs3stat":         new(Nfs3Collector),
 	"nfs4stat":         new(Nfs4Collector),
 	"numastats":        new(NUMAStatsCollector),
+	"beegfs_meta":      new(BeegfsMetaCollector),
+	"beegfs_storage":   new(BeegfsStorageCollector),
 }
 
 // Metric collector manager data structure

collectors/cpufreqCpuinfoMetric.go

@@ -12,6 +12,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 //
@@ -36,7 +37,8 @@ type CPUFreqCpuInfoCollectorTopology struct {
 
 type CPUFreqCpuInfoCollector struct {
 	metricCollector
-	topology []*CPUFreqCpuInfoCollectorTopology
+	topology              []*CPUFreqCpuInfoCollectorTopology
+	statsProcessedMetrics int64
 }
 
 func (m *CPUFreqCpuInfoCollector) Init(config json.RawMessage) error {
@@ -57,7 +59,7 @@ func (m *CPUFreqCpuInfoCollector) Init(config json.RawMessage) error {
 	const cpuInfoFile = "/proc/cpuinfo"
 	file, err := os.Open(cpuInfoFile)
 	if err != nil {
-		return fmt.Errorf("Failed to open file '%s': %v", cpuInfoFile, err)
+		return fmt.Errorf("failed to open file '%s': %v", cpuInfoFile, err)
 	}
 	defer file.Close()
 
@@ -106,14 +108,14 @@ func (m *CPUFreqCpuInfoCollector) Init(config json.RawMessage) error {
 			topology.coreID = coreID
 			topology.coreID_int, err = strconv.ParseInt(coreID, 10, 64)
 			if err != nil {
-				return fmt.Errorf("Unable to convert coreID '%s' to int64: %v", coreID, err)
+				return fmt.Errorf("unable to convert coreID '%s' to int64: %v", coreID, err)
 			}
 
 			// Physical package ID
 			topology.physicalPackageID = physicalPackageID
 			topology.physicalPackageID_int, err = strconv.ParseInt(physicalPackageID, 10, 64)
 			if err != nil {
-				return fmt.Errorf("Unable to convert physicalPackageID '%s' to int64: %v", physicalPackageID, err)
+				return fmt.Errorf("unable to convert physicalPackageID '%s' to int64: %v", physicalPackageID, err)
 			}
 
 			// increase maximun socket / package ID, when required
@@ -155,7 +157,7 @@ func (m *CPUFreqCpuInfoCollector) Init(config json.RawMessage) error {
 			"package_id": t.physicalPackageID,
 		}
 	}
-
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -196,6 +198,7 @@ func (m *CPUFreqCpuInfoCollector) Read(interval time.Duration, output chan lp.CC
 						return
 					}
 					if y, err := lp.New("cpufreq", t.tagSet, m.meta, map[string]interface{}{"value": value}, now); err == nil {
+						m.statsProcessedMetrics++
 						output <- y
 					}
 				}
@@ -203,6 +206,7 @@ func (m *CPUFreqCpuInfoCollector) Read(interval time.Duration, output chan lp.CC
 			}
 		}
 	}
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 }
 
 func (m *CPUFreqCpuInfoCollector) Close() {

collectors/cpufreqMetric.go

@@ -11,6 +11,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 	"golang.org/x/sys/unix"
 )
 
@@ -39,8 +40,9 @@ type CPUFreqCollectorTopology struct {
 //
 type CPUFreqCollector struct {
 	metricCollector
-	topology []CPUFreqCollectorTopology
-	config   struct {
+	topology              []CPUFreqCollectorTopology
+	statsProcessedMetrics int64
+	config                struct {
 		ExcludeMetrics []string `json:"exclude_metrics,omitempty"`
 	}
 }
@@ -70,10 +72,10 @@ func (m *CPUFreqCollector) Init(config json.RawMessage) error {
 	globPattern := filepath.Join(baseDir, "cpu[0-9]*")
 	cpuDirs, err := filepath.Glob(globPattern)
 	if err != nil {
-		return fmt.Errorf("Unable to glob files with pattern '%s': %v", globPattern, err)
+		return fmt.Errorf("unable to glob files with pattern '%s': %v", globPattern, err)
 	}
 	if cpuDirs == nil {
-		return fmt.Errorf("Unable to find any files with pattern '%s'", globPattern)
+		return fmt.Errorf("unable to find any files with pattern '%s'", globPattern)
 	}
 
 	// Initialize CPU topology
@@ -82,38 +84,38 @@ func (m *CPUFreqCollector) Init(config json.RawMessage) error {
 		processor := strings.TrimPrefix(cpuDir, "/sys/devices/system/cpu/cpu")
 		processor_int, err := strconv.ParseInt(processor, 10, 64)
 		if err != nil {
-			return fmt.Errorf("Unable to convert cpuID '%s' to int64: %v", processor, err)
+			return fmt.Errorf("unable to convert cpuID '%s' to int64: %v", processor, err)
 		}
 
 		// Read package ID
 		physicalPackageIDFile := filepath.Join(cpuDir, "topology", "physical_package_id")
 		line, err := ioutil.ReadFile(physicalPackageIDFile)
 		if err != nil {
-			return fmt.Errorf("Unable to read physical package ID from file '%s': %v", physicalPackageIDFile, err)
+			return fmt.Errorf("unable to read physical package ID from file '%s': %v", physicalPackageIDFile, err)
 		}
 		physicalPackageID := strings.TrimSpace(string(line))
 		physicalPackageID_int, err := strconv.ParseInt(physicalPackageID, 10, 64)
 		if err != nil {
-			return fmt.Errorf("Unable to convert packageID '%s' to int64: %v", physicalPackageID, err)
+			return fmt.Errorf("unable to convert packageID '%s' to int64: %v", physicalPackageID, err)
 		}
 
 		// Read core ID
 		coreIDFile := filepath.Join(cpuDir, "topology", "core_id")
 		line, err = ioutil.ReadFile(coreIDFile)
 		if err != nil {
-			return fmt.Errorf("Unable to read core ID from file '%s': %v", coreIDFile, err)
+			return fmt.Errorf("unable to read core ID from file '%s': %v", coreIDFile, err)
 		}
 		coreID := strings.TrimSpace(string(line))
 		coreID_int, err := strconv.ParseInt(coreID, 10, 64)
 		if err != nil {
-			return fmt.Errorf("Unable to convert coreID '%s' to int64: %v", coreID, err)
+			return fmt.Errorf("unable to convert coreID '%s' to int64: %v", coreID, err)
 		}
 
 		// Check access to current frequency file
 		scalingCurFreqFile := filepath.Join(cpuDir, "cpufreq", "scaling_cur_freq")
 		err = unix.Access(scalingCurFreqFile, unix.R_OK)
 		if err != nil {
-			return fmt.Errorf("Unable to access file '%s': %v", scalingCurFreqFile, err)
+			return fmt.Errorf("unable to access file '%s': %v", scalingCurFreqFile, err)
 		}
 
 		t := &m.topology[processor_int]
@@ -166,7 +168,7 @@ func (m *CPUFreqCollector) Init(config json.RawMessage) error {
 			"package_id": t.physicalPackageID,
 		}
 	}
-
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -203,9 +205,11 @@ func (m *CPUFreqCollector) Read(interval time.Duration, output chan lp.CCMetric)
 		}
 
 		if y, err := lp.New("cpufreq", t.tagSet, m.meta, map[string]interface{}{"value": cpuFreq}, now); err == nil {
+			m.statsProcessedMetrics++
 			output <- y
 		}
 	}
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 }
 
 func (m *CPUFreqCollector) Close() {

collectors/cpustatMetric.go

@@ -11,6 +11,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 const CPUSTATFILE = `/proc/stat`
@@ -21,11 +22,11 @@ type CpustatCollectorConfig struct {
 
 type CpustatCollector struct {
 	metricCollector
-	config          CpustatCollectorConfig
-	matches         map[string]int
-	cputags         map[string]map[string]string
-	nodetags        map[string]string
-	num_cpus_metric lp.CCMetric
+	config                CpustatCollectorConfig
+	matches               map[string]int
+	cputags               map[string]map[string]string
+	nodetags              map[string]string
+	statsProcessedMetrics int64
 }
 
 func (m *CpustatCollector) Init(config json.RawMessage) error {
@@ -87,6 +88,7 @@ func (m *CpustatCollector) Init(config json.RawMessage) error {
 			num_cpus++
 		}
 	}
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -107,6 +109,7 @@ func (m *CpustatCollector) parseStatLine(linefields []string, tags map[string]st
 	for name, value := range values {
 		y, err := lp.New(name, tags, m.meta, map[string]interface{}{"value": (value * 100.0) / total}, t)
 		if err == nil {
+			m.statsProcessedMetrics++
 			output <- y
 		}
 	}
@@ -142,8 +145,10 @@ func (m *CpustatCollector) Read(interval time.Duration, output chan lp.CCMetric)
 		time.Now(),
 	)
 	if err == nil {
+		m.statsProcessedMetrics++
 		output <- num_cpus_metric
 	}
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 }
 
 func (m *CpustatCollector) Close() {

collectors/customCmdMetric.go

@@ -10,6 +10,7 @@ import (
 	"time"
 
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 	influx "github.com/influxdata/line-protocol"
 )
 
@@ -23,11 +24,14 @@ type CustomCmdCollectorConfig struct {
 
 type CustomCmdCollector struct {
 	metricCollector
-	handler  *influx.MetricHandler
-	parser   *influx.Parser
-	config   CustomCmdCollectorConfig
-	commands []string
-	files    []string
+	handler                *influx.MetricHandler
+	parser                 *influx.Parser
+	config                 CustomCmdCollectorConfig
+	commands               []string
+	files                  []string
+	statsProcessedMetrics  int64
+	statsProcessedCommands int64
+	statsProcessedFiles    int64
 }
 
 func (m *CustomCmdCollector) Init(config json.RawMessage) error {
@@ -61,11 +65,14 @@ func (m *CustomCmdCollector) Init(config json.RawMessage) error {
 		}
 	}
 	if len(m.files) == 0 && len(m.commands) == 0 {
-		return errors.New("No metrics to collect")
+		return errors.New("no metrics to collect")
 	}
 	m.handler = influx.NewMetricHandler()
 	m.parser = influx.NewParser(m.handler)
 	m.parser.SetTimeFunc(DefaultTime)
+	m.statsProcessedMetrics = 0
+	m.statsProcessedFiles = 0
+	m.statsProcessedCommands = 0
 	m.init = true
 	return nil
 }
@@ -100,9 +107,13 @@ func (m *CustomCmdCollector) Read(interval time.Duration, output chan lp.CCMetri
 
 			y := lp.FromInfluxMetric(c)
 			if err == nil {
+				m.statsProcessedMetrics++
+				stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 				output <- y
 			}
 		}
+		m.statsProcessedCommands++
+		stats.ComponentStatInt(m.name, "processed_commands", m.statsProcessedCommands)
 	}
 	for _, file := range m.files {
 		buffer, err := ioutil.ReadFile(file)
@@ -122,9 +133,13 @@ func (m *CustomCmdCollector) Read(interval time.Duration, output chan lp.CCMetri
 			}
 			y := lp.FromInfluxMetric(f)
 			if err == nil {
+				m.statsProcessedMetrics++
+				stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 				output <- y
 			}
 		}
+		m.statsProcessedFiles++
+		stats.ComponentStatInt(m.name, "processed_files", m.statsProcessedFiles)
 	}
 }
 

collectors/diskstatMetric.go

@@ -11,6 +11,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 //	"log"
@@ -23,9 +24,8 @@ type DiskstatCollectorConfig struct {
 
 type DiskstatCollector struct {
 	metricCollector
-	//matches map[string]int
-	config IOstatCollectorConfig
-	//devices map[string]IOstatCollectorEntry
+	config                DiskstatCollectorConfig
+	statsProcessedMetrics int64
 }
 
 func (m *DiskstatCollector) Init(config json.RawMessage) error {
@@ -44,6 +44,7 @@ func (m *DiskstatCollector) Init(config json.RawMessage) error {
 		return err
 	}
 	defer file.Close()
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -89,12 +90,16 @@ func (m *DiskstatCollector) Read(interval time.Duration, output chan lp.CCMetric
 		y, err := lp.New("disk_total", tags, m.meta, map[string]interface{}{"value": total}, time.Now())
 		if err == nil {
 			y.AddMeta("unit", "GBytes")
+			m.statsProcessedMetrics++
+			stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 			output <- y
 		}
 		free := (stat.Bfree * uint64(stat.Bsize)) / uint64(1000000000)
 		y, err = lp.New("disk_free", tags, m.meta, map[string]interface{}{"value": free}, time.Now())
 		if err == nil {
 			y.AddMeta("unit", "GBytes")
+			m.statsProcessedMetrics++
+			stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 			output <- y
 		}
 		perc := (100 * (total - free)) / total
@@ -105,6 +110,8 @@ func (m *DiskstatCollector) Read(interval time.Duration, output chan lp.CCMetric
 	y, err := lp.New("part_max_used", map[string]string{"type": "node"}, m.meta, map[string]interface{}{"value": int(part_max_used)}, time.Now())
 	if err == nil {
 		y.AddMeta("unit", "percent")
+		m.statsProcessedMetrics++
+		stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 		output <- y
 	}
 }

collectors/gpfsMetric.go

@@ -15,16 +15,28 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
+const DEFAULT_GPFS_CMD = "mmpmon"
+
+type GpfsCollectorLastState struct {
+	bytesRead    int64
+	bytesWritten int64
+}
+
 type GpfsCollector struct {
 	metricCollector
 	tags   map[string]string
 	config struct {
 		Mmpmon            string   `json:"mmpmon_path,omitempty"`
 		ExcludeFilesystem []string `json:"exclude_filesystem,omitempty"`
+		SendBandwidths    bool     `json:"send_bandwidths"`
 	}
-	skipFS map[string]struct{}
+	skipFS                map[string]struct{}
+	lastTimestamp         time.Time // Store time stamp of last tick to derive bandwidths
+	lastState             map[string]GpfsCollectorLastState
+	statsProcessedMetrics int64
 }
 
 func (m *GpfsCollector) Init(config json.RawMessage) error {
@@ -38,7 +50,7 @@ func (m *GpfsCollector) Init(config json.RawMessage) error {
 	m.setup()
 
 	// Set default mmpmon binary
-	m.config.Mmpmon = "/usr/lpp/mmfs/bin/mmpmon"
+	m.config.Mmpmon = DEFAULT_GPFS_CMD
 
 	// Read JSON configuration
 	if len(config) > 0 {
@@ -64,18 +76,19 @@ func (m *GpfsCollector) Init(config json.RawMessage) error {
 	// GPFS / IBM Spectrum Scale file system statistics can only be queried by user root
 	user, err := user.Current()
 	if err != nil {
-		return fmt.Errorf("Failed to get current user: %v", err)
+		return fmt.Errorf("failed to get current user: %v", err)
 	}
 	if user.Uid != "0" {
 		return fmt.Errorf("GPFS file system statistics can only be queried by user root")
 	}
 
 	// Check if mmpmon is in executable search path
-	_, err = exec.LookPath(m.config.Mmpmon)
+	p, err := exec.LookPath(m.config.Mmpmon)
 	if err != nil {
-		return fmt.Errorf("Failed to find mmpmon binary '%s': %v", m.config.Mmpmon, err)
+		return fmt.Errorf("failed to find mmpmon binary '%s': %v", m.config.Mmpmon, err)
 	}
-
+	m.config.Mmpmon = p
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -86,6 +99,13 @@ func (m *GpfsCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 		return
 	}
 
+	// Current time stamp
+	now := time.Now()
+	// time difference to last time stamp
+	timeDiff := now.Sub(m.lastTimestamp).Seconds()
+	// Save current timestamp
+	m.lastTimestamp = now
+
 	// mmpmon:
 	// -p: generate output that can be parsed
 	// -s: suppress the prompt on input
@@ -145,6 +165,12 @@ func (m *GpfsCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 		}
 
 		m.tags["filesystem"] = filesystem
+		if _, ok := m.lastState[filesystem]; !ok {
+			m.lastState[filesystem] = GpfsCollectorLastState{
+				bytesRead:    -1,
+				bytesWritten: -1,
+			}
+		}
 
 		// return code
 		rc, err := strconv.Atoi(key_value["_rc_"])
@@ -187,6 +213,16 @@ func (m *GpfsCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 		}
 		if y, err := lp.New("gpfs_bytes_read", m.tags, m.meta, map[string]interface{}{"value": bytesRead}, timestamp); err == nil {
 			output <- y
+			m.statsProcessedMetrics++
+		}
+		if m.config.SendBandwidths {
+			if lastBytesRead := m.lastState[filesystem].bytesRead; lastBytesRead >= 0 {
+				bwRead := float64(bytesRead-lastBytesRead) / timeDiff
+				if y, err := lp.New("gpfs_bw_read", m.tags, m.meta, map[string]interface{}{"value": bwRead}, timestamp); err == nil {
+					output <- y
+					m.statsProcessedMetrics++
+				}
+			}
 		}
 
 		// bytes written
@@ -199,6 +235,23 @@ func (m *GpfsCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 		}
 		if y, err := lp.New("gpfs_bytes_written", m.tags, m.meta, map[string]interface{}{"value": bytesWritten}, timestamp); err == nil {
 			output <- y
+			m.statsProcessedMetrics++
+		}
+		if m.config.SendBandwidths {
+			if lastBytesWritten := m.lastState[filesystem].bytesRead; lastBytesWritten >= 0 {
+				bwWrite := float64(bytesWritten-lastBytesWritten) / timeDiff
+				if y, err := lp.New("gpfs_bw_write", m.tags, m.meta, map[string]interface{}{"value": bwWrite}, timestamp); err == nil {
+					output <- y
+					m.statsProcessedMetrics++
+				}
+			}
+		}
+
+		if m.config.SendBandwidths {
+			m.lastState[filesystem] = GpfsCollectorLastState{
+				bytesRead:    bytesRead,
+				bytesWritten: bytesWritten,
+			}
 		}
 
 		// number of opens
@@ -211,6 +264,7 @@ func (m *GpfsCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 		}
 		if y, err := lp.New("gpfs_num_opens", m.tags, m.meta, map[string]interface{}{"value": numOpens}, timestamp); err == nil {
 			output <- y
+			m.statsProcessedMetrics++
 		}
 
 		// number of closes
@@ -223,6 +277,7 @@ func (m *GpfsCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 		}
 		if y, err := lp.New("gpfs_num_closes", m.tags, m.meta, map[string]interface{}{"value": numCloses}, timestamp); err == nil {
 			output <- y
+			m.statsProcessedMetrics++
 		}
 
 		// number of reads
@@ -235,6 +290,7 @@ func (m *GpfsCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 		}
 		if y, err := lp.New("gpfs_num_reads", m.tags, m.meta, map[string]interface{}{"value": numReads}, timestamp); err == nil {
 			output <- y
+			m.statsProcessedMetrics++
 		}
 
 		// number of writes
@@ -247,6 +303,7 @@ func (m *GpfsCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 		}
 		if y, err := lp.New("gpfs_num_writes", m.tags, m.meta, map[string]interface{}{"value": numWrites}, timestamp); err == nil {
 			output <- y
+			m.statsProcessedMetrics++
 		}
 
 		// number of read directories
@@ -259,6 +316,7 @@ func (m *GpfsCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 		}
 		if y, err := lp.New("gpfs_num_readdirs", m.tags, m.meta, map[string]interface{}{"value": numReaddirs}, timestamp); err == nil {
 			output <- y
+			m.statsProcessedMetrics++
 		}
 
 		// Number of inode updates
@@ -270,9 +328,11 @@ func (m *GpfsCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 			continue
 		}
 		if y, err := lp.New("gpfs_num_inode_updates", m.tags, m.meta, map[string]interface{}{"value": numInodeUpdates}, timestamp); err == nil {
+			m.statsProcessedMetrics++
 			output <- y
 		}
 	}
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 }
 
 func (m *GpfsCollector) Close() {

collectors/gpfsMetric.md

@@ -5,7 +5,8 @@
     "mmpmon_path": "/path/to/mmpmon",
     "exclude_filesystem": [
       "fs1"
-    ]
+    ],
+    "send_bandwidths" : true
   }
 ```
 
@@ -16,15 +17,18 @@ The reported filesystems can be filtered with the `exclude_filesystem` option
 in the configuration.
 
 The path to the `mmpmon` command can be configured with the `mmpmon_path` option
-in the configuration.
+in the configuration. If nothing is set, the collector searches in `$PATH` for `mmpmon`.
+
 
 Metrics:
-* `bytes_read`
+* `gpfs_bytes_read`
 * `gpfs_bytes_written`
 * `gpfs_num_opens`
 * `gpfs_num_closes`
 * `gpfs_num_reads`
 * `gpfs_num_readdirs`
 * `gpfs_num_inode_updates`
+* `gpfs_bw_read` (if `send_bandwidths == true`)
+* `gpfs_bw_write` (if `send_bandwidths == true`)
 
 The collector adds a `filesystem` tag to all metrics

collectors/infinibandMetric.go

@@ -7,6 +7,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 	"golang.org/x/sys/unix"
 
 	"encoding/json"
@@ -16,22 +17,32 @@ import (
 	"time"
 )
 
-const IB_BASEPATH = `/sys/class/infiniband/`
+const IB_BASEPATH = "/sys/class/infiniband/"
+
+type InfinibandCollectorMetric struct {
+	path string
+	unit string
+}
 
 type InfinibandCollectorInfo struct {
-	LID              string            // IB local Identifier (LID)
-	device           string            // IB device
-	port             string            // IB device port
-	portCounterFiles map[string]string // mapping counter name -> sysfs file
-	tagSet           map[string]string // corresponding tag list
+	LID              string                               // IB local Identifier (LID)
+	device           string                               // IB device
+	port             string                               // IB device port
+	portCounterFiles map[string]InfinibandCollectorMetric // mapping counter name -> InfinibandCollectorMetric
+	tagSet           map[string]string                    // corresponding tag list
+	lastState        map[string]int64                     // State from last measurement
 }
 
 type InfinibandCollector struct {
 	metricCollector
 	config struct {
-		ExcludeDevices []string `json:"exclude_devices,omitempty"` // IB device to exclude e.g. mlx5_0
+		ExcludeDevices     []string `json:"exclude_devices,omitempty"` // IB device to exclude e.g. mlx5_0
+		SendAbsoluteValues bool     `json:"send_abs_values"`           // Send absolut values as read from sys filesystem
+		SendDerivedValues  bool     `json:"send_derived_values"`       // Send derived values e.g. rates
 	}
-	info []*InfinibandCollectorInfo
+	info                  []*InfinibandCollectorInfo
+	lastTimestamp         time.Time // Store time stamp of last tick to derive bandwidths
+	statsProcessedMetrics int64
 }
 
 // Init initializes the Infiniband collector by walking through files below IB_BASEPATH
@@ -49,6 +60,11 @@ func (m *InfinibandCollector) Init(config json.RawMessage) error {
 		"source": m.name,
 		"group":  "Network",
 	}
+
+	// Set default configuration,
+	m.config.SendAbsoluteValues = true
+	m.config.SendDerivedValues = false
+	// Read configuration file, allow overwriting default config
 	if len(config) > 0 {
 		err = json.Unmarshal(config, &m.config)
 		if err != nil {
@@ -60,10 +76,10 @@ func (m *InfinibandCollector) Init(config json.RawMessage) error {
 	globPattern := filepath.Join(IB_BASEPATH, "*", "ports", "*")
 	ibDirs, err := filepath.Glob(globPattern)
 	if err != nil {
-		return fmt.Errorf("Unable to glob files with pattern %s: %v", globPattern, err)
+		return fmt.Errorf("unable to glob files with pattern %s: %v", globPattern, err)
 	}
 	if ibDirs == nil {
-		return fmt.Errorf("Unable to find any directories with pattern %s", globPattern)
+		return fmt.Errorf("unable to find any directories with pattern %s", globPattern)
 	}
 
 	for _, path := range ibDirs {
@@ -97,19 +113,25 @@ func (m *InfinibandCollector) Init(config json.RawMessage) error {
 
 		// Check access to counter files
 		countersDir := filepath.Join(path, "counters")
-		portCounterFiles := map[string]string{
-			"ib_recv":      filepath.Join(countersDir, "port_rcv_data"),
-			"ib_xmit":      filepath.Join(countersDir, "port_xmit_data"),
-			"ib_recv_pkts": filepath.Join(countersDir, "port_rcv_packets"),
-			"ib_xmit_pkts": filepath.Join(countersDir, "port_xmit_packets"),
+		portCounterFiles := map[string]InfinibandCollectorMetric{
+			"ib_recv":      {path: filepath.Join(countersDir, "port_rcv_data"), unit: "bytes"},
+			"ib_xmit":      {path: filepath.Join(countersDir, "port_xmit_data"), unit: "bytes"},
+			"ib_recv_pkts": {path: filepath.Join(countersDir, "port_rcv_packets"), unit: "packets"},
+			"ib_xmit_pkts": {path: filepath.Join(countersDir, "port_xmit_packets"), unit: "packets"},
 		}
-		for _, counterFile := range portCounterFiles {
-			err := unix.Access(counterFile, unix.R_OK)
+		for _, counter := range portCounterFiles {
+			err := unix.Access(counter.path, unix.R_OK)
 			if err != nil {
-				return fmt.Errorf("Unable to access %s: %v", counterFile, err)
+				return fmt.Errorf("unable to access %s: %v", counter.path, err)
 			}
 		}
 
+		// Initialize last state
+		lastState := make(map[string]int64)
+		for counter := range portCounterFiles {
+			lastState[counter] = -1
+		}
+
 		m.info = append(m.info,
 			&InfinibandCollectorInfo{
 				LID:              LID,
@@ -122,13 +144,14 @@ func (m *InfinibandCollector) Init(config json.RawMessage) error {
 					"port":   port,
 					"lid":    LID,
 				},
+				lastState: lastState,
 			})
 	}
 
 	if len(m.info) == 0 {
-		return fmt.Errorf("Found no IB devices")
+		return fmt.Errorf("found no IB devices")
 	}
-
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -141,17 +164,27 @@ func (m *InfinibandCollector) Read(interval time.Duration, output chan lp.CCMetr
 		return
 	}
 
+	// Current time stamp
 	now := time.Now()
+	// time difference to last time stamp
+	timeDiff := now.Sub(m.lastTimestamp).Seconds()
+	// Save current timestamp
+	m.lastTimestamp = now
+
 	for _, info := range m.info {
-		for counterName, counterFile := range info.portCounterFiles {
-			line, err := ioutil.ReadFile(counterFile)
+		for counterName, counterDef := range info.portCounterFiles {
+
+			// Read counter file
+			line, err := ioutil.ReadFile(counterDef.path)
 			if err != nil {
 				cclog.ComponentError(
 					m.name,
-					fmt.Sprintf("Read(): Failed to read from file '%s': %v", counterFile, err))
+					fmt.Sprintf("Read(): Failed to read from file '%s': %v", counterDef.path, err))
 				continue
 			}
 			data := strings.TrimSpace(string(line))
+
+			// convert counter to int64
 			v, err := strconv.ParseInt(data, 10, 64)
 			if err != nil {
 				cclog.ComponentError(
@@ -159,12 +192,33 @@ func (m *InfinibandCollector) Read(interval time.Duration, output chan lp.CCMetr
 					fmt.Sprintf("Read(): Failed to convert Infininiband metrice %s='%s' to int64: %v", counterName, data, err))
 				continue
 			}
-			if y, err := lp.New(counterName, info.tagSet, m.meta, map[string]interface{}{"value": v}, now); err == nil {
-				output <- y
+
+			// Send absolut values
+			if m.config.SendAbsoluteValues {
+				if y, err := lp.New(counterName, info.tagSet, m.meta, map[string]interface{}{"value": v}, now); err == nil {
+					y.AddMeta("unit", counterDef.unit)
+					output <- y
+					m.statsProcessedMetrics++
+				}
+			}
+
+			// Send derived values
+			if m.config.SendDerivedValues {
+				if info.lastState[counterName] >= 0 {
+					rate := float64((v - info.lastState[counterName])) / timeDiff
+					if y, err := lp.New(counterName+"_bw", info.tagSet, m.meta, map[string]interface{}{"value": rate}, now); err == nil {
+						y.AddMeta("unit", counterDef.unit+"/sec")
+						output <- y
+						m.statsProcessedMetrics++
+					}
+				}
+				// Save current state
+				info.lastState[counterName] = v
 			}
 		}
 
 	}
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 }
 
 func (m *InfinibandCollector) Close() {

collectors/infinibandMetric.md

@@ -5,7 +5,9 @@
   "ibstat": {
     "exclude_devices": [
       "mlx4"
-    ]
+    ],
+    "send_abs_values": true,
+    "send_derived_values": true
   }
 ```
 
@@ -22,5 +24,9 @@ Metrics:
 * `ib_xmit`
 * `ib_recv_pkts`
 * `ib_xmit_pkts`
+* `ib_recv_bw` (if `send_derived_values == true`)
+* `ib_xmit_bw` (if `send_derived_values == true`)
+* `ib_recv_pkts_bw` (if `send_derived_values == true`)
+* `ib_xmit_pkts_bw` (if `send_derived_values == true`)
 
 The collector adds a `device` tag to all metrics

collectors/infinibandPerfQueryMetric.go

@@ -1,232 +0,0 @@
-package collectors
-
-import (
-	"fmt"
-	"io/ioutil"
-	"log"
-	"os/exec"
-
-	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
-
-	//	"os"
-	"encoding/json"
-	"errors"
-	"path/filepath"
-	"strconv"
-	"strings"
-	"time"
-)
-
-const PERFQUERY = `/usr/sbin/perfquery`
-
-type InfinibandPerfQueryCollector struct {
-	metricCollector
-	tags   map[string]string
-	lids   map[string]map[string]string
-	config struct {
-		ExcludeDevices []string `json:"exclude_devices,omitempty"`
-		PerfQueryPath  string   `json:"perfquery_path"`
-	}
-}
-
-func (m *InfinibandPerfQueryCollector) Init(config json.RawMessage) error {
-	var err error
-	m.name = "InfinibandCollectorPerfQuery"
-	m.setup()
-	m.meta = map[string]string{"source": m.name, "group": "Network"}
-	m.tags = map[string]string{"type": "node"}
-	if len(config) > 0 {
-		err = json.Unmarshal(config, &m.config)
-		if err != nil {
-			return err
-		}
-	}
-	if len(m.config.PerfQueryPath) == 0 {
-		path, err := exec.LookPath("perfquery")
-		if err == nil {
-			m.config.PerfQueryPath = path
-		}
-	}
-	m.lids = make(map[string]map[string]string)
-	p := fmt.Sprintf("%s/*/ports/*/lid", string(IB_BASEPATH))
-	files, err := filepath.Glob(p)
-	if err != nil {
-		return err
-	}
-	for _, f := range files {
-		lid, err := ioutil.ReadFile(f)
-		if err == nil {
-			plist := strings.Split(strings.Replace(f, string(IB_BASEPATH), "", -1), "/")
-			skip := false
-			for _, d := range m.config.ExcludeDevices {
-				if d == plist[0] {
-					skip = true
-				}
-			}
-			if !skip {
-				m.lids[plist[0]] = make(map[string]string)
-				m.lids[plist[0]][plist[2]] = string(lid)
-			}
-		}
-	}
-
-	for _, ports := range m.lids {
-		for port, lid := range ports {
-			args := fmt.Sprintf("-r %s %s 0xf000", lid, port)
-			command := exec.Command(m.config.PerfQueryPath, args)
-			command.Wait()
-			_, err := command.Output()
-			if err != nil {
-				return fmt.Errorf("Failed to execute %s: %v", m.config.PerfQueryPath, err)
-			}
-		}
-	}
-
-	if len(m.lids) == 0 {
-		return errors.New("No usable IB devices")
-	}
-
-	m.init = true
-	return nil
-}
-
-func (m *InfinibandPerfQueryCollector) doPerfQuery(cmd string, dev string, lid string, port string, tags map[string]string, output chan lp.CCMetric) error {
-
-	args := fmt.Sprintf("-r %s %s 0xf000", lid, port)
-	command := exec.Command(cmd, args)
-	command.Wait()
-	stdout, err := command.Output()
-	if err != nil {
-		log.Print(err)
-		return err
-	}
-	ll := strings.Split(string(stdout), "\n")
-
-	for _, line := range ll {
-		if strings.HasPrefix(line, "PortRcvData") || strings.HasPrefix(line, "RcvData") {
-			lv := strings.Fields(line)
-			v, err := strconv.ParseFloat(lv[1], 64)
-			if err == nil {
-				y, err := lp.New("ib_recv", tags, m.meta, map[string]interface{}{"value": float64(v)}, time.Now())
-				if err == nil {
-					output <- y
-				}
-			}
-		}
-		if strings.HasPrefix(line, "PortXmitData") || strings.HasPrefix(line, "XmtData") {
-			lv := strings.Fields(line)
-			v, err := strconv.ParseFloat(lv[1], 64)
-			if err == nil {
-				y, err := lp.New("ib_xmit", tags, m.meta, map[string]interface{}{"value": float64(v)}, time.Now())
-				if err == nil {
-					output <- y
-				}
-			}
-		}
-		if strings.HasPrefix(line, "PortRcvPkts") || strings.HasPrefix(line, "RcvPkts") {
-			lv := strings.Fields(line)
-			v, err := strconv.ParseFloat(lv[1], 64)
-			if err == nil {
-				y, err := lp.New("ib_recv_pkts", tags, m.meta, map[string]interface{}{"value": float64(v)}, time.Now())
-				if err == nil {
-					output <- y
-				}
-			}
-		}
-		if strings.HasPrefix(line, "PortXmitPkts") || strings.HasPrefix(line, "XmtPkts") {
-			lv := strings.Fields(line)
-			v, err := strconv.ParseFloat(lv[1], 64)
-			if err == nil {
-				y, err := lp.New("ib_xmit_pkts", tags, m.meta, map[string]interface{}{"value": float64(v)}, time.Now())
-				if err == nil {
-					output <- y
-				}
-			}
-		}
-		if strings.HasPrefix(line, "PortRcvPkts") || strings.HasPrefix(line, "RcvPkts") {
-			lv := strings.Fields(line)
-			v, err := strconv.ParseFloat(lv[1], 64)
-			if err == nil {
-				y, err := lp.New("ib_recv_pkts", tags, m.meta, map[string]interface{}{"value": float64(v)}, time.Now())
-				if err == nil {
-					output <- y
-				}
-			}
-		}
-		if strings.HasPrefix(line, "PortXmitPkts") || strings.HasPrefix(line, "XmtPkts") {
-			lv := strings.Fields(line)
-			v, err := strconv.ParseFloat(lv[1], 64)
-			if err == nil {
-				y, err := lp.New("ib_xmit_pkts", tags, m.meta, map[string]interface{}{"value": float64(v)}, time.Now())
-				if err == nil {
-					output <- y
-				}
-			}
-		}
-	}
-	return nil
-}
-
-func (m *InfinibandPerfQueryCollector) Read(interval time.Duration, output chan lp.CCMetric) {
-
-	if m.init {
-		for dev, ports := range m.lids {
-			for port, lid := range ports {
-				tags := map[string]string{
-					"type":   "node",
-					"device": dev,
-					"port":   port,
-					"lid":    lid}
-				path := fmt.Sprintf("%s/%s/ports/%s/counters/", string(IB_BASEPATH), dev, port)
-				buffer, err := ioutil.ReadFile(fmt.Sprintf("%s/port_rcv_data", path))
-				if err == nil {
-					data := strings.Replace(string(buffer), "\n", "", -1)
-					v, err := strconv.ParseFloat(data, 64)
-					if err == nil {
-						y, err := lp.New("ib_recv", tags, m.meta, map[string]interface{}{"value": float64(v)}, time.Now())
-						if err == nil {
-							output <- y
-						}
-					}
-				}
-				buffer, err = ioutil.ReadFile(fmt.Sprintf("%s/port_xmit_data", path))
-				if err == nil {
-					data := strings.Replace(string(buffer), "\n", "", -1)
-					v, err := strconv.ParseFloat(data, 64)
-					if err == nil {
-						y, err := lp.New("ib_xmit", tags, m.meta, map[string]interface{}{"value": float64(v)}, time.Now())
-						if err == nil {
-							output <- y
-						}
-					}
-				}
-				buffer, err = ioutil.ReadFile(fmt.Sprintf("%s/port_rcv_packets", path))
-				if err == nil {
-					data := strings.Replace(string(buffer), "\n", "", -1)
-					v, err := strconv.ParseFloat(data, 64)
-					if err == nil {
-						y, err := lp.New("ib_recv_pkts", tags, m.meta, map[string]interface{}{"value": float64(v)}, time.Now())
-						if err == nil {
-							output <- y
-						}
-					}
-				}
-				buffer, err = ioutil.ReadFile(fmt.Sprintf("%s/port_xmit_packets", path))
-				if err == nil {
-					data := strings.Replace(string(buffer), "\n", "", -1)
-					v, err := strconv.ParseFloat(data, 64)
-					if err == nil {
-						y, err := lp.New("ib_xmit_pkts", tags, m.meta, map[string]interface{}{"value": float64(v)}, time.Now())
-						if err == nil {
-							output <- y
-						}
-					}
-				}
-			}
-		}
-	}
-}
-
-func (m *InfinibandPerfQueryCollector) Close() {
-	m.init = false
-}

collectors/infinibandPerfQueryMetric.md

@@ -1,28 +0,0 @@
-
-## `ibstat_perfquery` collector
-
-```json
-  "ibstat_perfquery": {
-    "perfquery_path": "/path/to/perfquery",
-    "exclude_devices": [
-      "mlx4"
-    ]
-  }
-```
-
-The `ibstat_perfquery` collector includes all Infiniband devices that can be
-found below `/sys/class/infiniband/` and where any of the ports provides a
-LID file (`/sys/class/infiniband/<dev>/ports/<port>/lid`)
-
-The devices can be filtered with the `exclude_devices` option in the configuration.
-
-For each found LID the collector calls the `perfquery` command. The path to the
-`perfquery` command can be configured with the `perfquery_path` option in the configuration
-
-Metrics:
-* `ib_recv`
-* `ib_xmit`
-* `ib_recv_pkts`
-* `ib_xmit_pkts`
-
-The collector adds a `device` tag to all metrics

collectors/iostatMetric.go

@@ -6,6 +6,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 
 	//	"log"
 	"encoding/json"
@@ -29,9 +30,10 @@ type IOstatCollectorEntry struct {
 
 type IOstatCollector struct {
 	metricCollector
-	matches map[string]int
-	config  IOstatCollectorConfig
-	devices map[string]IOstatCollectorEntry
+	matches               map[string]int
+	config                IOstatCollectorConfig
+	devices               map[string]IOstatCollectorEntry
+	statsProcessedMetrics int64
 }
 
 func (m *IOstatCollector) Init(config json.RawMessage) error {
@@ -102,6 +104,7 @@ func (m *IOstatCollector) Init(config json.RawMessage) error {
 			lastValues: values,
 		}
 	}
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return err
 }
@@ -141,6 +144,7 @@ func (m *IOstatCollector) Read(interval time.Duration, output chan lp.CCMetric)
 					y, err := lp.New(name, entry.tags, m.meta, map[string]interface{}{"value": int(diff)}, time.Now())
 					if err == nil {
 						output <- y
+						m.statsProcessedMetrics++
 					}
 				}
 				entry.lastValues[name] = x
@@ -148,6 +152,7 @@ func (m *IOstatCollector) Read(interval time.Duration, output chan lp.CCMetric)
 		}
 		m.devices[device] = entry
 	}
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 }
 
 func (m *IOstatCollector) Close() {

collectors/ipmiMetric.go

@@ -11,6 +11,7 @@ import (
 	"time"
 
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 const IPMITOOL_PATH = `ipmitool`
@@ -26,9 +27,10 @@ type IpmiCollector struct {
 	metricCollector
 	//tags        map[string]string
 	//matches     map[string]string
-	config      IpmiCollectorConfig
-	ipmitool    string
-	ipmisensors string
+	config                IpmiCollectorConfig
+	ipmitool              string
+	ipmisensors           string
+	statsProcessedMetrics int64
 }
 
 func (m *IpmiCollector) Init(config json.RawMessage) error {
@@ -54,8 +56,9 @@ func (m *IpmiCollector) Init(config json.RawMessage) error {
 		m.ipmisensors = p
 	}
 	if len(m.ipmitool) == 0 && len(m.ipmisensors) == 0 {
-		return errors.New("No IPMI reader found")
+		return errors.New("no IPMI reader found")
 	}
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -94,6 +97,7 @@ func (m *IpmiCollector) readIpmiTool(cmd string, output chan lp.CCMetric) {
 			if err == nil {
 				y.AddMeta("unit", unit)
 				output <- y
+				m.statsProcessedMetrics++
 			}
 		}
 	}
@@ -123,6 +127,7 @@ func (m *IpmiCollector) readIpmiSensors(cmd string, output chan lp.CCMetric) {
 						y.AddMeta("unit", lv[4])
 					}
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -141,6 +146,7 @@ func (m *IpmiCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 			m.readIpmiSensors(m.config.IpmisensorsPath, output)
 		}
 	}
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 }
 
 func (m *IpmiCollector) Close() {

collectors/likwidMetric.go

@@ -15,9 +15,12 @@ import (
 	"io/ioutil"
 	"math"
 	"os"
-	"regexp"
+	"os/signal"
+	"sort"
 	"strconv"
 	"strings"
+	"sync"
+	"syscall"
 	"time"
 	"unsafe"
 
@@ -25,51 +28,10 @@ import (
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
 	topo "github.com/ClusterCockpit/cc-metric-collector/internal/ccTopology"
 	agg "github.com/ClusterCockpit/cc-metric-collector/internal/metricAggregator"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 	"github.com/NVIDIA/go-nvml/pkg/dl"
 )
 
-type MetricScope string
-
-const (
-	METRIC_SCOPE_HWTHREAD = iota
-	METRIC_SCOPE_CORE
-	METRIC_SCOPE_LLC
-	METRIC_SCOPE_NUMA
-	METRIC_SCOPE_DIE
-	METRIC_SCOPE_SOCKET
-	METRIC_SCOPE_NODE
-)
-
-func (ms MetricScope) String() string {
-	return string(ms)
-}
-
-func (ms MetricScope) Likwid() string {
-	LikwidDomains := map[string]string{
-		"cpu":        "",
-		"core":       "",
-		"llc":        "C",
-		"numadomain": "M",
-		"die":        "D",
-		"socket":     "S",
-		"node":       "N",
-	}
-	return LikwidDomains[string(ms)]
-}
-
-func (ms MetricScope) Granularity() int {
-	for i, g := range GetAllMetricScopes() {
-		if ms == g {
-			return i
-		}
-	}
-	return -1
-}
-
-func GetAllMetricScopes() []MetricScope {
-	return []MetricScope{"cpu" /*, "core", "llc", "numadomain", "die",*/, "socket", "node"}
-}
-
 const (
 	LIKWID_LIB_NAME       = "liblikwid.so"
 	LIKWID_LIB_DL_FLAGS   = dl.RTLD_LAZY | dl.RTLD_GLOBAL
@@ -77,18 +39,26 @@ const (
 )
 
 type LikwidCollectorMetricConfig struct {
-	Name string `json:"name"` // Name of the metric
-	Calc string `json:"calc"` // Calculation for the metric using
-	//Aggr        string      `json:"aggregation"` // if scope unequal to LIKWID metric scope, the values are combined (sum, min, max, mean or avg, median)
-	Scope       MetricScope `json:"scope"` // scope for calculation. subscopes are aggregated using the 'aggregation' function
-	Publish     bool        `json:"publish"`
-	granulatity MetricScope
+	Name    string `json:"name"` // Name of the metric
+	Calc    string `json:"calc"` // Calculation for the metric using
+	Type    string `json:"type"` // Metric type (aka node, socket, cpu, ...)
+	Publish bool   `json:"publish"`
+	Unit    string `json:"unit"` // Unit of metric if any
 }
 
 type LikwidCollectorEventsetConfig struct {
-	Events      map[string]string `json:"events"`
-	granulatity map[string]MetricScope
-	Metrics     []LikwidCollectorMetricConfig `json:"metrics"`
+	Events  map[string]string             `json:"events"`
+	Metrics []LikwidCollectorMetricConfig `json:"metrics"`
+}
+
+type LikwidEventsetConfig struct {
+	internal int
+	gid      C.int
+	eorder   []*C.char
+	estr     *C.char
+	go_estr  string
+	results  map[int]map[string]interface{}
+	metrics  map[int]map[string]float64
 }
 
 type LikwidCollectorConfig struct {
@@ -98,28 +68,32 @@ type LikwidCollectorConfig struct {
 	InvalidToZero  bool                            `json:"invalid_to_zero,omitempty"`
 	AccessMode     string                          `json:"access_mode,omitempty"`
 	DaemonPath     string                          `json:"accessdaemon_path,omitempty"`
+	LibraryPath    string                          `json:"liblikwid_path,omitempty"`
 }
 
 type LikwidCollector struct {
 	metricCollector
-	cpulist       []C.int
-	cpu2tid       map[int]int
-	sock2tid      map[int]int
-	scopeRespTids map[MetricScope]map[int]int
-	metrics       map[C.int]map[string]int
-	groups        []C.int
-	config        LikwidCollectorConfig
-	results       map[int]map[int]map[string]interface{}
-	mresults      map[int]map[int]map[string]float64
-	gmresults     map[int]map[string]float64
-	basefreq      float64
-	running       bool
+	cpulist               []C.int
+	cpu2tid               map[int]int
+	sock2tid              map[int]int
+	metrics               map[C.int]map[string]int
+	groups                []C.int
+	config                LikwidCollectorConfig
+	gmresults             map[int]map[string]float64
+	basefreq              float64
+	running               bool
+	initialized           bool
+	likwidGroups          map[C.int]LikwidEventsetConfig
+	lock                  sync.Mutex
+	statsMeasurements     int64
+	statsProcessedMetrics int64
+	statsPublishedMetrics int64
 }
 
 type LikwidMetric struct {
 	name      string
 	search    string
-	scope     MetricScope
+	scope     string
 	group_idx int
 }
 
@@ -131,152 +105,99 @@ func eventsToEventStr(events map[string]string) string {
 	return strings.Join(elist, ",")
 }
 
-func getGranularity(counter, event string) MetricScope {
-	if strings.HasPrefix(counter, "PMC") || strings.HasPrefix(counter, "FIXC") {
-		return "cpu"
-	} else if strings.Contains(counter, "BOX") || strings.Contains(counter, "DEV") {
-		return "socket"
-	} else if strings.HasPrefix(counter, "PWR") {
-		if event == "RAPL_CORE_ENERGY" {
-			return "cpu"
-		} else {
-			return "socket"
+func genLikwidEventSet(input LikwidCollectorEventsetConfig) LikwidEventsetConfig {
+	tmplist := make([]string, 0)
+	clist := make([]string, 0)
+	for k := range input.Events {
+		clist = append(clist, k)
+	}
+	sort.Strings(clist)
+	elist := make([]*C.char, 0)
+	for _, k := range clist {
+		v := input.Events[k]
+		tmplist = append(tmplist, fmt.Sprintf("%s:%s", v, k))
+		c_counter := C.CString(k)
+		elist = append(elist, c_counter)
+	}
+	estr := strings.Join(tmplist, ",")
+	res := make(map[int]map[string]interface{})
+	met := make(map[int]map[string]float64)
+	for _, i := range topo.CpuList() {
+		res[i] = make(map[string]interface{})
+		for k := range input.Events {
+			res[i][k] = 0.0
+		}
+		met[i] = make(map[string]float64)
+		for _, v := range input.Metrics {
+			res[i][v.Name] = 0.0
 		}
 	}
-	return "unknown"
+	return LikwidEventsetConfig{
+		gid:     -1,
+		eorder:  elist,
+		estr:    C.CString(estr),
+		go_estr: estr,
+		results: res,
+		metrics: met,
+	}
+}
+
+func testLikwidMetricFormula(formula string, params []string) bool {
+	myparams := make(map[string]interface{})
+	for _, p := range params {
+		myparams[p] = float64(1.0)
+	}
+	_, err := agg.EvalFloat64Condition(formula, myparams)
+	return err == nil
 }
 
 func getBaseFreq() float64 {
+	files := []string{
+		"/sys/devices/system/cpu/cpu0/cpufreq/bios_limit",
+		"/sys/devices/system/cpu/cpu0/cpufreq/base_frequency",
+	}
 	var freq float64 = math.NaN()
-	C.power_init(0)
-	info := C.get_powerInfo()
-	if float64(info.baseFrequency) != 0 {
-		freq = float64(info.baseFrequency) * 1e3
-	} else {
-		buffer, err := ioutil.ReadFile("/sys/devices/system/cpu/cpu0/cpufreq/bios_limit")
+	for _, f := range files {
+		buffer, err := ioutil.ReadFile(f)
 		if err == nil {
 			data := strings.Replace(string(buffer), "\n", "", -1)
 			x, err := strconv.ParseInt(data, 0, 64)
 			if err == nil {
-				freq = float64(x) * 1e3
+				freq = float64(x) * 1e6
 			}
 		}
 	}
+
+	if math.IsNaN(freq) {
+		C.power_init(0)
+		info := C.get_powerInfo()
+		if float64(info.baseFrequency) != 0 {
+			freq = float64(info.baseFrequency) * 1e6
+		}
+		C.power_finalize()
+	}
 	return freq
 }
 
-func (m *LikwidCollector) initGranularity() {
-	splitRegex := regexp.MustCompile("[+-/*()]")
-	for _, evset := range m.config.Eventsets {
-		evset.granulatity = make(map[string]MetricScope)
-		for counter, event := range evset.Events {
-			gran := getGranularity(counter, event)
-			if gran.Granularity() >= 0 {
-				evset.granulatity[counter] = gran
-			}
-		}
-		for i, metric := range evset.Metrics {
-			s := splitRegex.Split(metric.Calc, -1)
-			gran := MetricScope("cpu")
-			evset.Metrics[i].granulatity = gran
-			for _, x := range s {
-				if _, ok := evset.Events[x]; ok {
-					if evset.granulatity[x].Granularity() > gran.Granularity() {
-						gran = evset.granulatity[x]
-					}
-				}
-			}
-			evset.Metrics[i].granulatity = gran
-		}
-	}
-	for i, metric := range m.config.Metrics {
-		s := splitRegex.Split(metric.Calc, -1)
-		gran := MetricScope("cpu")
-		m.config.Metrics[i].granulatity = gran
-		for _, x := range s {
-			for _, evset := range m.config.Eventsets {
-				for _, m := range evset.Metrics {
-					if m.Name == x && m.granulatity.Granularity() > gran.Granularity() {
-						gran = m.granulatity
-					}
-				}
-			}
-		}
-		m.config.Metrics[i].granulatity = gran
-	}
-}
-
-type TopoResolveFunc func(cpuid int) int
-
-func (m *LikwidCollector) getResponsiblities() map[MetricScope]map[int]int {
-	get_cpus := func(scope MetricScope) map[int]int {
-		var slist []int
-		var cpu C.int
-		var input func(index int) string
-		switch scope {
-		case "node":
-			slist = []int{0}
-			input = func(index int) string { return "N:0" }
-		case "socket":
-			input = func(index int) string { return fmt.Sprintf("%s%d:0", scope.Likwid(), index) }
-			slist = topo.SocketList()
-		// case "numadomain":
-		// 	input = func(index int) string { return fmt.Sprintf("%s%d:0", scope.Likwid(), index) }
-		// 	slist = topo.NumaNodeList()
-		// 	cclog.Debug(scope, " ", input(0), " ", slist)
-		// case "die":
-		// 	input = func(index int) string { return fmt.Sprintf("%s%d:0", scope.Likwid(), index) }
-		// 	slist = topo.DieList()
-		// case "llc":
-		// 	input = fmt.Sprintf("%s%d:0", scope.Likwid(), s)
-		// 	slist = topo.LLCacheList()
-		case "cpu":
-			input = func(index int) string { return fmt.Sprintf("%d", index) }
-			slist = topo.CpuList()
-		case "hwthread":
-			input = func(index int) string { return fmt.Sprintf("%d", index) }
-			slist = topo.CpuList()
-		}
-		outmap := make(map[int]int)
-		for _, s := range slist {
-			t := C.CString(input(s))
-			clen := C.cpustr_to_cpulist(t, &cpu, 1)
-			if int(clen) == 1 {
-				outmap[s] = m.cpu2tid[int(cpu)]
-			} else {
-				cclog.Error(fmt.Sprintf("Cannot determine responsible CPU for %s", input(s)))
-				outmap[s] = -1
-			}
-			C.free(unsafe.Pointer(t))
-		}
-		return outmap
-	}
-
-	scopes := GetAllMetricScopes()
-	complete := make(map[MetricScope]map[int]int)
-	for _, s := range scopes {
-		complete[s] = get_cpus(s)
-	}
-	return complete
-}
-
 func (m *LikwidCollector) Init(config json.RawMessage) error {
-	var ret C.int
 	m.name = "LikwidCollector"
+	m.initialized = false
+	m.running = false
 	m.config.AccessMode = LIKWID_DEF_ACCESSMODE
+	m.config.LibraryPath = LIKWID_LIB_NAME
 	if len(config) > 0 {
 		err := json.Unmarshal(config, &m.config)
 		if err != nil {
 			return err
 		}
 	}
-	lib := dl.New(LIKWID_LIB_NAME, LIKWID_LIB_DL_FLAGS)
+	lib := dl.New(m.config.LibraryPath, LIKWID_LIB_DL_FLAGS)
 	if lib == nil {
-		return fmt.Errorf("error instantiating DynamicLibrary for %s", LIKWID_LIB_NAME)
+		return fmt.Errorf("error instantiating DynamicLibrary for %s", m.config.LibraryPath)
 	}
 	err := lib.Open()
 	if err != nil {
-		return fmt.Errorf("error opening %s: %v", LIKWID_LIB_NAME, err)
+		return fmt.Errorf("error opening %s: %v", m.config.LibraryPath, err)
 	}
 
 	if m.config.ForceOverwrite {
@@ -285,7 +206,7 @@ func (m *LikwidCollector) Init(config json.RawMessage) error {
 	}
 	m.setup()
 
-	m.meta = map[string]string{"source": m.name, "group": "PerfCounter"}
+	m.meta = map[string]string{"group": "PerfCounter"}
 	cclog.ComponentDebug(m.name, "Get cpulist and init maps and lists")
 	cpulist := topo.CpuList()
 	m.cpulist = make([]C.int, len(cpulist))
@@ -293,194 +214,175 @@ func (m *LikwidCollector) Init(config json.RawMessage) error {
 	for i, c := range cpulist {
 		m.cpulist[i] = C.int(c)
 		m.cpu2tid[c] = i
-
 	}
-	m.results = make(map[int]map[int]map[string]interface{})
-	m.mresults = make(map[int]map[int]map[string]float64)
+
+	m.likwidGroups = make(map[C.int]LikwidEventsetConfig)
+
+	// m.results = make(map[int]map[int]map[string]interface{})
+	// m.mresults = make(map[int]map[int]map[string]float64)
 	m.gmresults = make(map[int]map[string]float64)
-	cclog.ComponentDebug(m.name, "initialize LIKWID topology")
-	ret = C.topology_init()
-	if ret != 0 {
-		err := errors.New("failed to initialize LIKWID topology")
-		cclog.ComponentError(m.name, err.Error())
-		return err
-	}
-
-	// Determine which counter works at which level. PMC*: cpu, *BOX*: socket, ...
-	m.initGranularity()
-	// Generate map for MetricScope -> scope_id (like socket id) -> responsible id (offset in cpulist)
-	m.scopeRespTids = m.getResponsiblities()
-	switch m.config.AccessMode {
-	case "direct":
-		C.HPMmode(0)
-	case "accessdaemon":
-		if len(m.config.DaemonPath) > 0 {
-			p := os.Getenv("PATH")
-			os.Setenv("PATH", m.config.DaemonPath+":"+p)
-		}
-		C.HPMmode(1)
-	}
-
-	cclog.ComponentDebug(m.name, "initialize LIKWID perfmon module")
-	ret = C.perfmon_init(C.int(len(m.cpulist)), &m.cpulist[0])
-	if ret != 0 {
-		C.topology_finalize()
-		err := errors.New("failed to initialize LIKWID topology")
-		cclog.ComponentError(m.name, err.Error())
-		return err
+	for _, tid := range m.cpu2tid {
+		m.gmresults[tid] = make(map[string]float64)
 	}
 
 	// This is for the global metrics computation test
-	globalParams := make(map[string]interface{})
-	globalParams["time"] = float64(1.0)
-	globalParams["inverseClock"] = float64(1.0)
-	// While adding the events, we test the metrics whether they can be computed at all
-	for i, evset := range m.config.Eventsets {
-		estr := eventsToEventStr(evset.Events)
-		// Generate parameter list for the metric computing test
-		params := make(map[string]interface{})
-		params["time"] = float64(1.0)
-		params["inverseClock"] = float64(1.0)
-		for counter := range evset.Events {
-			params[counter] = float64(1.0)
-		}
-		for _, metric := range evset.Metrics {
-			// Try to evaluate the metric
-			_, err := agg.EvalFloat64Condition(metric.Calc, params)
-			if err != nil {
-				cclog.ComponentError(m.name, "Calculation for metric", metric.Name, "failed:", err.Error())
-				continue
+	totalMetrics := 0
+	// Generate parameter list for the metric computing test
+	params := make([]string, 0)
+	params = append(params, "time", "inverseClock")
+	// Generate parameter list for the global metric computing test
+	globalParams := make([]string, 0)
+	globalParams = append(globalParams, "time", "inverseClock")
+	// We test the eventset metrics whether they can be computed at all
+	for _, evset := range m.config.Eventsets {
+		if len(evset.Events) > 0 {
+			params = params[:2]
+			for counter := range evset.Events {
+				params = append(params, counter)
 			}
-			// If the metric is not in the parameter list for the global metrics, add it
-			if _, ok := globalParams[metric.Name]; !ok {
-				globalParams[metric.Name] = float64(1.0)
-			}
-		}
-		// Now we add the list of events to likwid
-		cstr := C.CString(estr)
-		gid := C.perfmon_addEventSet(cstr)
-		if gid >= 0 {
-			m.groups = append(m.groups, gid)
-		}
-		C.free(unsafe.Pointer(cstr))
-		m.results[i] = make(map[int]map[string]interface{})
-		m.mresults[i] = make(map[int]map[string]float64)
-		for tid := range m.cpulist {
-			m.results[i][tid] = make(map[string]interface{})
-			m.mresults[i][tid] = make(map[string]float64)
-			if i == 0 {
-				m.gmresults[tid] = make(map[string]float64)
+			for _, metric := range evset.Metrics {
+				// Try to evaluate the metric
+				if testLikwidMetricFormula(metric.Calc, params) {
+					// Add the computable metric to the parameter list for the global metrics
+					globalParams = append(globalParams, metric.Name)
+					totalMetrics++
+				} else {
+					metric.Calc = ""
+				}
 			}
+		} else {
+			cclog.ComponentError(m.name, "Invalid Likwid eventset config, no events given")
+			continue
 		}
 	}
 	for _, metric := range m.config.Metrics {
 		// Try to evaluate the global metric
-		_, err := agg.EvalFloat64Condition(metric.Calc, globalParams)
-		if err != nil {
-			cclog.ComponentError(m.name, "Calculation for metric", metric.Name, "failed:", err.Error())
-			continue
+		if !testLikwidMetricFormula(metric.Calc, globalParams) {
+			cclog.ComponentError(m.name, "Calculation for metric", metric.Name, "failed")
+			metric.Calc = ""
+		} else {
+			totalMetrics++
 		}
 	}
 
 	// If no event set could be added, shut down LikwidCollector
-	if len(m.groups) == 0 {
-		C.perfmon_finalize()
-		C.topology_finalize()
-		err := errors.New("no LIKWID performance group initialized")
+	if totalMetrics == 0 {
+		err := errors.New("no LIKWID eventset or metric usable")
 		cclog.ComponentError(m.name, err.Error())
 		return err
 	}
-	m.basefreq = getBaseFreq()
-	cclog.ComponentDebug(m.name, "BaseFreq", m.basefreq)
+	m.statsMeasurements = 0
+	m.statsProcessedMetrics = 0
+	m.statsPublishedMetrics = 0
 	m.init = true
 	return nil
 }
 
 // take a measurement for 'interval' seconds of event set index 'group'
-func (m *LikwidCollector) takeMeasurement(group int, interval time.Duration) error {
+func (m *LikwidCollector) takeMeasurement(evset LikwidEventsetConfig, interval time.Duration) (bool, error) {
 	var ret C.int
-	gid := m.groups[group]
-	ret = C.perfmon_setupCounters(gid)
-	if ret != 0 {
-		gctr := C.GoString(C.perfmon_getGroupName(gid))
-		err := fmt.Errorf("failed to setup performance group %d (%s)", gid, gctr)
-		return err
+
+	m.lock.Lock()
+	if m.initialized {
+		ret = C.perfmon_setupCounters(evset.gid)
+		if ret != 0 {
+			var err error = nil
+			var skip bool = false
+			if ret == -37 {
+				skip = true
+			} else {
+				err = fmt.Errorf("failed to setup performance group %d", evset.gid)
+			}
+			m.lock.Unlock()
+			return skip, err
+		}
+		ret = C.perfmon_startCounters()
+		if ret != 0 {
+			var err error = nil
+			var skip bool = false
+			if ret == -37 {
+				skip = true
+			} else {
+				err = fmt.Errorf("failed to setup performance group %d", evset.gid)
+			}
+			m.lock.Unlock()
+			return skip, err
+		}
+		m.running = true
+		time.Sleep(interval)
+		m.running = false
+		ret = C.perfmon_stopCounters()
+		if ret != 0 {
+			var err error = nil
+			var skip bool = false
+			if ret == -37 {
+				skip = true
+			} else {
+				err = fmt.Errorf("failed to setup performance group %d", evset.gid)
+			}
+			m.lock.Unlock()
+			return skip, err
+		}
 	}
-	ret = C.perfmon_startCounters()
-	if ret != 0 {
-		gctr := C.GoString(C.perfmon_getGroupName(gid))
-		err := fmt.Errorf("failed to start performance group %d (%s)", gid, gctr)
-		return err
-	}
-	m.running = true
-	time.Sleep(interval)
-	m.running = false
-	ret = C.perfmon_stopCounters()
-	if ret != 0 {
-		gctr := C.GoString(C.perfmon_getGroupName(gid))
-		err := fmt.Errorf("failed to stop performance group %d (%s)", gid, gctr)
-		return err
-	}
-	return nil
+	m.lock.Unlock()
+	m.statsMeasurements++
+	stats.ComponentStatInt(m.name, "measurements", m.statsMeasurements)
+	return false, nil
 }
 
 // Get all measurement results for an event set, derive the metric values out of the measurement results and send it
-func (m *LikwidCollector) calcEventsetMetrics(group int, interval time.Duration, output chan lp.CCMetric) error {
-	var eidx C.int
-	evset := m.config.Eventsets[group]
-	gid := m.groups[group]
+func (m *LikwidCollector) calcEventsetMetrics(evset LikwidEventsetConfig, interval time.Duration, output chan lp.CCMetric) error {
 	invClock := float64(1.0 / m.basefreq)
 
 	// Go over events and get the results
-	for eidx = 0; int(eidx) < len(evset.Events); eidx++ {
-		ctr := C.perfmon_getCounterName(gid, eidx)
-		ev := C.perfmon_getEventName(gid, eidx)
-		gctr := C.GoString(ctr)
-		gev := C.GoString(ev)
-		// MetricScope for the counter (and if needed the event)
-		scope := getGranularity(gctr, gev)
-		// Get the map scope-id -> tids
-		// This way we read less counters like only the responsible hardware thread for a socket
-		scopemap := m.scopeRespTids[scope]
-		for _, tid := range scopemap {
-			if tid >= 0 {
-				m.results[group][tid]["time"] = interval.Seconds()
-				m.results[group][tid]["inverseClock"] = invClock
-				res := C.perfmon_getLastResult(gid, eidx, C.int(tid))
-				m.results[group][tid][gctr] = float64(res)
-			}
+	for eidx, counter := range evset.eorder {
+		gctr := C.GoString(counter)
+		for _, tid := range m.cpu2tid {
+			res := C.perfmon_getLastResult(evset.gid, C.int(eidx), C.int(tid))
+			evset.results[tid][gctr] = float64(res)
+			evset.results[tid]["time"] = interval.Seconds()
+			evset.results[tid]["inverseClock"] = invClock
 		}
 	}
 
 	// Go over the event set metrics, derive the value out of the event:counter values and send it
-	for _, metric := range evset.Metrics {
+	for _, metric := range m.config.Eventsets[evset.internal].Metrics {
 		// The metric scope is determined in the Init() function
 		// Get the map scope-id -> tids
-		scopemap := m.scopeRespTids[metric.Scope]
+		scopemap := m.cpu2tid
+		if metric.Type == "socket" {
+			scopemap = m.sock2tid
+		}
 		for domain, tid := range scopemap {
-			if tid >= 0 {
-				value, err := agg.EvalFloat64Condition(metric.Calc, m.results[group][tid])
+			if tid >= 0 && len(metric.Calc) > 0 {
+				value, err := agg.EvalFloat64Condition(metric.Calc, evset.results[tid])
 				if err != nil {
 					cclog.ComponentError(m.name, "Calculation for metric", metric.Name, "failed:", err.Error())
 					continue
 				}
-				m.mresults[group][tid][metric.Name] = value
+				evset.metrics[tid][metric.Name] = value
 				if m.config.InvalidToZero && math.IsNaN(value) {
 					value = 0.0
 				}
 				if m.config.InvalidToZero && math.IsInf(value, 0) {
 					value = 0.0
 				}
+				m.statsProcessedMetrics++
+				stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 				// Now we have the result, send it with the proper tags
 				if !math.IsNaN(value) {
 					if metric.Publish {
-						tags := map[string]string{"type": metric.Scope.String()}
-						if metric.Scope != "node" {
-							tags["type-id"] = fmt.Sprintf("%d", domain)
-						}
 						fields := map[string]interface{}{"value": value}
-						y, err := lp.New(metric.Name, tags, m.meta, fields, time.Now())
+						y, err := lp.New(metric.Name, map[string]string{"type": metric.Type}, m.meta, fields, time.Now())
 						if err == nil {
+							if metric.Type != "node" {
+								y.AddTag("type-id", fmt.Sprintf("%d", domain))
+							}
+							if len(metric.Unit) > 0 {
+								y.AddMeta("unit", metric.Unit)
+							}
+							m.statsPublishedMetrics++
+							stats.ComponentStatInt(m.name, "published_metrics", m.statsPublishedMetrics)
 							output <- y
 						}
 					}
@@ -495,13 +397,16 @@ func (m *LikwidCollector) calcEventsetMetrics(group int, interval time.Duration,
 // Go over the global metrics, derive the value out of the event sets' metric values and send it
 func (m *LikwidCollector) calcGlobalMetrics(interval time.Duration, output chan lp.CCMetric) error {
 	for _, metric := range m.config.Metrics {
-		scopemap := m.scopeRespTids[metric.Scope]
+		scopemap := m.cpu2tid
+		if metric.Type == "socket" {
+			scopemap = m.sock2tid
+		}
 		for domain, tid := range scopemap {
 			if tid >= 0 {
 				// Here we generate parameter list
 				params := make(map[string]interface{})
-				for j := range m.groups {
-					for mname, mres := range m.mresults[j][tid] {
+				for _, evset := range m.likwidGroups {
+					for mname, mres := range evset.metrics[tid] {
 						params[mname] = mres
 					}
 				}
@@ -518,16 +423,23 @@ func (m *LikwidCollector) calcGlobalMetrics(interval time.Duration, output chan
 				if m.config.InvalidToZero && math.IsInf(value, 0) {
 					value = 0.0
 				}
+				m.statsProcessedMetrics++
+				stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 				// Now we have the result, send it with the proper tags
 				if !math.IsNaN(value) {
 					if metric.Publish {
-						tags := map[string]string{"type": metric.Scope.String()}
-						if metric.Scope != "node" {
-							tags["type-id"] = fmt.Sprintf("%d", domain)
-						}
+						tags := map[string]string{"type": metric.Type}
 						fields := map[string]interface{}{"value": value}
 						y, err := lp.New(metric.Name, tags, m.meta, fields, time.Now())
 						if err == nil {
+							if metric.Type != "node" {
+								y.AddTag("type-id", fmt.Sprintf("%d", domain))
+							}
+							if len(metric.Unit) > 0 {
+								y.AddMeta("unit", metric.Unit)
+							}
+							m.statsPublishedMetrics++
+							stats.ComponentStatInt(m.name, "published_metrics", m.statsPublishedMetrics)
 							output <- y
 						}
 					}
@@ -538,38 +450,163 @@ func (m *LikwidCollector) calcGlobalMetrics(interval time.Duration, output chan
 	return nil
 }
 
+func (m *LikwidCollector) LateInit() error {
+	var ret C.int
+	if m.initialized {
+		return nil
+	}
+	switch m.config.AccessMode {
+	case "direct":
+		C.HPMmode(0)
+	case "accessdaemon":
+		if len(m.config.DaemonPath) > 0 {
+			p := os.Getenv("PATH")
+			os.Setenv("PATH", m.config.DaemonPath+":"+p)
+		}
+		C.HPMmode(1)
+	}
+	cclog.ComponentDebug(m.name, "initialize LIKWID topology")
+	ret = C.topology_init()
+	if ret != 0 {
+		err := errors.New("failed to initialize LIKWID topology")
+		cclog.ComponentError(m.name, err.Error())
+		return err
+	}
+
+	m.sock2tid = make(map[int]int)
+	tmp := make([]C.int, 1)
+	for _, sid := range topo.SocketList() {
+		cstr := C.CString(fmt.Sprintf("S%d:0", sid))
+		ret = C.cpustr_to_cpulist(cstr, &tmp[0], 1)
+		if ret > 0 {
+			m.sock2tid[sid] = m.cpu2tid[int(tmp[0])]
+		}
+		C.free(unsafe.Pointer(cstr))
+	}
+
+	m.basefreq = getBaseFreq()
+	cclog.ComponentDebug(m.name, "BaseFreq", m.basefreq)
+
+	cclog.ComponentDebug(m.name, "initialize LIKWID perfmon module")
+	ret = C.perfmon_init(C.int(len(m.cpulist)), &m.cpulist[0])
+	if ret != 0 {
+		var err error = nil
+		C.topology_finalize()
+		if ret != -22 {
+			err = errors.New("failed to initialize LIKWID perfmon")
+			cclog.ComponentError(m.name, err.Error())
+		} else {
+			err = errors.New("access to LIKWID perfmon locked")
+		}
+		return err
+	}
+
+	// While adding the events, we test the metrics whether they can be computed at all
+	for i, evset := range m.config.Eventsets {
+		var gid C.int
+		if len(evset.Events) > 0 {
+			skip := false
+			likwidGroup := genLikwidEventSet(evset)
+			for _, g := range m.likwidGroups {
+				if likwidGroup.go_estr == g.go_estr {
+					skip = true
+					break
+				}
+			}
+			if skip {
+				continue
+			}
+			// Now we add the list of events to likwid
+			gid = C.perfmon_addEventSet(likwidGroup.estr)
+			if gid >= 0 {
+				likwidGroup.gid = gid
+				likwidGroup.internal = i
+				m.likwidGroups[gid] = likwidGroup
+			}
+		} else {
+			cclog.ComponentError(m.name, "Invalid Likwid eventset config, no events given")
+			continue
+		}
+
+	}
+
+	// If no event set could be added, shut down LikwidCollector
+	if len(m.likwidGroups) == 0 {
+		C.perfmon_finalize()
+		C.topology_finalize()
+		err := errors.New("no LIKWID performance group initialized")
+		cclog.ComponentError(m.name, err.Error())
+		return err
+	}
+	sigchan := make(chan os.Signal, 1)
+	signal.Notify(sigchan, syscall.SIGCHLD)
+	signal.Notify(sigchan, os.Interrupt)
+	go func() {
+		<-sigchan
+
+		signal.Stop(sigchan)
+		m.initialized = false
+	}()
+	m.initialized = true
+	return nil
+}
+
 // main read function taking multiple measurement rounds, each 'interval' seconds long
 func (m *LikwidCollector) Read(interval time.Duration, output chan lp.CCMetric) {
+	var skip bool = false
+	var err error
 	if !m.init {
 		return
 	}
 
-	for i := range m.groups {
-		// measure event set 'i' for 'interval' seconds
-		err := m.takeMeasurement(i, interval)
+	if !m.initialized {
+		m.lock.Lock()
+		err = m.LateInit()
 		if err != nil {
-			cclog.ComponentError(m.name, err.Error())
+			m.lock.Unlock()
 			return
 		}
-		// read measurements and derive event set metrics
-		m.calcEventsetMetrics(i, interval, output)
+		m.initialized = true
+		m.lock.Unlock()
+	}
+
+	if m.initialized && !skip {
+		for _, evset := range m.likwidGroups {
+			if !skip {
+				// measure event set 'i' for 'interval' seconds
+				skip, err = m.takeMeasurement(evset, interval)
+				if err != nil {
+					cclog.ComponentError(m.name, err.Error())
+					return
+				}
+			}
+
+			if !skip {
+				// read measurements and derive event set metrics
+				m.calcEventsetMetrics(evset, interval, output)
+			}
+		}
+		if !skip {
+			// use the event set metrics to derive the global metrics
+			m.calcGlobalMetrics(interval, output)
+		}
 	}
-	// use the event set metrics to derive the global metrics
-	m.calcGlobalMetrics(interval, output)
 }
 
 func (m *LikwidCollector) Close() {
 	if m.init {
-		cclog.ComponentDebug(m.name, "Closing ...")
 		m.init = false
-		if m.running {
-			cclog.ComponentDebug(m.name, "Stopping counters")
-			C.perfmon_stopCounters()
+		cclog.ComponentDebug(m.name, "Closing ...")
+		m.lock.Lock()
+		if m.initialized {
+			cclog.ComponentDebug(m.name, "Finalize LIKWID perfmon module")
+			C.perfmon_finalize()
+			m.initialized = false
 		}
-		cclog.ComponentDebug(m.name, "Finalize LIKWID perfmon module")
-		C.perfmon_finalize()
+		m.lock.Unlock()
 		cclog.ComponentDebug(m.name, "Finalize LIKWID topology module")
 		C.topology_finalize()
+
 		cclog.ComponentDebug(m.name, "Closing done")
 	}
 }

collectors/likwidMetric.md

@@ -4,14 +4,17 @@
 The `likwid` collector is probably the most complicated collector. The LIKWID library is included as static library with *direct* access mode. The *direct* access mode is suitable if the daemon is executed by a root user. The static library does not contain the performance groups, so all information needs to be provided in the configuration.
 
 The `likwid` configuration consists of two parts, the "eventsets" and "globalmetrics":
-- An event set list itself has two parts, the "events" and a set of derivable "metrics". Each of the "events" is a counter:event pair in LIKWID's syntax. The "metrics" are a list of formulas to derive the metric value from the measurements of the "events". Each metric has a name, the formula, a scope and a publish flag. Counter names can be used like variables in the formulas, so `PMC0+PMC1` sums the measurements for the both events configured in the counters `PMC0` and `PMC1`. The scope tells the Collector whether it is a metric for each hardware thread (`cpu`) or each CPU socket (`socket`). The last one is the publishing flag. It tells the collector whether a metric should be sent to the router.
-- The global metrics are metrics which require data from all event set measurements to be derived. The inputs are the metrics in the event sets. Similar to the metrics in the event sets, the global metrics are defined by a name, a formula, a scope and a publish flag. See event set metrics for details. The only difference is that there is no access to the raw event measurements anymore but only to the metrics. So, the idea is to derive a metric in the "eventsets" section and reuse it in the "globalmetrics" part. If you need a metric only for deriving the global metrics, disable forwarding of the event set metrics. **Be aware** that the combination might be misleading because the "behavior" of a metric changes over time and the multiple measurements might count different computing phases.
+- An event set list itself has two parts, the "events" and a set of derivable "metrics". Each of the "events" is a counter:event pair in LIKWID's syntax. The "metrics" are a list of formulas to derive the metric value from the measurements of the "events". Each metric has a name, the formula, a scope and a publish flag. Counter names can be used like variables in the formulas, so `PMC0+PMC1` sums the measurements for the both events configured in the counters `PMC0` and `PMC1`. The scope tells the Collector whether it is a metric for each hardware thread (`cpu`) or each CPU socket (`socket`). You may specify a unit for the metric with `unit`. The last one is the publishing flag. It tells the collector whether a metric should be sent to the router.
+- The global metrics are metrics which require data from all event set measurements to be derived. The inputs are the metrics in the event sets. Similar to the metrics in the event sets, the global metrics are defined by a name, a formula, a scope and a publish flag. See event set metrics for details. The only difference is that there is no access to the raw event measurements anymore but only to the metrics. So, the idea is to derive a metric in the "eventsets" section and reuse it in the "globalmetrics" part. If you need a metric only for deriving the global metrics, disable forwarding of the event set metrics (`publish=false`). **Be aware** that the combination might be misleading because the "behavior" of a metric changes over time and the multiple measurements might count different computing phases. Similar to the metrics in the eventset, you can specify a metric unit with the `unit` field.
 
 Additional options:
 - `access_mode` : Method to use for hardware performance monitoring (`direct` access as root user, `accessdaemon` for the daemon mode)
 - `accessdaemon_path`: Folder with the access daemon `likwid-accessD`, commonly `$LIKWID_INSTALL_LOC/sbin`
 - `force_overwrite`: Same as setting `LIKWID_FORCE=1`. In case counters are already in-use, LIKWID overwrites their configuration to do its measurements
 - `invalid_to_zero`: In some cases, the calculations result in `NaN` or `Inf`. With this option, all `NaN` and `Inf` values are replaces with `0.0`.
+- `access_mode`: Specify LIKWID access mode: `direct` for direct register access as root user or `accessdaemon`
+- `accessdaemon_path`: Folder of the accessDaemon `likwid-accessD`
+- `liblikwid_path`: Location of `liblikwid.so`
 
 ### Available metric scopes
 
@@ -54,7 +57,8 @@ $ scripts/likwid_perfgroup_to_cc_config.py ICX MEM_DP
       "calc": "time",
       "name": "Runtime (RDTSC) [s]",
       "publish": true,
-      "scope": "hwthread"
+      "unit": "seconds"
+      "scope": "cpu"
     },
     {
       "..." : "..."
@@ -104,25 +108,28 @@ $ chwon $CCUSER /var/run/likwid.lock
           {
             "name": "ipc",
             "calc": "PMC0/PMC1",
-            "scope": "cpu",
+            "type": "cpu",
             "publish": true
           },
           {
             "name": "flops_any",
             "calc": "0.000001*PMC2/time",
-            "scope": "cpu",
+            "unit": "MFlops/s",
+            "type": "cpu",
             "publish": true
           },
           {
-            "name": "clock_mhz",
+            "name": "clock",
             "calc": "0.000001*(FIXC1/FIXC2)/inverseClock",
-            "scope": "cpu",
+            "type": "cpu",
+            "unit": "MHz",
             "publish": true
           },
           {
             "name": "mem1",
             "calc": "0.000001*(DFC0+DFC1+DFC2+DFC3)*64.0/time",
-            "scope": "socket",
+            "unit": "Mbyte/s",
+            "type": "socket",
             "publish": false
           }
         ]
@@ -140,19 +147,22 @@ $ chwon $CCUSER /var/run/likwid.lock
           {
             "name": "pwr_core",
             "calc": "PWR0/time",
-            "scope": "socket",
+            "unit": "Watt"
+            "type": "socket",
             "publish": true
           },
           {
             "name": "pwr_pkg",
             "calc": "PWR1/time",
-            "scope": "socket",
+            "type": "socket",
+            "unit": "Watt"
             "publish": true
           },
           {
             "name": "mem2",
             "calc": "0.000001*(DFC0+DFC1+DFC2+DFC3)*64.0/time",
-            "scope": "socket",
+            "unit": "Mbyte/s",
+            "type": "socket",
             "publish": false
           }
         ]
@@ -162,7 +172,8 @@ $ chwon $CCUSER /var/run/likwid.lock
       {
         "name": "mem_bw",
         "calc": "mem1+mem2",
-        "scope": "socket",
+        "type": "socket",
+        "unit": "Mbyte/s",
         "publish": true
       }
     ]
@@ -198,3 +209,4 @@ IPC   PMC0/PMC1                  ->     {
                                  ->   ]
 ```
 
+The script `scripts/likwid_perfgroup_to_cc_config.py` might help you.

collectors/loadavgMetric.go

@@ -10,6 +10,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 //
@@ -32,6 +33,7 @@ type LoadavgCollector struct {
 	config       struct {
 		ExcludeMetrics []string `json:"exclude_metrics,omitempty"`
 	}
+	statsProcessedMetrics int64
 }
 
 func (m *LoadavgCollector) Init(config json.RawMessage) error {
@@ -63,6 +65,7 @@ func (m *LoadavgCollector) Init(config json.RawMessage) error {
 	for i, name := range m.proc_matches {
 		_, m.proc_skips[i] = stringArrayContains(m.config.ExcludeMetrics, name)
 	}
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -98,6 +101,7 @@ func (m *LoadavgCollector) Read(interval time.Duration, output chan lp.CCMetric)
 		y, err := lp.New(name, m.tags, m.meta, map[string]interface{}{"value": x}, now)
 		if err == nil {
 			output <- y
+			m.statsProcessedMetrics++
 		}
 	}
 
@@ -117,9 +121,10 @@ func (m *LoadavgCollector) Read(interval time.Duration, output chan lp.CCMetric)
 		y, err := lp.New(name, m.tags, m.meta, map[string]interface{}{"value": x}, now)
 		if err == nil {
 			output <- y
+			m.statsProcessedMetrics++
 		}
-
 	}
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 }
 
 func (m *LoadavgCollector) Close() {

collectors/lustreMetric.go

@@ -12,6 +12,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 const LUSTRE_SYSFS = `/sys/fs/lustre`
@@ -19,20 +20,32 @@ const LCTL_CMD = `lctl`
 const LCTL_OPTION = `get_param`
 
 type LustreCollectorConfig struct {
-	LCtlCommand    string   `json:"lctl_command"`
-	ExcludeMetrics []string `json:"exclude_metrics"`
-	SendAllMetrics bool     `json:"send_all_metrics"`
-	Sudo           bool     `json:"use_sudo"`
+	LCtlCommand        string   `json:"lctl_command,omitempty"`
+	ExcludeMetrics     []string `json:"exclude_metrics,omitempty"`
+	Sudo               bool     `json:"use_sudo,omitempty"`
+	SendAbsoluteValues bool     `json:"send_abs_values,omitempty"`
+	SendDerivedValues  bool     `json:"send_derived_values,omitempty"`
+	SendDiffValues     bool     `json:"send_diff_values,omitempty"`
+}
+
+type LustreMetricDefinition struct {
+	name       string
+	lineprefix string
+	lineoffset int
+	unit       string
+	calc       string
 }
 
 type LustreCollector struct {
 	metricCollector
-	tags    map[string]string
-	matches map[string]map[string]int
-	stats   map[string]map[string]int64
-	config  LustreCollectorConfig
-	lctl    string
-	sudoCmd string
+	tags                  map[string]string
+	config                LustreCollectorConfig
+	lctl                  string
+	sudoCmd               string
+	lastTimestamp         time.Time                   // Store time stamp of last tick to derive bandwidths
+	definitions           []LustreMetricDefinition    // Combined list without excluded metrics
+	stats                 map[string]map[string]int64 // Data for last value per device and metric
+	statsProcessedMetrics int64
 }
 
 func (m *LustreCollector) getDeviceDataCommand(device string) []string {
@@ -75,6 +88,16 @@ func (m *LustreCollector) getDevices() []string {
 	return devices
 }
 
+func getMetricData(lines []string, prefix string, offset int) (int64, error) {
+	for _, line := range lines {
+		if strings.HasPrefix(line, prefix) {
+			lf := strings.Fields(line)
+			return strconv.ParseInt(lf[offset], 0, 64)
+		}
+	}
+	return 0, errors.New("no such line in data")
+}
+
 // //Version reading the stats data of a device from sysfs
 // func (m *LustreCollector) getDeviceDataSysfs(device string) []string {
 // 	llitedir := filepath.Join(LUSTRE_SYSFS, "llite")
@@ -87,6 +110,183 @@ func (m *LustreCollector) getDevices() []string {
 // 	return strings.Split(string(buffer), "\n")
 // }
 
+var LustreAbsMetrics = []LustreMetricDefinition{
+	{
+		name:       "lustre_read_requests",
+		lineprefix: "read_bytes",
+		lineoffset: 1,
+		unit:       "requests",
+		calc:       "none",
+	},
+	{
+		name:       "lustre_write_requests",
+		lineprefix: "write_bytes",
+		lineoffset: 1,
+		unit:       "requests",
+		calc:       "none",
+	},
+	{
+		name:       "lustre_read_bytes",
+		lineprefix: "read_bytes",
+		lineoffset: 6,
+		unit:       "bytes",
+		calc:       "none",
+	},
+	{
+		name:       "lustre_write_bytes",
+		lineprefix: "write_bytes",
+		lineoffset: 6,
+		unit:       "bytes",
+		calc:       "none",
+	},
+	{
+		name:       "lustre_open",
+		lineprefix: "open",
+		lineoffset: 1,
+		unit:       "",
+		calc:       "none",
+	},
+	{
+		name:       "lustre_close",
+		lineprefix: "close",
+		lineoffset: 1,
+		unit:       "",
+		calc:       "none",
+	},
+	{
+		name:       "lustre_setattr",
+		lineprefix: "setattr",
+		lineoffset: 1,
+		unit:       "",
+		calc:       "none",
+	},
+	{
+		name:       "lustre_getattr",
+		lineprefix: "getattr",
+		lineoffset: 1,
+		unit:       "",
+		calc:       "none",
+	},
+	{
+		name:       "lustre_statfs",
+		lineprefix: "statfs",
+		lineoffset: 1,
+		unit:       "",
+		calc:       "none",
+	},
+	{
+		name:       "lustre_inode_permission",
+		lineprefix: "inode_permission",
+		lineoffset: 1,
+		unit:       "",
+		calc:       "none",
+	},
+}
+
+var LustreDiffMetrics = []LustreMetricDefinition{
+	{
+		name:       "lustre_read_requests_diff",
+		lineprefix: "read_bytes",
+		lineoffset: 1,
+		unit:       "requests",
+		calc:       "difference",
+	},
+	{
+		name:       "lustre_write_requests_diff",
+		lineprefix: "write_bytes",
+		lineoffset: 1,
+		unit:       "requests",
+		calc:       "difference",
+	},
+	{
+		name:       "lustre_read_bytes_diff",
+		lineprefix: "read_bytes",
+		lineoffset: 6,
+		unit:       "bytes",
+		calc:       "difference",
+	},
+	{
+		name:       "lustre_write_bytes_diff",
+		lineprefix: "write_bytes",
+		lineoffset: 6,
+		unit:       "bytes",
+		calc:       "difference",
+	},
+	{
+		name:       "lustre_open_diff",
+		lineprefix: "open",
+		lineoffset: 1,
+		unit:       "",
+		calc:       "difference",
+	},
+	{
+		name:       "lustre_close_diff",
+		lineprefix: "close",
+		lineoffset: 1,
+		unit:       "",
+		calc:       "difference",
+	},
+	{
+		name:       "lustre_setattr_diff",
+		lineprefix: "setattr",
+		lineoffset: 1,
+		unit:       "",
+		calc:       "difference",
+	},
+	{
+		name:       "lustre_getattr_diff",
+		lineprefix: "getattr",
+		lineoffset: 1,
+		unit:       "",
+		calc:       "difference",
+	},
+	{
+		name:       "lustre_statfs_diff",
+		lineprefix: "statfs",
+		lineoffset: 1,
+		unit:       "",
+		calc:       "difference",
+	},
+	{
+		name:       "lustre_inode_permission_diff",
+		lineprefix: "inode_permission",
+		lineoffset: 1,
+		unit:       "",
+		calc:       "difference",
+	},
+}
+
+var LustreDeriveMetrics = []LustreMetricDefinition{
+	{
+		name:       "lustre_read_requests_rate",
+		lineprefix: "read_bytes",
+		lineoffset: 1,
+		unit:       "requests/sec",
+		calc:       "derivative",
+	},
+	{
+		name:       "lustre_write_requests_rate",
+		lineprefix: "write_bytes",
+		lineoffset: 1,
+		unit:       "requests/sec",
+		calc:       "derivative",
+	},
+	{
+		name:       "lustre_read_bw",
+		lineprefix: "read_bytes",
+		lineoffset: 6,
+		unit:       "bytes/sec",
+		calc:       "derivative",
+	},
+	{
+		name:       "lustre_write_bw",
+		lineprefix: "write_bytes",
+		lineoffset: 6,
+		unit:       "bytes/sec",
+		calc:       "derivative",
+	},
+}
+
 func (m *LustreCollector) Init(config json.RawMessage) error {
 	var err error
 	m.name = "LustreCollector"
@@ -99,42 +299,28 @@ func (m *LustreCollector) Init(config json.RawMessage) error {
 	m.setup()
 	m.tags = map[string]string{"type": "node"}
 	m.meta = map[string]string{"source": m.name, "group": "Lustre"}
-	defmatches := map[string]map[string]int{
-		"read_bytes":       {"lustre_read_bytes": 6, "lustre_read_requests": 1},
-		"write_bytes":      {"lustre_write_bytes": 6, "lustre_write_requests": 1},
-		"open":             {"lustre_open": 1},
-		"close":            {"lustre_close": 1},
-		"setattr":          {"lustre_setattr": 1},
-		"getattr":          {"lustre_getattr": 1},
-		"statfs":           {"lustre_statfs": 1},
-		"inode_permission": {"lustre_inode_permission": 1}}
 
 	// Lustre file system statistics can only be queried by user root
-	user, err := user.Current()
-	if err != nil {
-		cclog.ComponentError(m.name, "Failed to get current user:", err.Error())
-		return err
-	}
-	if user.Uid != "0" {
-		cclog.ComponentError(m.name, "Lustre file system statistics can only be queried by user root")
-		return err
+	// or with password-less sudo
+	if !m.config.Sudo {
+		user, err := user.Current()
+		if err != nil {
+			cclog.ComponentError(m.name, "Failed to get current user:", err.Error())
+			return err
+		}
+		if user.Uid != "0" {
+			cclog.ComponentError(m.name, "Lustre file system statistics can only be queried by user root")
+			return err
+		}
+	} else {
+		p, err := exec.LookPath("sudo")
+		if err != nil {
+			cclog.ComponentError(m.name, "Cannot find 'sudo'")
+			return err
+		}
+		m.sudoCmd = p
 	}
 
-	m.matches = make(map[string]map[string]int)
-	for lineprefix, names := range defmatches {
-		for metricname, offset := range names {
-			_, skip := stringArrayContains(m.config.ExcludeMetrics, metricname)
-			if skip {
-				continue
-			}
-			if _, prefixExist := m.matches[lineprefix]; !prefixExist {
-				m.matches[lineprefix] = make(map[string]int)
-			}
-			if _, metricExist := m.matches[lineprefix][metricname]; !metricExist {
-				m.matches[lineprefix][metricname] = offset
-			}
-		}
-	}
 	p, err := exec.LookPath(m.config.LCtlCommand)
 	if err != nil {
 		p, err = exec.LookPath(LCTL_CMD)
@@ -143,26 +329,52 @@ func (m *LustreCollector) Init(config json.RawMessage) error {
 		}
 	}
 	m.lctl = p
-	if m.config.Sudo {
-		p, err := exec.LookPath("sudo")
-		if err != nil {
-			m.sudoCmd = p
+
+	m.definitions = []LustreMetricDefinition{}
+	if m.config.SendAbsoluteValues {
+		for _, def := range LustreAbsMetrics {
+			if _, skip := stringArrayContains(m.config.ExcludeMetrics, def.name); !skip {
+				m.definitions = append(m.definitions, def)
+			}
 		}
 	}
+	if m.config.SendDiffValues {
+		for _, def := range LustreDiffMetrics {
+			if _, skip := stringArrayContains(m.config.ExcludeMetrics, def.name); !skip {
+				m.definitions = append(m.definitions, def)
+			}
+		}
+	}
+	if m.config.SendDerivedValues {
+		for _, def := range LustreDeriveMetrics {
+			if _, skip := stringArrayContains(m.config.ExcludeMetrics, def.name); !skip {
+				m.definitions = append(m.definitions, def)
+			}
+		}
+	}
+	if len(m.definitions) == 0 {
+		return errors.New("no metrics to collect")
+	}
 
 	devices := m.getDevices()
 	if len(devices) == 0 {
-		return errors.New("no metrics to collect")
+		return errors.New("no Lustre devices found")
 	}
 	m.stats = make(map[string]map[string]int64)
 	for _, d := range devices {
 		m.stats[d] = make(map[string]int64)
-		for _, names := range m.matches {
-			for metricname := range names {
-				m.stats[d][metricname] = 0
+		data := m.getDeviceDataCommand(d)
+		for _, def := range m.definitions {
+			x, err := getMetricData(data, def.lineprefix, def.lineoffset)
+			if err == nil {
+				m.stats[d][def.name] = x
+			} else {
+				m.stats[d][def.name] = 0
 			}
 		}
 	}
+	m.lastTimestamp = time.Now()
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -171,54 +383,51 @@ func (m *LustreCollector) Read(interval time.Duration, output chan lp.CCMetric)
 	if !m.init {
 		return
 	}
+	now := time.Now()
+	tdiff := now.Sub(m.lastTimestamp)
 	for device, devData := range m.stats {
-		stats := m.getDeviceDataCommand(device)
-		processed := []string{}
-
-		for _, line := range stats {
-			lf := strings.Fields(line)
-			if len(lf) > 1 {
-				if fields, ok := m.matches[lf[0]]; ok {
-					for name, idx := range fields {
-						x, err := strconv.ParseInt(lf[idx], 0, 64)
-						if err != nil {
-							continue
-						}
-						value := x - devData[name]
-						devData[name] = x
-						if value < 0 {
-							value = 0
-						}
-						y, err := lp.New(name, m.tags, m.meta, map[string]interface{}{"value": value}, time.Now())
-						if err == nil {
-							y.AddTag("device", device)
-							if strings.Contains(name, "byte") {
-								y.AddMeta("unit", "Byte")
-							}
-							output <- y
-							if m.config.SendAllMetrics {
-								processed = append(processed, name)
-							}
-						}
-					}
-				}
+		data := m.getDeviceDataCommand(device)
+		for _, def := range m.definitions {
+			var use_x int64
+			var err error
+			var y lp.CCMetric
+			x, err := getMetricData(data, def.lineprefix, def.lineoffset)
+			if err == nil {
+				use_x = x
+			} else {
+				use_x = devData[def.name]
 			}
-		}
-		if m.config.SendAllMetrics {
-			for name := range devData {
-				if _, done := stringArrayContains(processed, name); !done {
-					y, err := lp.New(name, m.tags, m.meta, map[string]interface{}{"value": 0}, time.Now())
-					if err == nil {
-						y.AddTag("device", device)
-						if strings.Contains(name, "byte") {
-							y.AddMeta("unit", "Byte")
-						}
-						output <- y
-					}
+			var value interface{}
+			switch def.calc {
+			case "none":
+				value = use_x
+				y, err = lp.New(def.name, m.tags, m.meta, map[string]interface{}{"value": value}, time.Now())
+			case "difference":
+				value = use_x - devData[def.name]
+				if value.(int64) < 0 {
+					value = 0
 				}
+				y, err = lp.New(def.name, m.tags, m.meta, map[string]interface{}{"value": value}, time.Now())
+			case "derivative":
+				value = float64(use_x-devData[def.name]) / tdiff.Seconds()
+				if value.(float64) < 0 {
+					value = 0
+				}
+				y, err = lp.New(def.name, m.tags, m.meta, map[string]interface{}{"value": value}, time.Now())
 			}
+			if err == nil {
+				y.AddTag("device", device)
+				if len(def.unit) > 0 {
+					y.AddMeta("unit", def.unit)
+				}
+				output <- y
+				m.statsProcessedMetrics++
+			}
+			devData[def.name] = use_x
 		}
 	}
+	m.lastTimestamp = now
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 }
 
 func (m *LustreCollector) Close() {

collectors/lustreMetric.md

@@ -3,27 +3,44 @@
 
 ```json
   "lustrestat": {
-    "procfiles" : [
-      "/proc/fs/lustre/llite/lnec-XXXXXX/stats"
-    ],
+    "lctl_command": "/path/to/lctl",
     "exclude_metrics": [
       "setattr",
       "getattr"
-    ]
+    ],
+    "send_abs_values" : true,
+    "send_derived_values" : true,
+    "send_diff_values": true,
+    "use_sudo": false
   }
 ```
 
-The `lustrestat` collector reads from the procfs stat files for Lustre like `/proc/fs/lustre/llite/lnec-XXXXXX/stats`.
+The `lustrestat` collector uses the `lctl` application with the `get_param` option to get all `llite` metrics (Lustre client). The `llite` metrics are only available for root users. If password-less sudo is configured, you can enable `sudo` in the configuration.
 
 Metrics:
-* `read_bytes`
-* `read_requests`
-* `write_bytes`
-* `write_requests`
-* `open`
-* `close`
-* `getattr`
-* `setattr`
-* `statfs`
-* `inode_permission`
+* `lustre_read_bytes` (unit `bytes`)
+* `lustre_read_requests` (unit `requests`)
+* `lustre_write_bytes` (unit `bytes`)
+* `lustre_write_requests` (unit `requests`)
+* `lustre_open`
+* `lustre_close`
+* `lustre_getattr`
+* `lustre_setattr`
+* `lustre_statfs`
+* `lustre_inode_permission`
+* `lustre_read_bw` (if `send_derived_values == true`, unit `bytes/sec`)
+* `lustre_write_bw` (if `send_derived_values == true`, unit `bytes/sec`)
+* `lustre_read_requests_rate` (if `send_derived_values == true`, unit `requests/sec`)
+* `lustre_write_requests_rate` (if `send_derived_values == true`, unit `requests/sec`)
+* `lustre_read_bytes_diff` (if `send_diff_values == true`, unit `bytes`)
+* `lustre_read_requests_diff` (if `send_diff_values == true`, unit `requests`)
+* `lustre_write_bytes_diff` (if `send_diff_values == true`, unit `bytes`)
+* `lustre_write_requests_diff` (if `send_diff_values == true`, unit `requests`)
+* `lustre_open_diff` (if `send_diff_values == true`)
+* `lustre_close_diff` (if `send_diff_values == true`)
+* `lustre_getattr_diff` (if `send_diff_values == true`)
+* `lustre_setattr_diff` (if `send_diff_values == true`)
+* `lustre_statfs_diff` (if `send_diff_values == true`)
+* `lustre_inode_permission_diff` (if `send_diff_values == true`)
 
+This collector adds an `device` tag.

collectors/memstatMetric.go

@@ -14,6 +14,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 const MEMSTATFILE = "/proc/meminfo"
@@ -32,15 +33,22 @@ type MemstatCollectorNode struct {
 
 type MemstatCollector struct {
 	metricCollector
-	stats     map[string]int64
-	tags      map[string]string
-	matches   map[string]string
-	config    MemstatCollectorConfig
-	nodefiles map[int]MemstatCollectorNode
+	stats                 map[string]int64
+	tags                  map[string]string
+	matches               map[string]string
+	config                MemstatCollectorConfig
+	nodefiles             map[int]MemstatCollectorNode
+	sendMemUsed           bool
+	statsProcessedMetrics int64
 }
 
-func getStats(filename string) map[string]float64 {
-	stats := make(map[string]float64)
+type MemstatStats struct {
+	value float64
+	unit  string
+}
+
+func getStats(filename string) map[string]MemstatStats {
+	stats := make(map[string]MemstatStats)
 	file, err := os.Open(filename)
 	if err != nil {
 		cclog.Error(err.Error())
@@ -54,12 +62,18 @@ func getStats(filename string) map[string]float64 {
 		if len(linefields) == 3 {
 			v, err := strconv.ParseFloat(linefields[1], 64)
 			if err == nil {
-				stats[strings.Trim(linefields[0], ":")] = v
+				stats[strings.Trim(linefields[0], ":")] = MemstatStats{
+					value: v,
+					unit:  linefields[2],
+				}
 			}
 		} else if len(linefields) == 5 {
 			v, err := strconv.ParseFloat(linefields[3], 64)
 			if err == nil {
-				stats[strings.Trim(linefields[0], ":")] = v
+				stats[strings.Trim(linefields[0], ":")] = MemstatStats{
+					value: v,
+					unit:  linefields[4],
+				}
 			}
 		}
 	}
@@ -77,7 +91,7 @@ func (m *MemstatCollector) Init(config json.RawMessage) error {
 			return err
 		}
 	}
-	m.meta = map[string]string{"source": m.name, "group": "Memory", "unit": "kByte"}
+	m.meta = map[string]string{"source": m.name, "group": "Memory"}
 	m.stats = make(map[string]int64)
 	m.matches = make(map[string]string)
 	m.tags = map[string]string{"type": "node"}
@@ -99,6 +113,10 @@ func (m *MemstatCollector) Init(config json.RawMessage) error {
 			m.matches[k] = v
 		}
 	}
+	m.sendMemUsed = false
+	if _, skip := stringArrayContains(m.config.ExcludeMetrics, "mem_used"); !skip {
+		m.sendMemUsed = true
+	}
 	if len(m.matches) == 0 {
 		return errors.New("no metrics to collect")
 	}
@@ -137,6 +155,7 @@ func (m *MemstatCollector) Init(config json.RawMessage) error {
 			}
 		}
 	}
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return err
 }
@@ -146,29 +165,55 @@ func (m *MemstatCollector) Read(interval time.Duration, output chan lp.CCMetric)
 		return
 	}
 
-	sendStats := func(stats map[string]float64, tags map[string]string) {
+	sendStats := func(stats map[string]MemstatStats, tags map[string]string) {
 		for match, name := range m.matches {
 			var value float64 = 0
+			var unit string = ""
 			if v, ok := stats[match]; ok {
-				value = v
+				value = v.value
+				if len(v.unit) > 0 {
+					unit = v.unit
+				}
 			}
+
 			y, err := lp.New(name, tags, m.meta, map[string]interface{}{"value": value}, time.Now())
 			if err == nil {
+				if len(unit) > 0 {
+					y.AddMeta("unit", unit)
+				}
+				m.statsProcessedMetrics++
 				output <- y
 			}
 		}
-		if _, skip := stringArrayContains(m.config.ExcludeMetrics, "mem_used"); !skip {
-			if freeVal, free := stats["MemFree"]; free {
-				if bufVal, buffers := stats["Buffers"]; buffers {
-					if cacheVal, cached := stats["Cached"]; cached {
-						memUsed := stats["MemTotal"] - (freeVal + bufVal + cacheVal)
-						y, err := lp.New("mem_used", tags, m.meta, map[string]interface{}{"value": memUsed}, time.Now())
-						if err == nil {
-							output <- y
+		if m.sendMemUsed {
+			memUsed := 0.0
+			unit := ""
+			if totalVal, total := stats["MemTotal"]; total {
+				if freeVal, free := stats["MemFree"]; free {
+					if bufVal, buffers := stats["Buffers"]; buffers {
+						if cacheVal, cached := stats["Cached"]; cached {
+							memUsed = totalVal.value - (freeVal.value + bufVal.value + cacheVal.value)
+							if len(totalVal.unit) > 0 {
+								unit = totalVal.unit
+							} else if len(freeVal.unit) > 0 {
+								unit = freeVal.unit
+							} else if len(bufVal.unit) > 0 {
+								unit = bufVal.unit
+							} else if len(cacheVal.unit) > 0 {
+								unit = cacheVal.unit
+							}
 						}
 					}
 				}
 			}
+			y, err := lp.New("mem_used", tags, m.meta, map[string]interface{}{"value": memUsed}, time.Now())
+			if err == nil {
+				if len(unit) > 0 {
+					y.AddMeta("unit", unit)
+				}
+				m.statsProcessedMetrics++
+				output <- y
+			}
 		}
 	}
 
@@ -183,6 +228,7 @@ func (m *MemstatCollector) Read(interval time.Duration, output chan lp.CCMetric)
 			sendStats(stats, nodeConf.tags)
 		}
 	}
+	stats.ComponentStatInt(m.name, "collected_metrics", m.statsProcessedMetrics)
 }
 
 func (m *MemstatCollector) Close() {

collectors/metricCollector.go

@@ -125,5 +125,5 @@ func RemoveFromStringList(s []string, r string) ([]string, error) {
 			return append(s[:i], s[i+1:]...), nil
 		}
 	}
-	return s, fmt.Errorf("No such string in list")
+	return s, fmt.Errorf("no such string in list")
 }

collectors/netstatMetric.go

@@ -11,40 +11,65 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
-const NETSTATFILE = `/proc/net/dev`
+const NETSTATFILE = "/proc/net/dev"
 
 type NetstatCollectorConfig struct {
-	IncludeDevices []string `json:"include_devices"`
+	IncludeDevices     []string `json:"include_devices"`
+	SendAbsoluteValues bool     `json:"send_abs_values"`
+	SendDerivedValues  bool     `json:"send_derived_values"`
 }
 
 type NetstatCollectorMetric struct {
-	index     int
-	lastValue float64
+	name       string
+	index      int
+	tags       map[string]string
+	meta       map[string]string
+	meta_rates map[string]string
+	lastValue  int64
 }
 
 type NetstatCollector struct {
 	metricCollector
-	config        NetstatCollectorConfig
-	matches       map[string]map[string]NetstatCollectorMetric
-	devtags       map[string]map[string]string
-	lastTimestamp time.Time
+	config                NetstatCollectorConfig
+	matches               map[string][]NetstatCollectorMetric
+	lastTimestamp         time.Time
+	statsProcessedMetrics int64
 }
 
 func (m *NetstatCollector) Init(config json.RawMessage) error {
 	m.name = "NetstatCollector"
 	m.setup()
 	m.lastTimestamp = time.Now()
-	m.meta = map[string]string{"source": m.name, "group": "Network"}
-	m.devtags = make(map[string]map[string]string)
-	nameIndexMap := map[string]int{
-		"net_bytes_in":  1,
-		"net_pkts_in":   2,
-		"net_bytes_out": 9,
-		"net_pkts_out":  10,
-	}
-	m.matches = make(map[string]map[string]NetstatCollectorMetric)
+
+	const (
+		fieldInterface = iota
+		fieldReceiveBytes
+		fieldReceivePackets
+		fieldReceiveErrs
+		fieldReceiveDrop
+		fieldReceiveFifo
+		fieldReceiveFrame
+		fieldReceiveCompressed
+		fieldReceiveMulticast
+		fieldTransmitBytes
+		fieldTransmitPackets
+		fieldTransmitErrs
+		fieldTransmitDrop
+		fieldTransmitFifo
+		fieldTransmitColls
+		fieldTransmitCarrier
+		fieldTransmitCompressed
+	)
+
+	m.matches = make(map[string][]NetstatCollectorMetric)
+
+	// Set default configuration,
+	m.config.SendAbsoluteValues = true
+	m.config.SendDerivedValues = false
+	// Read configuration file, allow overwriting default config
 	if len(config) > 0 {
 		err := json.Unmarshal(config, &m.config)
 		if err != nil {
@@ -52,7 +77,9 @@ func (m *NetstatCollector) Init(config json.RawMessage) error {
 			return err
 		}
 	}
-	file, err := os.Open(string(NETSTATFILE))
+
+	// Check access to net statistic file
+	file, err := os.Open(NETSTATFILE)
 	if err != nil {
 		cclog.ComponentError(m.name, err.Error())
 		return err
@@ -62,25 +89,68 @@ func (m *NetstatCollector) Init(config json.RawMessage) error {
 	scanner := bufio.NewScanner(file)
 	for scanner.Scan() {
 		l := scanner.Text()
+
+		// Skip lines with no net device entry
 		if !strings.Contains(l, ":") {
 			continue
 		}
+
+		// Split line into fields
 		f := strings.Fields(l)
+
+		// Get net device entry
 		dev := strings.Trim(f[0], ": ")
+
+		// Check if device is a included device
 		if _, ok := stringArrayContains(m.config.IncludeDevices, dev); ok {
-			m.matches[dev] = make(map[string]NetstatCollectorMetric)
-			for name, idx := range nameIndexMap {
-				m.matches[dev][name] = NetstatCollectorMetric{
-					index:     idx,
-					lastValue: 0,
-				}
+			tags := map[string]string{"device": dev, "type": "node"}
+			meta_unit_byte := map[string]string{"source": m.name, "group": "Network", "unit": "bytes"}
+			meta_unit_byte_per_sec := map[string]string{"source": m.name, "group": "Network", "unit": "bytes/sec"}
+			meta_unit_pkts := map[string]string{"source": m.name, "group": "Network", "unit": "packets"}
+			meta_unit_pkts_per_sec := map[string]string{"source": m.name, "group": "Network", "unit": "packets/sec"}
+
+			m.matches[dev] = []NetstatCollectorMetric{
+				{
+					name:       "net_bytes_in",
+					index:      fieldReceiveBytes,
+					lastValue:  -1,
+					tags:       tags,
+					meta:       meta_unit_byte,
+					meta_rates: meta_unit_byte_per_sec,
+				},
+				{
+					name:       "net_pkts_in",
+					index:      fieldReceivePackets,
+					lastValue:  -1,
+					tags:       tags,
+					meta:       meta_unit_pkts,
+					meta_rates: meta_unit_pkts_per_sec,
+				},
+				{
+					name:       "net_bytes_out",
+					index:      fieldTransmitBytes,
+					lastValue:  -1,
+					tags:       tags,
+					meta:       meta_unit_byte,
+					meta_rates: meta_unit_byte_per_sec,
+				},
+				{
+					name:       "net_pkts_out",
+					index:      fieldTransmitPackets,
+					lastValue:  -1,
+					tags:       tags,
+					meta:       meta_unit_pkts,
+					meta_rates: meta_unit_pkts_per_sec,
+				},
 			}
-			m.devtags[dev] = map[string]string{"device": dev, "type": "node"}
 		}
+
 	}
-	if len(m.devtags) == 0 {
+
+	if len(m.matches) == 0 {
 		return errors.New("no devices to collector metrics found")
 	}
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -89,50 +159,65 @@ func (m *NetstatCollector) Read(interval time.Duration, output chan lp.CCMetric)
 	if !m.init {
 		return
 	}
+	// Current time stamp
 	now := time.Now()
+	// time difference to last time stamp
+	timeDiff := now.Sub(m.lastTimestamp).Seconds()
+	// Save current timestamp
+	m.lastTimestamp = now
+
 	file, err := os.Open(string(NETSTATFILE))
 	if err != nil {
 		cclog.ComponentError(m.name, err.Error())
 		return
 	}
 	defer file.Close()
-	tdiff := now.Sub(m.lastTimestamp)
 
 	scanner := bufio.NewScanner(file)
 	for scanner.Scan() {
 		l := scanner.Text()
+
+		// Skip lines with no net device entry
 		if !strings.Contains(l, ":") {
 			continue
 		}
+
+		// Split line into fields
 		f := strings.Fields(l)
+
+		// Get net device entry
 		dev := strings.Trim(f[0], ":")
 
+		// Check if device is a included device
 		if devmetrics, ok := m.matches[dev]; ok {
-			for name, data := range devmetrics {
-				v, err := strconv.ParseFloat(f[data.index], 64)
-				if err == nil {
-					vdiff := v - data.lastValue
-					value := vdiff / tdiff.Seconds()
-					if data.lastValue == 0 {
-						value = 0
-					}
-					data.lastValue = v
-					y, err := lp.New(name, m.devtags[dev], m.meta, map[string]interface{}{"value": value}, now)
-					if err == nil {
-						switch {
-						case strings.Contains(name, "byte"):
-							y.AddMeta("unit", "bytes/sec")
-						case strings.Contains(name, "pkt"):
-							y.AddMeta("unit", "packets/sec")
-						}
+			for i := range devmetrics {
+				metric := &devmetrics[i]
+
+				// Read value
+				v, err := strconv.ParseInt(f[metric.index], 10, 64)
+				if err != nil {
+					continue
+				}
+				if m.config.SendAbsoluteValues {
+					if y, err := lp.New(metric.name, metric.tags, metric.meta, map[string]interface{}{"value": v}, now); err == nil {
 						output <- y
+						m.statsProcessedMetrics++
 					}
-					devmetrics[name] = data
+				}
+				if m.config.SendDerivedValues {
+					if metric.lastValue >= 0 {
+						rate := float64(v-metric.lastValue) / timeDiff
+						if y, err := lp.New(metric.name+"_bw", metric.tags, metric.meta_rates, map[string]interface{}{"value": rate}, now); err == nil {
+							output <- y
+							m.statsProcessedMetrics++
+						}
+					}
+					metric.lastValue = v
 				}
 			}
 		}
 	}
-	m.lastTimestamp = time.Now()
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 }
 
 func (m *NetstatCollector) Close() {

collectors/netstatMetric.md

@@ -5,17 +5,23 @@
   "netstat": {
     "include_devices": [
       "eth0"
-    ]
+    ],
+    "send_abs_values" : true,
+    "send_derived_values" : true
   }
 ```
 
 The `netstat` collector reads data from `/proc/net/dev` and outputs a handful **node** metrics. With the `include_devices` list you can specify which network devices should be measured. **Note**: Most other collectors use an _exclude_ list instead of an include list.
 
 Metrics:
-* `net_bytes_in` (`unit=bytes/sec`)
-* `net_bytes_out` (`unit=bytes/sec`)
-* `net_pkts_in` (`unit=packets/sec`)
-* `net_pkts_out` (`unit=packets/sec`)
+* `net_bytes_in` (`unit=bytes`)
+* `net_bytes_out` (`unit=bytes`)
+* `net_pkts_in` (`unit=packets`)
+* `net_pkts_out` (`unit=packets`)
+* `net_bytes_in_bw` (`unit=bytes/sec` if `send_derived_values == true`)
+* `net_bytes_out_bw` (`unit=bytes/sec` if `send_derived_values == true`)
+* `net_pkts_in_bw` (`unit=packets/sec` if `send_derived_values == true`)
+* `net_pkts_out_bw` (`unit=packets/sec` if `send_derived_values == true`)
 
 The device name is added as tag `device`.
 

collectors/nfsMetric.go

@@ -12,6 +12,7 @@ import (
 	"time"
 
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 // First part contains the code for the general NfsCollector.
@@ -32,11 +33,12 @@ type nfsCollector struct {
 		Nfsstats       string   `json:"nfsstat"`
 		ExcludeMetrics []string `json:"exclude_metrics,omitempty"`
 	}
-	data map[string]NfsCollectorData
+	data                  map[string]NfsCollectorData
+	statsProcessedMetrics int64
 }
 
 func (m *nfsCollector) initStats() error {
-	cmd := exec.Command(m.config.Nfsstats, `-l`)
+	cmd := exec.Command(m.config.Nfsstats, `-l`, `--all`)
 	cmd.Wait()
 	buffer, err := cmd.Output()
 	if err == nil {
@@ -52,7 +54,7 @@ func (m *nfsCollector) initStats() error {
 					if err == nil {
 						x := m.data[name]
 						x.current = value
-						x.last = 0
+						x.last = value
 						m.data[name] = x
 					}
 				}
@@ -63,7 +65,7 @@ func (m *nfsCollector) initStats() error {
 }
 
 func (m *nfsCollector) updateStats() error {
-	cmd := exec.Command(m.config.Nfsstats, `-l`)
+	cmd := exec.Command(m.config.Nfsstats, `-l`, `--all`)
 	cmd.Wait()
 	buffer, err := cmd.Output()
 	if err == nil {
@@ -113,6 +115,7 @@ func (m *nfsCollector) MainInit(config json.RawMessage) error {
 	}
 	m.data = make(map[string]NfsCollectorData)
 	m.initStats()
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -143,8 +146,10 @@ func (m *nfsCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 		if err == nil {
 			y.AddMeta("version", m.version)
 			output <- y
+			m.statsProcessedMetrics++
 		}
 	}
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 }
 
 func (m *nfsCollector) Close() {

collectors/numastatsMetric.go

@@ -12,6 +12,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 //
@@ -44,7 +45,8 @@ type NUMAStatsCollectorTopolgy struct {
 
 type NUMAStatsCollector struct {
 	metricCollector
-	topology []NUMAStatsCollectorTopolgy
+	topology              []NUMAStatsCollectorTopolgy
+	statsProcessedMetrics int64
 }
 
 func (m *NUMAStatsCollector) Init(config json.RawMessage) error {
@@ -80,7 +82,7 @@ func (m *NUMAStatsCollector) Init(config json.RawMessage) error {
 				tagSet: map[string]string{"memoryDomain": node},
 			})
 	}
-
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -127,11 +129,13 @@ func (m *NUMAStatsCollector) Read(interval time.Duration, output chan lp.CCMetri
 			)
 			if err == nil {
 				output <- y
+				m.statsProcessedMetrics++
 			}
 		}
 
 		file.Close()
 	}
+	stats.ComponentStatInt(m.name, "collected_metrics", m.statsProcessedMetrics)
 }
 
 func (m *NUMAStatsCollector) Close() {

collectors/nvidiaMetric.go

@@ -9,6 +9,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 	"github.com/NVIDIA/go-nvml/pkg/nvml"
 )
 
@@ -26,9 +27,10 @@ type NvidiaCollectorDevice struct {
 
 type NvidiaCollector struct {
 	metricCollector
-	num_gpus int
-	config   NvidiaCollectorConfig
-	gpus     []NvidiaCollectorDevice
+	num_gpus              int
+	config                NvidiaCollectorConfig
+	gpus                  []NvidiaCollectorDevice
+	statsProcessedMetrics int64
 }
 
 func (m *NvidiaCollector) CatchPanic() {
@@ -120,7 +122,7 @@ func (m *NvidiaCollector) Init(config json.RawMessage) error {
 				pciInfo.Device)
 		}
 	}
-
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -151,6 +153,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 					if err == nil {
 						y.AddMeta("unit", "%")
 						output <- y
+						m.statsProcessedMetrics++
 					}
 				}
 				if !device.excludeMetrics["nv_mem_util"] {
@@ -158,6 +161,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 					if err == nil {
 						y.AddMeta("unit", "%")
 						output <- y
+						m.statsProcessedMetrics++
 					}
 				}
 			}
@@ -186,6 +190,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 					if err == nil {
 						y.AddMeta("unit", "MByte")
 						output <- y
+						m.statsProcessedMetrics++
 					}
 				}
 
@@ -195,6 +200,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 					if err == nil {
 						y.AddMeta("unit", "MByte")
 						output <- y
+						m.statsProcessedMetrics++
 					}
 				}
 			}
@@ -212,6 +218,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				if err == nil {
 					y.AddMeta("unit", "degC")
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -232,6 +239,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				if err == nil {
 					y.AddMeta("unit", "%")
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -258,11 +266,13 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				}
 				if err == nil {
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			} else if ret == nvml.ERROR_NOT_SUPPORTED {
 				y, err := lp.New("nv_ecc_mode", device.tags, m.meta, map[string]interface{}{"value": "N/A"}, time.Now())
 				if err == nil {
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -280,6 +290,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				y, err := lp.New("nv_perf_state", device.tags, m.meta, map[string]interface{}{"value": fmt.Sprintf("P%d", int(pState))}, time.Now())
 				if err == nil {
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -296,6 +307,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				if err == nil {
 					y.AddMeta("unit", "watts")
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -313,6 +325,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				if err == nil {
 					y.AddMeta("unit", "MHz")
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -324,6 +337,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				if err == nil {
 					y.AddMeta("unit", "MHz")
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -335,6 +349,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				if err == nil {
 					y.AddMeta("unit", "MHz")
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -357,6 +372,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				if err == nil {
 					y.AddMeta("unit", "MHz")
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -368,6 +384,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				if err == nil {
 					y.AddMeta("unit", "MHz")
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -379,6 +396,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				if err == nil {
 					y.AddMeta("unit", "MHz")
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -398,6 +416,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				y, err := lp.New("nv_ecc_db_error", device.tags, m.meta, map[string]interface{}{"value": float64(ecc_db)}, time.Now())
 				if err == nil {
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -408,6 +427,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				y, err := lp.New("nv_ecc_sb_error", device.tags, m.meta, map[string]interface{}{"value": float64(ecc_sb)}, time.Now())
 				if err == nil {
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -425,6 +445,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				if err == nil {
 					y.AddMeta("unit", "watts")
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -441,6 +462,7 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				if err == nil {
 					y.AddMeta("unit", "%")
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
@@ -457,11 +479,12 @@ func (m *NvidiaCollector) Read(interval time.Duration, output chan lp.CCMetric)
 				if err == nil {
 					y.AddMeta("unit", "%")
 					output <- y
+					m.statsProcessedMetrics++
 				}
 			}
 		}
 	}
-
+	stats.ComponentStatInt(m.name, "collected_metrics", m.statsProcessedMetrics)
 }
 
 func (m *NvidiaCollector) Close() {

collectors/sampleMetric.go

@@ -6,6 +6,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 // These are the fields we read from the JSON configuration
@@ -17,9 +18,10 @@ type SampleCollectorConfig struct {
 // defined by metricCollector (name, init, ...)
 type SampleCollector struct {
 	metricCollector
-	config SampleTimerCollectorConfig // the configuration structure
-	meta   map[string]string          // default meta information
-	tags   map[string]string          // default tags
+	config     SampleTimerCollectorConfig // the configuration structure
+	meta       map[string]string          // default meta information
+	tags       map[string]string          // default tags
+	statsCount int64
 }
 
 // Functions to implement MetricCollector interface
@@ -58,6 +60,9 @@ func (m *SampleCollector) Init(config json.RawMessage) error {
 	// for all topological entities (sockets, NUMA domains, ...)
 	// Return some useful error message in case of any failures
 
+	// Initialize counts for statistics
+	m.statsCount = 0
+
 	// Set this flag only if everything is initialized properly, all required files exist, ...
 	m.init = true
 	return err
@@ -80,8 +85,11 @@ func (m *SampleCollector) Read(interval time.Duration, output chan lp.CCMetric)
 	if err == nil {
 		// Send it to output channel
 		output <- y
+		// increment count for each sent metric or any other operation
+		m.statsCount++
 	}
-
+	// Set stats for the component
+	stats.ComponentStatInt(m.name, "count", m.statsCount)
 }
 
 // Close metric collector: close network connection, close files, close libraries, ...

collectors/tempMetric.go

@@ -11,6 +11,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 // See: https://www.kernel.org/doc/html/latest/hwmon/sysfs-interface.html
@@ -40,7 +41,8 @@ type TempCollector struct {
 		ReportMaxTemp      bool                         `json:"report_max_temperature"`
 		ReportCriticalTemp bool                         `json:"report_critical_temperature"`
 	}
-	sensors []*TempCollectorSensor
+	sensors               []*TempCollectorSensor
+	statsProcessedMetrics int64
 }
 
 func (m *TempCollector) Init(config json.RawMessage) error {
@@ -70,10 +72,10 @@ func (m *TempCollector) Init(config json.RawMessage) error {
 	globPattern := filepath.Join("/sys/class/hwmon", "*", "temp*_input")
 	inputFiles, err := filepath.Glob(globPattern)
 	if err != nil {
-		return fmt.Errorf("Unable to glob files with pattern '%s': %v", globPattern, err)
+		return fmt.Errorf("unable to glob files with pattern '%s': %v", globPattern, err)
 	}
 	if inputFiles == nil {
-		return fmt.Errorf("Unable to find any files with pattern '%s'", globPattern)
+		return fmt.Errorf("unable to find any files with pattern '%s'", globPattern)
 	}
 
 	// Get sensor name for each temperature sensor file
@@ -158,10 +160,11 @@ func (m *TempCollector) Init(config json.RawMessage) error {
 
 	// Empty sensors map
 	if len(m.sensors) == 0 {
-		return fmt.Errorf("No temperature sensors found")
+		return fmt.Errorf("no temperature sensors found")
 	}
 
 	// Finished initialization
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -194,6 +197,7 @@ func (m *TempCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 		)
 		if err == nil {
 			output <- y
+			m.statsProcessedMetrics++
 		}
 
 		// max temperature
@@ -207,6 +211,7 @@ func (m *TempCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 			)
 			if err == nil {
 				output <- y
+				m.statsProcessedMetrics++
 			}
 		}
 
@@ -221,10 +226,11 @@ func (m *TempCollector) Read(interval time.Duration, output chan lp.CCMetric) {
 			)
 			if err == nil {
 				output <- y
+				m.statsProcessedMetrics++
 			}
 		}
 	}
-
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 }
 
 func (m *TempCollector) Close() {

collectors/topprocsMetric.go

@@ -10,6 +10,7 @@ import (
 	"time"
 
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 const MAX_NUM_PROCS = 10
@@ -21,8 +22,9 @@ type TopProcsCollectorConfig struct {
 
 type TopProcsCollector struct {
 	metricCollector
-	tags   map[string]string
-	config TopProcsCollectorConfig
+	tags                  map[string]string
+	config                TopProcsCollectorConfig
+	statsProcessedMetrics int64
 }
 
 func (m *TopProcsCollector) Init(config json.RawMessage) error {
@@ -39,15 +41,16 @@ func (m *TopProcsCollector) Init(config json.RawMessage) error {
 		m.config.Num_procs = int(DEFAULT_NUM_PROCS)
 	}
 	if m.config.Num_procs <= 0 || m.config.Num_procs > MAX_NUM_PROCS {
-		return errors.New(fmt.Sprintf("num_procs option must be set in 'topprocs' config (range: 1-%d)", MAX_NUM_PROCS))
+		return fmt.Errorf("num_procs option must be set in 'topprocs' config (range: 1-%d)", MAX_NUM_PROCS)
 	}
 	m.setup()
 	command := exec.Command("ps", "-Ao", "comm", "--sort=-pcpu")
 	command.Wait()
 	_, err = command.Output()
 	if err != nil {
-		return errors.New("Failed to execute command")
+		return errors.New("failed to execute command")
 	}
+	m.statsProcessedMetrics = 0
 	m.init = true
 	return nil
 }
@@ -70,8 +73,10 @@ func (m *TopProcsCollector) Read(interval time.Duration, output chan lp.CCMetric
 		y, err := lp.New(name, m.tags, m.meta, map[string]interface{}{"value": string(lines[i])}, time.Now())
 		if err == nil {
 			output <- y
+			m.statsProcessedMetrics++
 		}
 	}
+	stats.ComponentStatInt(m.name, "processed_metrics", m.statsProcessedMetrics)
 }
 
 func (m *TopProcsCollector) Close() {

docs/configuration.md

@@ -0,0 +1,189 @@
+# Configuring the CC metric collector
+
+The configuration of the CC metric collector consists of five configuration files: one global file and four component related files.
+
+## Global configuration
+
+The global file contains the paths to the other four files and some global options. 
+
+```json
+{
+  "sinks": "sinks.json",
+  "collectors" : "collectors.json",
+  "receivers" : "receivers.json",
+  "router" : "router.json",
+  "interval": 10,
+  "duration": 1
+}
+```
+
+Be aware that the paths are relative to the execution folder of the cc-metric-collector binary, so it is recommended to use absolute paths.
+
+## Component configuration
+
+The others are mainly list of of subcomponents: the collectors, the receivers, the router and the sinks. Their role is best shown in a picture:
+
+```mermaid
+flowchart LR
+
+  subgraph col ["Collectors"]
+  direction TB
+  cpustat["cpustat"]
+  memstat["memstat"]
+  tempstat["tempstat"]
+  misc["..."]
+  end
+  
+  subgraph Receivers ["Receivers"]
+  direction TB
+  nats["NATS"]
+  httprecv["HTTP"]
+  miscrecv[...]
+  end
+
+  subgraph calc["Aggregator"]
+  direction LR
+  cache["Cache"]
+  agg["Calculator"]
+  end
+
+  subgraph sinks ["Sinks"]
+  direction RL
+  influx["InfluxDB"]
+  ganglia["Ganglia"]
+  logger["Logfile"]
+  miscsink["..."]
+  end
+
+  cpustat --> CollectorManager["CollectorManager"]
+  memstat --> CollectorManager
+  tempstat --> CollectorManager
+  misc --> CollectorManager
+
+  nats  --> ReceiverManager["ReceiverManager"]
+  httprecv --> ReceiverManager
+  miscrecv --> ReceiverManager
+
+  CollectorManager --> newrouter["Router"]
+  ReceiverManager -.-> newrouter
+  calc -.-> newrouter
+  newrouter --> SinkManager["SinkManager"]
+  newrouter -.-> calc
+
+  SinkManager --> influx
+  SinkManager --> ganglia
+  SinkManager --> logger
+  SinkManager --> miscsink
+
+
+```
+
+There are four parts:
+- The collectors read data from files, execute commands and call dynamically loaded library function and send it to the router
+- The router can process metrics by cacheing and evaluating functions and conditions on them
+- The sinks send the metrics to storage backends
+- The receivers can be used to receive metrics from other collectors and forward them to the router. They can be used to create a tree-like structure of collectors.
+
+(A maybe better differentiation between collectors and receivers is that the collectors are called periodically while the receivers have their own logic and submit metrics at any time)
+
+
+### Collectors configuration file
+
+The collectors configuration file tells which metrics should be queried from the system. The metric gathering is logically grouped in so called 'Collectors'. So there are Collectors to read CPU, memory or filesystem statistics. The collectors configuration file is a list of these collectors with collector-specific configurations:
+
+```json
+{
+  "cpustat" : {},
+  "diskstat": {
+    "exclude_metrics": [
+      "disk_total"
+    ]
+  }
+}
+```
+
+The first one is the CPU statistics collector without any collector-specific setting. The second one enables disk mount statistics but excludes the metric `disk_total`.
+
+All names and possible collector-specific configuration options can be found [here](../collectors/README.md).
+
+Some collectors might dynamically load shared libraries. In order to enable these collectors, make sure that the shared library path is part of the `LD_LIBRARY_PATH` environment variable.
+
+### Sinks configuration file
+
+The sinks define the output/sending of metrics. The metrics can be forwarded to multiple sinks, even to sinks of the same type. The sinks configuration file is a list of these sinks, each with an individual name.
+
+```json
+{
+  "myinflux" : {
+    "type" : "influxasync",
+    "host": "localhost",
+    "port": "8086",
+    "organization" : "testorga",
+    "database" : "testbucket",
+    "password" : "<my secret JWT>"
+  },
+  "companyinflux" : {
+    "type" : "influxasync",
+    "host": "companyhost",
+    "port": "8086",
+    "organization" : "company",
+    "database" : "main",
+    "password" : "<company's secret JWT>"
+  }
+}
+```
+
+The above example configuration file defines two sink, both ot type `influxasync`. They are differentiated internally by the names: `myinflux` and `companyinflux`.
+
+All types and possible sink-specific configuration options can be found [here](../sinks/README.md).
+
+Some sinks might dynamically load shared libraries. In order to enable these sinks, make sure that the shared library path is part of the `LD_LIBRARY_PATH` environment variable.
+
+### Router configuration file
+
+The collectors and the sinks are connected through the router. The router forwards the metrics to the sinks but enables some data processing. A common example is to tag all passing metrics like adding `cluster=mycluster`. But also aggregations like "take the average of all 'ipc' metrics" (ipc -> Instructions Per Cycle). Since the configurations of these aggregations can be quite complicated, we refer to the router's [README](../internal/metricRouter/README.md).
+
+A simple router configuration file to start with looks like this:
+
+```json
+{
+    "add_tags" : [
+        {
+            "key" : "cluster",
+            "value" : "mycluster",
+            "if" : "*"
+        }
+    ],
+    "interval_timestamp" : false,
+    "num_cache_intervals" : 0
+}
+```
+
+With the `add_tags` section, we tell to attach the `cluster=mycluster` tag to each (`*` metric). The `interval_timestamp` tell the router to not touch the timestamp of metrics. It is possible to send all metrics within an interval with a common time stamp to avoid later alignment issues. The `num_cache_intervals` diables the cache completely. The cache is only required if you want to do complex metric aggregations.
+
+All configuration options can be found [here](../internal/metricRouter/README.md).
+
+### Receivers configuration file
+
+The receivers are a special feature of the CC Metric Collector to enable simpler integration into exising setups. While collectors query data from the local system, the receivers commonly get data from other systems through some network technology like HTTP or NATS. The idea is keep the current setup but send it to a CC Metric Collector which forwards it to the the destination system (if a sink exists for it). For most setups, the receivers are not required and an the receiver config file should contain only an empty JSON map (`{}`). 
+
+```json
+{
+  "nats_rack0": {
+    "type": "nats",
+    "address" : "nats-server.example.org",
+    "port" : "4222",
+    "subject" : "rack0",
+  },
+  "nats_rack1": {
+    "type": "nats",
+    "address" : "nats-server.example.org",
+    "port" : "4222",
+    "subject" : "rack1",
+  }
+}
+```
+
+This example configuration creates two receivers with the names `nats_rack0` and `nats_rack1`. While one subscribes to metrics published with the `rack0` subject, the other one subscribes to the `rack0` subject. The NATS server is the same as it manages all subjects in a subnet. (As example, the router could add tags `rack=0` and `rack=1` respectively to the received metrics.)
+
+All types and possible receiver-specific configuration options can be found [here](../receivers/README.md).

internal/ccMetric/ccMetric.go

@@ -24,8 +24,8 @@ type ccMetric struct {
 
 // ccMetric access functions
 type CCMetric interface {
-	ToPoint(metaAsTags bool) *write.Point  // Generate influxDB point for data type ccMetric
-	ToLineProtocol(metaAsTags bool) string // Generate influxDB line protocol for data type ccMetric
+	ToPoint(metaAsTags map[string]bool) *write.Point  // Generate influxDB point for data type ccMetric
+	ToLineProtocol(metaAsTags map[string]bool) string // Generate influxDB line protocol for data type ccMetric
 
 	Name() string        // Get metric name
 	SetName(name string) // Set metric name
@@ -61,25 +61,18 @@ func (m *ccMetric) String() string {
 }
 
 // ToLineProtocol generates influxDB line protocol for data type ccMetric
-func (m *ccMetric) ToPoint(metaAsTags bool) (p *write.Point) {
-
-	if !metaAsTags {
-		p = influxdb2.NewPoint(m.name, m.tags, m.fields, m.tm)
-	} else {
-		tags := make(map[string]string, len(m.tags)+len(m.meta))
-		for key, value := range m.tags {
-			tags[key] = value
+func (m *ccMetric) ToPoint(metaAsTags map[string]bool) (p *write.Point) {
+	p = influxdb2.NewPoint(m.name, m.tags, m.fields, m.tm)
+	for key, ok1 := range metaAsTags {
+		if val, ok2 := m.GetMeta(key); ok1 && ok2 {
+			p.AddTag(key, val)
 		}
-		for key, value := range m.meta {
-			tags[key] = value
-		}
-		p = influxdb2.NewPoint(m.name, tags, m.fields, m.tm)
 	}
-	return
+	return p
 }
 
 // ToLineProtocol generates influxDB line protocol for data type ccMetric
-func (m *ccMetric) ToLineProtocol(metaAsTags bool) string {
+func (m *ccMetric) ToLineProtocol(metaAsTags map[string]bool) string {
 
 	return write.PointToLineProtocol(
 		m.ToPoint(metaAsTags),

internal/ccTopology/ccTopology.go

@@ -169,7 +169,10 @@ func DieList() []int {
 			}
 		}
 	}
-	return dielist
+	if len(dielist) > 0 {
+		return dielist
+	}
+	return SocketList()
 }
 
 type CpuEntry struct {
@@ -261,7 +264,7 @@ func CpuData() []CpuEntry {
 	for _, c := range CpuList() {
 		clist = append(clist, CpuEntry{Cpuid: c})
 	}
-	for _, centry := range clist {
+	for i, centry := range clist {
 		centry.Socket = -1
 		centry.Numadomain = -1
 		centry.Die = -1
@@ -289,6 +292,8 @@ func CpuData() []CpuEntry {
 		// Lookup NUMA domain id
 		centry.Numadomain = getNumaDomain(base)
 
+		// Update values in output list
+		clist[i] = centry
 	}
 	return clist
 }

internal/metricRouter/README.md

@@ -8,6 +8,8 @@ The CCMetric router sits in between the collectors and the sinks and can be used
 {
     "num_cache_intervals" : 1,
     "interval_timestamp" : true,
+    "hostname_tag" : "hostname",
+    "max_forward" : 50,
     "add_tags" : [
         {
             "key" : "cluster",
@@ -55,6 +57,20 @@ The CCMetric router sits in between the collectors and the sinks and can be used
 ```
 
 There are three main options `add_tags`, `delete_tags` and `interval_timestamp`. `add_tags` and `delete_tags` are lists consisting of dicts with `key`, `value` and `if`. The `value` can be omitted in the `delete_tags` part as it only uses the `key` for removal. The `interval_timestamp` setting means that a unique timestamp is applied to all metrics traversing the router during an interval.
+
+# Processing order in the router
+
+- Add the `hostname_tag` tag (if sent by collectors or cache)
+- If `interval_timestamp == true`, change time of metrics
+- Check if metric should be dropped (`drop_metrics` and `drop_metrics_if`)
+- Add tags from `add_tags`
+- Delete tags from `del_tags`
+- Rename metric based on `rename_metrics` and store old name as `oldname` in meta information
+- Add tags from `add_tags` (if you used the new name in the `if` condition)
+- Delete tags from `del_tags` (if you used the new name in the `if` condition)
+- Send to sinks
+- Move to cache (if `num_cache_intervals > 0`)
+
 # The `interval_timestamp` option
 
 The collectors' `Read()` functions are not called simultaneously and therefore the metrics gathered in an interval can have different timestamps. If you want to avoid that and have a common timestamp (the beginning of the interval), set this option to `true` and the MetricRouter sets the time.
@@ -65,6 +81,14 @@ If the MetricRouter should buffer metrics of intervals in a MetricCache, this op
 
 A `num_cache_intervals > 0` is required to use the `interval_aggregates` option.
 
+# The `hostname_tag` option
+
+By default, the router tags metrics with the hostname for all locally created metrics. The default tag name is `hostname`, but it can be changed if your organization wants anything else
+
+# The `max_forward` option
+
+Every time the router receives a metric through any of the channels, it tries to directly read up to `max_forward` metrics from the same channel. This was done as the router thread would go to sleep and wake up with every arriving metric. The default are `50` metrics at once and `max_forward` needs to greater than `1`.
+
 # The `rename_metrics` option
 
 In the ClusterCockpit world we specified a set of standard metrics. Since some collectors determine the metric names based on files, execuables and libraries, they might change from system to system (or installation to installtion, OS to OS, ...). In order to get the common names, you can rename incoming metrics before sending them to the sink. If the metric name matches the `oldname`, it is changed to `newname`

internal/metricRouter/StatsApi.md

@@ -0,0 +1,17 @@
+# Stats API
+
+The Stats API can be used for debugging. It publishes counts at an HTTP endpoint as JSON from different componenets of the CC Metric Collector.
+
+# Configuration
+
+The Stats API has an own configuration file to specify the listen host and port. The defaults are `localhost` and `8080`.
+
+```json
+{
+  "bindhost" : "",
+  "port" : "8080",
+  "publish_collectorstate" : true
+}
+```
+
+The `bindhost` and `port` can be used to specify the listen host and port. The `publish_collectorstate` needs to be `true`, otherwise nothing is presented. This option is for future use if we need to publish more infos using different domains.

internal/metricRouter/metricApi.go

@@ -0,0 +1,232 @@
+package metricRouter
+
+import (
+	"context"
+	"encoding/json"
+	"fmt"
+	"io"
+	"net/http"
+	"os"
+	"sync"
+	"time"
+
+	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
+	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	mct "github.com/ClusterCockpit/cc-metric-collector/internal/multiChanTicker"
+	"github.com/gorilla/mux"
+)
+
+type statsApiConfig struct {
+	PublishCollectorState bool   `json:"publish_collectorstate"`
+	Host                  string `json:"bindhost"`
+	Port                  string `json:"port"`
+}
+
+// Metric cache data structure
+type statsApi struct {
+	name     string
+	input    chan lp.CCMetric
+	indone   chan bool
+	outdone  chan bool
+	config   statsApiConfig
+	wg       *sync.WaitGroup
+	statsWg  sync.WaitGroup
+	ticker   mct.MultiChanTicker
+	tickchan chan time.Time
+	server   *http.Server
+	router   *mux.Router
+	lock     sync.Mutex
+	baseurl  string
+	stats    map[string]map[string]int64
+	outStats map[string]map[string]int64
+}
+
+type StatsApi interface {
+	Start()
+	Close()
+	StatsFunc(w http.ResponseWriter, r *http.Request)
+}
+
+var statsApiServer *statsApi = nil
+
+func (a *statsApi) updateStats(point lp.CCMetric) {
+	switch point.Name() {
+	case "_stats":
+		if name, nok := point.GetMeta("source"); nok {
+			var compStats map[string]int64
+			var ok bool
+
+			if compStats, ok = a.stats[name]; !ok {
+				a.stats[name] = make(map[string]int64)
+				compStats = a.stats[name]
+			}
+			for k, v := range point.Fields() {
+				switch value := v.(type) {
+				case int:
+					compStats[k] = int64(value)
+				case uint:
+					compStats[k] = int64(value)
+				case int32:
+					compStats[k] = int64(value)
+				case uint32:
+					compStats[k] = int64(value)
+				case int64:
+					compStats[k] = int64(value)
+				case uint64:
+					compStats[k] = int64(value)
+				default:
+					cclog.ComponentDebug(a.name, "Unusable stats for", k, ". Values should be int64")
+				}
+			}
+			a.stats[name] = compStats
+		}
+	}
+}
+
+func (a *statsApi) Start() {
+	a.ticker.AddChannel(a.tickchan)
+	a.wg.Add(1)
+	a.statsWg.Add(1)
+	go func() {
+		a.stats = make(map[string]map[string]int64)
+		defer a.statsWg.Done()
+		for {
+			select {
+			case <-a.indone:
+				cclog.ComponentDebug(a.name, "INPUT DONE")
+				close(a.indone)
+				return
+			case p := <-a.input:
+				a.lock.Lock()
+				a.updateStats(p)
+				a.lock.Unlock()
+			}
+		}
+	}()
+	a.statsWg.Add(1)
+	go func() {
+		a.outStats = make(map[string]map[string]int64)
+		defer a.statsWg.Done()
+		a.lock.Lock()
+		for comp, compData := range a.stats {
+			var outData map[string]int64
+			var ok bool
+			if outData, ok = a.outStats[comp]; !ok {
+				outData = make(map[string]int64)
+			}
+			for k, v := range compData {
+				outData[k] = v
+			}
+			a.outStats[comp] = outData
+		}
+		a.lock.Unlock()
+		for {
+			select {
+			case <-a.outdone:
+				cclog.ComponentDebug(a.name, "OUTPUT DONE")
+				close(a.outdone)
+				return
+			case <-a.tickchan:
+				a.lock.Lock()
+				for comp, compData := range a.stats {
+					var outData map[string]int64
+					var ok bool
+					if outData, ok = a.outStats[comp]; !ok {
+						outData = make(map[string]int64)
+					}
+					for k, v := range compData {
+						outData[k] = v
+					}
+					a.outStats[comp] = outData
+				}
+				a.lock.Unlock()
+			}
+		}
+	}()
+	a.statsWg.Add(1)
+	go func() {
+		defer a.statsWg.Done()
+		err := a.server.ListenAndServe()
+		if err != nil && err.Error() != "http: Server closed" {
+			cclog.ComponentError(a.name, err.Error())
+		}
+		cclog.ComponentDebug(a.name, "SERVER DONE")
+	}()
+	cclog.ComponentDebug(a.name, "STARTED")
+}
+
+func (a *statsApi) StatsFunc(w http.ResponseWriter, r *http.Request) {
+	data, err := json.Marshal(a.outStats)
+	if err == nil {
+		w.Header().Set("Content-Type", "application/json")
+		io.WriteString(w, string(data))
+	}
+}
+
+// Close finishes / stops the metric cache
+func (a *statsApi) Close() {
+	cclog.ComponentDebug(a.name, "CLOSE")
+	a.indone <- true
+	a.outdone <- true
+	a.server.Shutdown(context.Background())
+	// wait for close of channel r.done
+	<-a.indone
+	<-a.outdone
+	a.statsWg.Wait()
+	a.wg.Done()
+
+	//a.wg.Wait()
+}
+
+func NewStatsApi(ticker mct.MultiChanTicker, wg *sync.WaitGroup, statsApiConfigfile string) (StatsApi, error) {
+	a := new(statsApi)
+	a.name = "StatsApi"
+	a.config.Host = "localhost"
+	a.config.Port = "8080"
+	configFile, err := os.Open(statsApiConfigfile)
+	if err != nil {
+		cclog.ComponentError(a.name, err.Error())
+		return nil, err
+	}
+	defer configFile.Close()
+	jsonParser := json.NewDecoder(configFile)
+	err = jsonParser.Decode(&a.config)
+	if err != nil {
+		cclog.ComponentError(a.name, err.Error())
+		return nil, err
+	}
+	a.input = make(chan lp.CCMetric)
+	a.ticker = ticker
+	a.tickchan = make(chan time.Time)
+	a.wg = wg
+	a.indone = make(chan bool)
+	a.outdone = make(chan bool)
+	a.router = mux.NewRouter()
+	a.baseurl = fmt.Sprintf("%s:%s", a.config.Host, a.config.Port)
+	a.server = &http.Server{Addr: a.baseurl, Handler: a.router}
+	if a.config.PublishCollectorState {
+		a.router.HandleFunc("/", a.StatsFunc)
+	}
+	statsApiServer = a
+	return a, nil
+}
+
+func ComponentStatInt(component string, key string, value int64) {
+	if statsApiServer == nil {
+		return
+	}
+	y, err := lp.New("_stats", map[string]string{}, map[string]string{"source": component}, map[string]interface{}{key: value}, time.Now())
+	if err == nil {
+		statsApiServer.input <- y
+	}
+}
+
+func ComponentStatString(component string, key string, value int64) {
+	if statsApiServer == nil {
+		return
+	}
+	y, err := lp.New("_stats", map[string]string{}, map[string]string{"source": component}, map[string]interface{}{key: value}, time.Now())
+	if err == nil {
+		statsApiServer.input <- y
+	}
+}

internal/metricRouter/metricRouter.go

@@ -25,6 +25,7 @@ type metricRouterTagConfig struct {
 
 // Metric router configuration
 type metricRouterConfig struct {
+	HostnameTagName   string                               `json:"hostname_tag"`        // Key name used when adding the hostname to a metric (default 'hostname')
 	AddTags           []metricRouterTagConfig              `json:"add_tags"`            // List of tags that are added when the condition is met
 	DelTags           []metricRouterTagConfig              `json:"delete_tags"`         // List of tags that are removed when the condition is met
 	IntervalAgg       []agg.MetricAggregatorIntervalConfig `json:"interval_aggregates"` // List of aggregation function processed at the end of an interval
@@ -33,25 +34,32 @@ type metricRouterConfig struct {
 	RenameMetrics     map[string]string                    `json:"rename_metrics"`      // Map to rename metric name from key to value
 	IntervalStamp     bool                                 `json:"interval_timestamp"`  // Update timestamp periodically by ticker each interval?
 	NumCacheIntervals int                                  `json:"num_cache_intervals"` // Number of intervals of cached metrics for evaluation
+	MaxForward        int                                  `json:"max_forward"`         // Number of maximal forwarded metrics at one select
 	dropMetrics       map[string]bool                      // Internal map for O(1) lookup
 }
 
 // Metric router data structure
 type metricRouter struct {
-	hostname    string              // Hostname used in tags
-	coll_input  chan lp.CCMetric    // Input channel from CollectorManager
-	recv_input  chan lp.CCMetric    // Input channel from ReceiveManager
-	cache_input chan lp.CCMetric    // Input channel from MetricCache
-	outputs     []chan lp.CCMetric  // List of all output channels
-	done        chan bool           // channel to finish / stop metric router
-	wg          *sync.WaitGroup     // wait group for all goroutines in cc-metric-collector
-	timestamp   time.Time           // timestamp periodically updated by ticker each interval
-	timerdone   chan bool           // channel to finish / stop timestamp updater
-	ticker      mct.MultiChanTicker // periodically ticking once each interval
-	config      metricRouterConfig  // json encoded config for metric router
-	cache       MetricCache         // pointer to MetricCache
-	cachewg     sync.WaitGroup      // wait group for MetricCache
-	maxForward  int                 // number of metrics to forward maximally in one iteration
+	hostname          string              // Hostname used in tags
+	coll_input        chan lp.CCMetric    // Input channel from CollectorManager
+	recv_input        chan lp.CCMetric    // Input channel from ReceiveManager
+	cache_input       chan lp.CCMetric    // Input channel from MetricCache
+	outputs           []chan lp.CCMetric  // List of all output channels
+	done              chan bool           // channel to finish / stop metric router
+	wg                *sync.WaitGroup     // wait group for all goroutines in cc-metric-collector
+	timestamp         time.Time           // timestamp periodically updated by ticker each interval
+	timerdone         chan bool           // channel to finish / stop timestamp updater
+	ticker            mct.MultiChanTicker // periodically ticking once each interval
+	config            metricRouterConfig  // json encoded config for metric router
+	cache             MetricCache         // pointer to MetricCache
+	cachewg           sync.WaitGroup      // wait group for MetricCache
+	maxForward        int                 // number of metrics to forward maximally in one iteration
+	statsCollForward  int64
+	statsRecvForward  int64
+	statsCacheForward int64
+	statsTotalForward int64
+	statsDropped      int64
+	statsRenamed      int64
 }
 
 // MetricRouter access functions
@@ -76,7 +84,8 @@ func (r *metricRouter) Init(ticker mct.MultiChanTicker, wg *sync.WaitGroup, rout
 	r.cache_input = make(chan lp.CCMetric)
 	r.wg = wg
 	r.ticker = ticker
-	r.maxForward = ROUTER_MAX_FORWARD
+	r.config.MaxForward = ROUTER_MAX_FORWARD
+	r.config.HostnameTagName = "hostname"
 
 	// Set hostname
 	hostname, err := os.Hostname()
@@ -100,6 +109,10 @@ func (r *metricRouter) Init(ticker mct.MultiChanTicker, wg *sync.WaitGroup, rout
 		cclog.ComponentError("MetricRouter", err.Error())
 		return err
 	}
+	r.maxForward = 1
+	if r.config.MaxForward > r.maxForward {
+		r.maxForward = r.config.MaxForward
+	}
 	if r.config.NumCacheIntervals > 0 {
 		r.cache, err = NewCache(r.cache_input, r.ticker, &r.cachewg, r.config.NumCacheIntervals)
 		if err != nil {
@@ -114,6 +127,12 @@ func (r *metricRouter) Init(ticker mct.MultiChanTicker, wg *sync.WaitGroup, rout
 	for _, mname := range r.config.DropMetrics {
 		r.config.dropMetrics[mname] = true
 	}
+	r.statsCollForward = 0
+	r.statsRecvForward = 0
+	r.statsCacheForward = 0
+	r.statsTotalForward = 0
+	r.statsDropped = 0
+	r.statsRenamed = 0
 	return nil
 }
 
@@ -133,6 +152,7 @@ func (r *metricRouter) StartTimer() {
 				cclog.ComponentDebug("MetricRouter", "TIMER DONE")
 				return
 			case t := <-m:
+				cclog.ComponentDebug("MetricRouter", "INTERVAL_TICK", t.Unix())
 				r.timestamp = t
 			}
 		}
@@ -244,8 +264,12 @@ func (r *metricRouter) Start() {
 		cclog.ComponentDebug("MetricRouter", "FORWARD", point)
 		r.DoAddTags(point)
 		r.DoDelTags(point)
-		if new, ok := r.config.RenameMetrics[point.Name()]; ok {
+		name := point.Name()
+		if new, ok := r.config.RenameMetrics[name]; ok {
+			r.statsRenamed++
+			ComponentStatInt("MetricRouter", "renamed", r.statsRenamed)
 			point.SetName(new)
+			point.AddMeta("oldname", name)
 		}
 		r.DoAddTags(point)
 		r.DoDelTags(point)
@@ -258,12 +282,19 @@ func (r *metricRouter) Start() {
 	// Foward message received from collector channel
 	coll_forward := func(p lp.CCMetric) {
 		// receive from metric collector
-		p.AddTag("hostname", r.hostname)
+		p.AddTag(r.config.HostnameTagName, r.hostname)
 		if r.config.IntervalStamp {
 			p.SetTime(r.timestamp)
 		}
 		if !r.dropMetric(p) {
+			r.statsCollForward++
+			r.statsTotalForward++
+			ComponentStatInt("MetricRouter", "collector_forward", r.statsCollForward)
+			ComponentStatInt("MetricRouter", "total_forward", r.statsTotalForward)
 			forward(p)
+		} else {
+			r.statsDropped++
+			ComponentStatInt("MetricRouter", "dropped", r.statsDropped)
 		}
 		// even if the metric is dropped, it is stored in the cache for
 		// aggregations
@@ -279,7 +310,14 @@ func (r *metricRouter) Start() {
 			p.SetTime(r.timestamp)
 		}
 		if !r.dropMetric(p) {
+			r.statsRecvForward++
+			r.statsTotalForward++
+			ComponentStatInt("MetricRouter", "receiver_forward", r.statsRecvForward)
+			ComponentStatInt("MetricRouter", "total_forward", r.statsTotalForward)
 			forward(p)
+		} else {
+			r.statsDropped++
+			ComponentStatInt("MetricRouter", "dropped", r.statsDropped)
 		}
 	}
 
@@ -287,8 +325,15 @@ func (r *metricRouter) Start() {
 	cache_forward := func(p lp.CCMetric) {
 		// receive from metric collector
 		if !r.dropMetric(p) {
-			p.AddTag("hostname", r.hostname)
+			p.AddTag(r.config.HostnameTagName, r.hostname)
+			r.statsCacheForward++
+			r.statsTotalForward++
+			ComponentStatInt("MetricRouter", "cache_forward", r.statsCacheForward)
+			ComponentStatInt("MetricRouter", "total_forward", r.statsTotalForward)
 			forward(p)
+		} else {
+			r.statsDropped++
+			ComponentStatInt("MetricRouter", "dropped", r.statsDropped)
 		}
 	}
 
@@ -309,19 +354,19 @@ func (r *metricRouter) Start() {
 
 			case p := <-r.coll_input:
 				coll_forward(p)
-				for i := 0; len(r.coll_input) > 0 && i < r.maxForward; i++ {
+				for i := 0; len(r.coll_input) > 0 && i < (r.maxForward-1); i++ {
 					coll_forward(<-r.coll_input)
 				}
 
 			case p := <-r.recv_input:
 				recv_forward(p)
-				for i := 0; len(r.recv_input) > 0 && i < r.maxForward; i++ {
+				for i := 0; len(r.recv_input) > 0 && i < (r.maxForward-1); i++ {
 					recv_forward(<-r.recv_input)
 				}
 
 			case p := <-r.cache_input:
 				cache_forward(p)
-				for i := 0; len(r.cache_input) > 0 && i < r.maxForward; i++ {
+				for i := 0; len(r.cache_input) > 0 && i < (r.maxForward-1); i++ {
 					cache_forward(<-r.cache_input)
 				}
 			}

receivers/README.md

@@ -21,6 +21,7 @@ This allows to specify
 
 - [`nats`](./natsReceiver.md): Receive metrics from the NATS network
 - [`prometheus`](./prometheusReceiver.md): Scrape data from a Prometheus client
+- [`http`](./httpReceiver.md): Listen for HTTP Post requests transporting metrics in InfluxDB line protocol
 
 # Contributing own receivers
 A receiver contains a few functions and is derived from the type `Receiver` (in `metricReceiver.go`):

receivers/httpReceiver.go

@@ -0,0 +1,118 @@
+package receivers
+
+import (
+	"context"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"io/ioutil"
+	"net/http"
+	"strings"
+	"sync"
+
+	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
+	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	"github.com/gorilla/mux"
+	influx "github.com/influxdata/line-protocol"
+)
+
+const HTTP_RECEIVER_PORT = "8080"
+
+type HttpReceiverConfig struct {
+	Type string `json:"type"`
+	Addr string `json:"address"`
+	Port string `json:"port"`
+	Path string `json:"path"`
+}
+
+type HttpReceiver struct {
+	receiver
+	handler *influx.MetricHandler
+	parser  *influx.Parser
+	meta    map[string]string
+	config  HttpReceiverConfig
+	router  *mux.Router
+	server  *http.Server
+	wg      sync.WaitGroup
+}
+
+func (r *HttpReceiver) Init(name string, config json.RawMessage) error {
+	r.name = fmt.Sprintf("HttpReceiver(%s)", name)
+	r.config.Port = HTTP_RECEIVER_PORT
+	if len(config) > 0 {
+		err := json.Unmarshal(config, &r.config)
+		if err != nil {
+			cclog.ComponentError(r.name, "Error reading config:", err.Error())
+			return err
+		}
+	}
+	if len(r.config.Port) == 0 {
+		return errors.New("not all configuration variables set required by HttpReceiver")
+	}
+	r.meta = map[string]string{"source": r.name}
+	p := r.config.Path
+	if !strings.HasPrefix(p, "/") {
+		p = "/" + p
+	}
+	uri := fmt.Sprintf("%s:%s%s", r.config.Addr, r.config.Port, p)
+	cclog.ComponentDebug(r.name, "INIT", uri)
+	r.handler = influx.NewMetricHandler()
+	r.parser = influx.NewParser(r.handler)
+	r.parser.SetTimeFunc(DefaultTime)
+
+	r.router = mux.NewRouter()
+	r.router.Path(p).HandlerFunc(r.ServerHttp)
+	r.server = &http.Server{Addr: uri, Handler: r.router}
+	return nil
+}
+
+func (r *HttpReceiver) Start() {
+	cclog.ComponentDebug(r.name, "START")
+	r.wg.Add(1)
+	go func() {
+		err := r.server.ListenAndServe()
+		if err != nil && err.Error() != "http: Server closed" {
+			cclog.ComponentError(r.name, err.Error())
+		}
+		r.wg.Done()
+	}()
+}
+
+func (r *HttpReceiver) ServerHttp(w http.ResponseWriter, req *http.Request) {
+	if req.Method != http.MethodPost {
+		http.Error(w, "Method Not Allowed", http.StatusMethodNotAllowed)
+		return
+	}
+
+	body, err := ioutil.ReadAll(req.Body)
+	if err != nil {
+		http.Error(w, err.Error(), http.StatusInternalServerError)
+		return
+	}
+	metrics, err := r.parser.Parse(body)
+	if err != nil {
+		http.Error(w, err.Error(), http.StatusBadRequest)
+	}
+
+	for _, m := range metrics {
+		y := lp.FromInfluxMetric(m)
+		for k, v := range r.meta {
+			y.AddMeta(k, v)
+		}
+		if r.sink != nil {
+			r.sink <- y
+		}
+	}
+
+	w.WriteHeader(http.StatusOK)
+}
+
+func (r *HttpReceiver) Close() {
+	r.server.Shutdown(context.Background())
+}
+
+func NewHttpReceiver(name string, config json.RawMessage) (Receiver, error) {
+	r := new(HttpReceiver)
+	err := r.Init(name, config)
+	return r, err
+}

receivers/httpReceiver.md

@@ -0,0 +1,23 @@
+## `http` receiver
+
+The `http` receiver can be used receive metrics through HTTP POST requests.
+
+### Configuration structure
+
+```json
+{
+  "<name>": {
+    "type": "http",
+    "address" : "",
+    "port" : "8080",
+    "path" : "/write"
+  }
+}
+```
+
+- `type`: makes the receiver a `http` receiver
+- `address`: Listen address
+- `port`: Listen port
+- `path`: URL path for the write endpoint
+
+The HTTP endpoint listens to `http://<address>:<port>/<path>`

scripts/cc-metric-collector.deb.control

@@ -0,0 +1,12 @@
+Package: cc-metric-collector
+Version: {VERSION}
+Installed-Size: {INSTALLED_SIZE}
+Architecture: {ARCH}
+Maintainer: thomas.gruber@fau.de
+Depends: libc6 (>= 2.2.1)
+Build-Depends: debhelper-compat (= 13), git, golang-go
+Description: Metric collection daemon from the ClusterCockpit suite
+Homepage: https://github.com/ClusterCockpit/cc-metric-collector
+Source: cc-metric-collector
+Rules-Requires-Root: no
+

scripts/cc-metric-collector.spec

@@ -8,8 +8,8 @@ Source0:        %{name}-%{version}.tar.gz
 
 BuildRequires:  go-toolset
 BuildRequires:  systemd-rpm-macros
-# for internal LIKWID installation
-BuildRequires:  wget perl-Data-Dumper
+# for header downloads
+BuildRequires:  wget
 
 Provides:       %{name} = %{version}
 
@@ -53,7 +53,7 @@ install -Dpm 0644 scripts/%{name}.sysusers %{buildroot}%{_sysusersdir}/%{name}.c
 %files
 # Binary
 %attr(-,clustercockpit,clustercockpit) %{_sbindir}/%{name}
-# Configuration
+# Config
 %dir %{_sysconfdir}/%{name}
 %attr(0600,clustercockpit,clustercockpit) %config(noreplace) %{_sysconfdir}/%{name}/%{name}.json
 %attr(0600,clustercockpit,clustercockpit) %config(noreplace) %{_sysconfdir}/%{name}/collectors.json
@@ -61,9 +61,9 @@ install -Dpm 0644 scripts/%{name}.sysusers %{buildroot}%{_sysusersdir}/%{name}.c
 %attr(0600,clustercockpit,clustercockpit) %config(noreplace) %{_sysconfdir}/%{name}/receivers.json
 %attr(0600,clustercockpit,clustercockpit) %config(noreplace) %{_sysconfdir}/%{name}/router.json
 # Systemd
-%{_sysusersdir}/%{name}.conf
 %{_unitdir}/%{name}.service
 %{_sysconfdir}/default/%{name}
+%{_sysusersdir}/%{name}.conf
 
 %changelog
 * Thu Mar 03 2022 Thomas Gruber - 0.3

sinks/gangliaCommon.go

@@ -148,10 +148,14 @@ type GangliaMetricConfig struct {
 	Unit  string
 	Group string
 	Value string
+	Name  string
 }
 
 func GetCommonGangliaConfig(point lp.CCMetric) GangliaMetricConfig {
 	mname := GangliaMetricRename(point.Name())
+	if oldname, ok := point.GetMeta("oldname"); ok {
+		mname = GangliaMetricRename(oldname)
+	}
 	for _, group := range CommonGangliaMetrics {
 		for _, metric := range group.Metrics {
 			if metric.Name == mname {
@@ -187,6 +191,7 @@ func GetCommonGangliaConfig(point lp.CCMetric) GangliaMetricConfig {
 					Tmax:  metric.Tmax,
 					Unit:  metric.Unit,
 					Value: valueStr,
+					Name:  GangliaMetricRename(mname),
 				}
 			}
 		}
@@ -198,10 +203,15 @@ func GetCommonGangliaConfig(point lp.CCMetric) GangliaMetricConfig {
 		Tmax:  0,
 		Unit:  "",
 		Value: "",
+		Name:  "",
 	}
 }
 
 func GetGangliaConfig(point lp.CCMetric) GangliaMetricConfig {
+	mname := GangliaMetricRename(point.Name())
+	if oldname, ok := point.GetMeta("oldname"); ok {
+		mname = GangliaMetricRename(oldname)
+	}
 	group := ""
 	if g, ok := point.GetMeta("group"); ok {
 		group = g
@@ -254,5 +264,6 @@ func GetGangliaConfig(point lp.CCMetric) GangliaMetricConfig {
 		Tmax:  DEFAULT_GANGLIA_METRIC_TMAX,
 		Unit:  unit,
 		Value: valueStr,
+		Name:  GangliaMetricRename(mname),
 	}
 }

sinks/gangliaSink.go

@@ -11,6 +11,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 const GMETRIC_EXEC = `gmetric`
@@ -29,9 +30,10 @@ type GangliaSinkConfig struct {
 
 type GangliaSink struct {
 	sink
-	gmetric_path   string
-	gmetric_config string
-	config         GangliaSinkConfig
+	gmetric_path     string
+	gmetric_config   string
+	config           GangliaSinkConfig
+	statsSentMetrics int64
 }
 
 func (s *GangliaSink) Write(point lp.CCMetric) error {
@@ -39,16 +41,13 @@ func (s *GangliaSink) Write(point lp.CCMetric) error {
 	//var tagsstr []string
 	var argstr []string
 
-	// Get metric name
-	metricname := GangliaMetricRename(point.Name())
-
 	// Get metric config (type, value, ... in suitable format)
 	conf := GetCommonGangliaConfig(point)
 	if len(conf.Type) == 0 {
 		conf = GetGangliaConfig(point)
 	}
 	if len(conf.Type) == 0 {
-		return fmt.Errorf("metric %s has no 'value' field", metricname)
+		return fmt.Errorf("metric %q (Ganglia name %q) has no 'value' field", point.Name(), conf.Name)
 	}
 
 	if s.config.AddGangliaGroup {
@@ -70,7 +69,7 @@ func (s *GangliaSink) Write(point lp.CCMetric) error {
 	if s.config.AddTypeToName {
 		argstr = append(argstr, fmt.Sprintf("--name=%s", GangliaMetricName(point)))
 	} else {
-		argstr = append(argstr, fmt.Sprintf("--name=%s", metricname))
+		argstr = append(argstr, fmt.Sprintf("--name=%s", conf.Name))
 	}
 	argstr = append(argstr, fmt.Sprintf("--slope=%s", conf.Slope))
 	argstr = append(argstr, fmt.Sprintf("--value=%s", conf.Value))
@@ -81,6 +80,8 @@ func (s *GangliaSink) Write(point lp.CCMetric) error {
 	command := exec.Command(s.gmetric_path, argstr...)
 	command.Wait()
 	_, err = command.Output()
+	s.statsSentMetrics++
+	stats.ComponentStatInt(s.name, "sent_metrics", s.statsSentMetrics)
 	return err
 }
 
@@ -123,5 +124,6 @@ func NewGangliaSink(name string, config json.RawMessage) (Sink, error) {
 	if len(s.config.GmetricConfig) > 0 {
 		s.gmetric_config = s.config.GmetricConfig
 	}
+	s.statsSentMetrics = 0
 	return s, nil
 }

sinks/httpSink.go

@@ -11,6 +11,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 	influx "github.com/influxdata/line-protocol"
 )
 
@@ -36,32 +37,37 @@ type HttpSink struct {
 	idleConnTimeout time.Duration
 	timeout         time.Duration
 	flushDelay      time.Duration
+	statsProcessed  int64
+	statsFlushes    int64
 }
 
 func (s *HttpSink) Write(m lp.CCMetric) error {
 	if s.buffer.Len() == 0 && s.flushDelay != 0 {
 		// This is the first write since the last flush, start the flushTimer!
 		if s.flushTimer != nil && s.flushTimer.Stop() {
-			cclog.ComponentDebug("HttpSink", "unexpected: the flushTimer was already running?")
+			cclog.ComponentDebug(s.name, "unexpected: the flushTimer was already running?")
 		}
 
 		// Run a batched flush for all lines that have arrived in the last second
 		s.flushTimer = time.AfterFunc(s.flushDelay, func() {
 			if err := s.Flush(); err != nil {
-				cclog.ComponentError("HttpSink", "flush failed:", err.Error())
+				cclog.ComponentError(s.name, "flush failed:", err.Error())
 			}
 		})
 	}
 
-	p := m.ToPoint(s.config.MetaAsTags)
+	p := m.ToPoint(s.meta_as_tags)
 
 	s.lock.Lock()
 	_, err := s.encoder.Encode(p)
 	s.lock.Unlock() // defer does not work here as Flush() takes the lock as well
 
 	if err != nil {
+		cclog.ComponentError(s.name, "encoding failed:", err.Error())
 		return err
 	}
+	s.statsProcessed++
+	stats.ComponentStatInt(s.name, "processed_metrics", s.statsProcessed)
 
 	// Flush synchronously if "flush_delay" is zero
 	if s.flushDelay == 0 {
@@ -84,6 +90,7 @@ func (s *HttpSink) Flush() error {
 	// Create new request to send buffer
 	req, err := http.NewRequest(http.MethodPost, s.config.URL, s.buffer)
 	if err != nil {
+		cclog.ComponentError(s.name, "failed to create request:", err.Error())
 		return err
 	}
 
@@ -100,13 +107,18 @@ func (s *HttpSink) Flush() error {
 
 	// Handle transport/tcp errors
 	if err != nil {
+		cclog.ComponentError(s.name, "transport/tcp error:", err.Error())
 		return err
 	}
 
 	// Handle application errors
 	if res.StatusCode != http.StatusOK {
-		return errors.New(res.Status)
+		err = errors.New(res.Status)
+		cclog.ComponentError(s.name, "application error:", err.Error())
+		return err
 	}
+	s.statsFlushes++
+	stats.ComponentStatInt(s.name, "flushes", s.statsFlushes)
 
 	return nil
 }
@@ -114,7 +126,7 @@ func (s *HttpSink) Flush() error {
 func (s *HttpSink) Close() {
 	s.flushTimer.Stop()
 	if err := s.Flush(); err != nil {
-		cclog.ComponentError("HttpSink", "flush failed:", err.Error())
+		cclog.ComponentError(s.name, "flush failed:", err.Error())
 	}
 	s.client.CloseIdleConnections()
 }
@@ -159,6 +171,11 @@ func NewHttpSink(name string, config json.RawMessage) (Sink, error) {
 			s.flushDelay = t
 		}
 	}
+	// Create lookup map to use meta infos as tags in the output metric
+	s.meta_as_tags = make(map[string]bool)
+	for _, k := range s.config.MetaAsTags {
+		s.meta_as_tags[k] = true
+	}
 	tr := &http.Transport{
 		MaxIdleConns:    s.maxIdleConns,
 		IdleConnTimeout: s.idleConnTimeout,
@@ -167,5 +184,7 @@ func NewHttpSink(name string, config json.RawMessage) (Sink, error) {
 	s.buffer = &bytes.Buffer{}
 	s.encoder = influx.NewEncoder(s.buffer)
 	s.encoder.SetPrecision(time.Second)
+	s.statsFlushes = 0
+	s.statsProcessed = 0
 	return s, nil
 }

sinks/influxAsyncSink.go

@@ -6,9 +6,11 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
+	"time"
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 	influxdb2 "github.com/influxdata/influxdb-client-go/v2"
 	influxdb2Api "github.com/influxdata/influxdb-client-go/v2/api"
 )
@@ -26,15 +28,24 @@ type InfluxAsyncSinkConfig struct {
 	// Maximum number of points sent to server in single request. Default 5000
 	BatchSize uint `json:"batch_size,omitempty"`
 	// Interval, in ms, in which is buffer flushed if it has not been already written (by reaching batch size) . Default 1000ms
-	FlushInterval uint `json:"flush_interval,omitempty"`
+	FlushInterval         uint   `json:"flush_interval,omitempty"`
+	InfluxRetryInterval   string `json:"retry_interval,omitempty"`
+	InfluxExponentialBase uint   `json:"retry_exponential_base,omitempty"`
+	InfluxMaxRetries      uint   `json:"max_retries,omitempty"`
+	InfluxMaxRetryTime    string `json:"max_retry_time,omitempty"`
 }
 
 type InfluxAsyncSink struct {
 	sink
-	client   influxdb2.Client
-	writeApi influxdb2Api.WriteAPI
-	errors   <-chan error
-	config   InfluxAsyncSinkConfig
+	client              influxdb2.Client
+	writeApi            influxdb2Api.WriteAPI
+	errors              <-chan error
+	config              InfluxAsyncSinkConfig
+	influxRetryInterval uint
+	influxMaxRetryTime  uint
+	sentMetrics         int64
+	statsFlushes        int64
+	statsErrors         int64
 }
 
 func (s *InfluxAsyncSink) connect() error {
@@ -53,16 +64,35 @@ func (s *InfluxAsyncSink) connect() error {
 	cclog.ComponentDebug(s.name, "Using URI", uri, "Org", s.config.Organization, "Bucket", s.config.Database)
 	clientOptions := influxdb2.DefaultOptions()
 	if s.config.BatchSize != 0 {
+		cclog.ComponentDebug(s.name, "Batch size", s.config.BatchSize)
 		clientOptions.SetBatchSize(s.config.BatchSize)
 	}
 	if s.config.FlushInterval != 0 {
+		cclog.ComponentDebug(s.name, "Flush interval", s.config.FlushInterval)
 		clientOptions.SetFlushInterval(s.config.FlushInterval)
 	}
+	if s.influxRetryInterval != 0 {
+		cclog.ComponentDebug(s.name, "MaxRetryInterval", s.influxRetryInterval)
+		clientOptions.SetMaxRetryInterval(s.influxRetryInterval)
+	}
+	if s.influxMaxRetryTime != 0 {
+		cclog.ComponentDebug(s.name, "MaxRetryTime", s.influxMaxRetryTime)
+		clientOptions.SetMaxRetryTime(s.influxMaxRetryTime)
+	}
+	if s.config.InfluxExponentialBase != 0 {
+		cclog.ComponentDebug(s.name, "Exponential Base", s.config.InfluxExponentialBase)
+		clientOptions.SetExponentialBase(s.config.InfluxExponentialBase)
+	}
+	if s.config.InfluxMaxRetries != 0 {
+		cclog.ComponentDebug(s.name, "Max Retries", s.config.InfluxMaxRetries)
+		clientOptions.SetMaxRetries(s.config.InfluxMaxRetries)
+	}
 	clientOptions.SetTLSConfig(
 		&tls.Config{
 			InsecureSkipVerify: true,
 		},
-	)
+	).SetPrecision(time.Second)
+
 	s.client = influxdb2.NewClientWithOptions(uri, auth, clientOptions)
 	s.writeApi = s.client.WriteAPI(s.config.Organization, s.config.Database)
 	ok, err := s.client.Ping(context.Background())
@@ -77,13 +107,17 @@ func (s *InfluxAsyncSink) connect() error {
 
 func (s *InfluxAsyncSink) Write(m lp.CCMetric) error {
 	s.writeApi.WritePoint(
-		m.ToPoint(s.config.MetaAsTags),
+		m.ToPoint(s.meta_as_tags),
 	)
+	s.sentMetrics++
+	stats.ComponentStatInt(s.name, "send_metrics", s.sentMetrics)
 	return nil
 }
 
 func (s *InfluxAsyncSink) Flush() error {
 	s.writeApi.Flush()
+	s.statsFlushes++
+	stats.ComponentStatInt(s.name, "flushes", s.statsFlushes)
 	return nil
 }
 
@@ -98,7 +132,35 @@ func NewInfluxAsyncSink(name string, config json.RawMessage) (Sink, error) {
 	s.name = fmt.Sprintf("InfluxSink(%s)", name)
 
 	// Set default for maximum number of points sent to server in single request.
-	s.config.BatchSize = 100
+	s.config.BatchSize = 0
+	s.influxRetryInterval = 0
+	//s.config.InfluxRetryInterval = "1s"
+	s.influxMaxRetryTime = 0
+	//s.config.InfluxMaxRetryTime = "168h"
+	s.config.InfluxMaxRetries = 0
+	s.config.InfluxExponentialBase = 0
+	s.config.FlushInterval = 0
+
+	// Default retry intervals (in seconds)
+	// 1 2
+	// 2 4
+	// 4 8
+	// 8 16
+	// 16 32
+	// 32 64
+	// 64 128
+	// 128 256
+	// 256 512
+	// 512 1024
+	// 1024 2048
+	// 2048 4096
+	// 4096 8192
+	// 8192 16384
+	// 16384 32768
+	// 32768 65536
+	// 65536 131072
+	// 131072 262144
+	// 262144 524288
 
 	if len(config) > 0 {
 		err := json.Unmarshal(config, &s.config)
@@ -113,6 +175,21 @@ func NewInfluxAsyncSink(name string, config json.RawMessage) (Sink, error) {
 		len(s.config.Password) == 0 {
 		return nil, errors.New("not all configuration variables set required by InfluxAsyncSink")
 	}
+	// Create lookup map to use meta infos as tags in the output metric
+	s.meta_as_tags = make(map[string]bool)
+	for _, k := range s.config.MetaAsTags {
+		s.meta_as_tags[k] = true
+	}
+
+	toUint := func(duration string, def uint) uint {
+		t, err := time.ParseDuration(duration)
+		if err == nil {
+			return uint(t.Milliseconds())
+		}
+		return def
+	}
+	s.influxRetryInterval = toUint(s.config.InfluxRetryInterval, s.influxRetryInterval)
+	s.influxMaxRetryTime = toUint(s.config.InfluxMaxRetryTime, s.influxMaxRetryTime)
 
 	// Connect to InfluxDB server
 	if err := s.connect(); err != nil {
@@ -120,12 +197,17 @@ func NewInfluxAsyncSink(name string, config json.RawMessage) (Sink, error) {
 	}
 
 	// Start background: Read from error channel
+	s.statsErrors = 0
 	s.errors = s.writeApi.Errors()
 	go func() {
 		for err := range s.errors {
+			s.statsErrors++
+			stats.ComponentStatInt(s.name, "errors", s.statsErrors)
 			cclog.ComponentError(s.name, err.Error())
 		}
 	}()
 
+	s.sentMetrics = 0
+	s.statsFlushes = 0
 	return s, nil
 }

sinks/influxAsyncSink.md

@@ -18,6 +18,10 @@ The `influxasync` sink uses the official [InfluxDB golang client](https://pkg.go
     "organization": "myorg",
     "ssl": true,
     "batch_size": 200,
+    "retry_interval" : "1s",
+    "retry_exponential_base" : 2,
+    "max_retries": 20,
+    "max_retry_time" : "168h"
   }
 }
 ```
@@ -31,4 +35,10 @@ The `influxasync` sink uses the official [InfluxDB golang client](https://pkg.go
 - `password`: Password for basic authentification
 - `organization`: Organization in the InfluxDB
 - `ssl`: Use SSL connection
-- `batch_size`: batch up metrics internally, default 100
+- `batch_size`: batch up metrics internally, default 100
+- `retry_interval`: Base retry interval for failed write requests, default 1s
+- `retry_exponential_base`: The retry interval is exponentially increased with this base, default 2
+- `max_retries`: Maximal number of retry attempts
+- `max_retry_time`: Maximal time to retry failed writes, default 168h (one week)
+
+For information about the calculation of the retry interval settings, see [offical influxdb-client-go documentation](https://github.com/influxdata/influxdb-client-go#handling-of-failed-async-writes)

sinks/influxSink.go

@@ -6,11 +6,15 @@ import (
 	"encoding/json"
 	"errors"
 	"fmt"
+	"sync"
+	"time"
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 	influxdb2 "github.com/influxdata/influxdb-client-go/v2"
 	influxdb2Api "github.com/influxdata/influxdb-client-go/v2/api"
+	"github.com/influxdata/influxdb-client-go/v2/api/write"
 )
 
 type InfluxSinkConfig struct {
@@ -22,14 +26,30 @@ type InfluxSinkConfig struct {
 	Password     string `json:"password,omitempty"`
 	Organization string `json:"organization,omitempty"`
 	SSL          bool   `json:"ssl,omitempty"`
+	FlushDelay   string `json:"flush_delay,omitempty"`
+	BatchSize    int    `json:"batch_size,omitempty"`
 	RetentionPol string `json:"retention_policy,omitempty"`
+	// InfluxRetryInterval   string `json:"retry_interval"`
+	// InfluxExponentialBase uint   `json:"retry_exponential_base"`
+	// InfluxMaxRetries      uint   `json:"max_retries"`
+	// InfluxMaxRetryTime    string `json:"max_retry_time"`
+	//InfluxMaxRetryDelay  string `json:"max_retry_delay"` // It is mentioned in the docs but there is no way to set it
 }
 
 type InfluxSink struct {
 	sink
-	client   influxdb2.Client
-	writeApi influxdb2Api.WriteAPIBlocking
-	config   InfluxSinkConfig
+	client                influxdb2.Client
+	writeApi              influxdb2Api.WriteAPIBlocking
+	config                InfluxSinkConfig
+	influxRetryInterval   uint
+	influxMaxRetryTime    uint
+	batch                 []*write.Point
+	flushTimer            *time.Timer
+	flushDelay            time.Duration
+	lock                  sync.Mutex // Flush() runs in another goroutine, so this lock has to protect the buffer
+	statsSentMetrics      int64
+	statsProcessedMetrics int64
+	//influxMaxRetryDelay uint
 }
 
 func (s *InfluxSink) connect() error {
@@ -47,11 +67,32 @@ func (s *InfluxSink) connect() error {
 	}
 	cclog.ComponentDebug(s.name, "Using URI", uri, "Org", s.config.Organization, "Bucket", s.config.Database)
 	clientOptions := influxdb2.DefaultOptions()
+
+	// if s.influxRetryInterval != 0 {
+	// 	cclog.ComponentDebug(s.name, "MaxRetryInterval", s.influxRetryInterval)
+	// 	clientOptions.SetMaxRetryInterval(s.influxRetryInterval)
+	// }
+	// if s.influxMaxRetryTime != 0 {
+	// 	cclog.ComponentDebug(s.name, "MaxRetryTime", s.influxMaxRetryTime)
+	// 	clientOptions.SetMaxRetryTime(s.influxMaxRetryTime)
+	// }
+	// if s.config.InfluxExponentialBase != 0 {
+	// 	cclog.ComponentDebug(s.name, "Exponential Base", s.config.InfluxExponentialBase)
+	// 	clientOptions.SetExponentialBase(s.config.InfluxExponentialBase)
+	// }
+	// if s.config.InfluxMaxRetries != 0 {
+	// 	cclog.ComponentDebug(s.name, "Max Retries", s.config.InfluxMaxRetries)
+	// 	clientOptions.SetMaxRetries(s.config.InfluxMaxRetries)
+	// }
+
 	clientOptions.SetTLSConfig(
 		&tls.Config{
 			InsecureSkipVerify: true,
 		},
 	)
+
+	clientOptions.SetPrecision(time.Second)
+
 	s.client = influxdb2.NewClientWithOptions(uri, auth, clientOptions)
 	s.writeApi = s.client.WriteAPIBlocking(s.config.Organization, s.config.Database)
 	ok, err := s.client.Ping(context.Background())
@@ -65,32 +106,81 @@ func (s *InfluxSink) connect() error {
 }
 
 func (s *InfluxSink) Write(m lp.CCMetric) error {
-	err :=
-		s.writeApi.WritePoint(
-			context.Background(),
-			m.ToPoint(s.config.MetaAsTags),
-		)
-	return err
+	// err :=
+	// 	s.writeApi.WritePoint(
+	// 		context.Background(),
+	// 		m.ToPoint(s.meta_as_tags),
+	// 	)
+	if len(s.batch) == 0 && s.flushDelay != 0 {
+		// This is the first write since the last flush, start the flushTimer!
+		if s.flushTimer != nil && s.flushTimer.Stop() {
+			cclog.ComponentDebug(s.name, "unexpected: the flushTimer was already running?")
+		}
+
+		// Run a batched flush for all lines that have arrived in the last second
+		s.flushTimer = time.AfterFunc(s.flushDelay, func() {
+			if err := s.Flush(); err != nil {
+				cclog.ComponentError(s.name, "flush failed:", err.Error())
+			}
+		})
+	}
+	p := m.ToPoint(s.meta_as_tags)
+	s.lock.Lock()
+	s.statsProcessedMetrics++
+	s.batch = append(s.batch, p)
+	s.lock.Unlock()
+	stats.ComponentStatInt(s.name, "processed_metrics", s.statsProcessedMetrics)
+
+	// Flush synchronously if "flush_delay" is zero
+	if s.flushDelay == 0 {
+		return s.Flush()
+	}
+
+	return nil
 }
 
 func (s *InfluxSink) Flush() error {
+	s.lock.Lock()
+	defer s.lock.Unlock()
+	if len(s.batch) == 0 {
+		return nil
+	}
+	err := s.writeApi.WritePoint(context.Background(), s.batch...)
+	if err != nil {
+		cclog.ComponentError(s.name, "flush failed:", err.Error())
+		return err
+	}
+	s.statsSentMetrics += int64(len(s.batch))
+	stats.ComponentStatInt(s.name, "sent_metrics", s.statsSentMetrics)
+	s.batch = s.batch[:0]
 	return nil
 }
 
 func (s *InfluxSink) Close() {
 	cclog.ComponentDebug(s.name, "Closing InfluxDB connection")
+	s.flushTimer.Stop()
+	s.Flush()
 	s.client.Close()
 }
 
 func NewInfluxSink(name string, config json.RawMessage) (Sink, error) {
 	s := new(InfluxSink)
 	s.name = fmt.Sprintf("InfluxSink(%s)", name)
+	s.config.BatchSize = 100
+	s.config.FlushDelay = "1s"
 	if len(config) > 0 {
 		err := json.Unmarshal(config, &s.config)
 		if err != nil {
 			return nil, err
 		}
 	}
+	s.influxRetryInterval = 0
+	s.influxMaxRetryTime = 0
+	// s.config.InfluxRetryInterval = ""
+	// s.config.InfluxMaxRetryTime = ""
+	// s.config.InfluxMaxRetries = 0
+	// s.config.InfluxExponentialBase = 0
+
 	if len(s.config.Host) == 0 ||
 		len(s.config.Port) == 0 ||
 		len(s.config.Database) == 0 ||
@@ -98,10 +188,37 @@ func NewInfluxSink(name string, config json.RawMessage) (Sink, error) {
 		len(s.config.Password) == 0 {
 		return nil, errors.New("not all configuration variables set required by InfluxSink")
 	}
+	// Create lookup map to use meta infos as tags in the output metric
+	s.meta_as_tags = make(map[string]bool)
+	for _, k := range s.config.MetaAsTags {
+		s.meta_as_tags[k] = true
+	}
+
+	// toUint := func(duration string, def uint) uint {
+	// 	if len(duration) > 0 {
+	// 		t, err := time.ParseDuration(duration)
+	// 		if err == nil {
+	// 			return uint(t.Milliseconds())
+	// 		}
+	// 	}
+	// 	return def
+	// }
+	// s.influxRetryInterval = toUint(s.config.InfluxRetryInterval, s.influxRetryInterval)
+	// s.influxMaxRetryTime = toUint(s.config.InfluxMaxRetryTime, s.influxMaxRetryTime)
+
+	if len(s.config.FlushDelay) > 0 {
+		t, err := time.ParseDuration(s.config.FlushDelay)
+		if err == nil {
+			s.flushDelay = t
+		}
+	}
+	s.batch = make([]*write.Point, 0, s.config.BatchSize)
 
 	// Connect to InfluxDB server
 	if err := s.connect(); err != nil {
 		return nil, fmt.Errorf("unable to connect: %v", err)
 	}
+	s.statsSentMetrics = 0
+	s.statsProcessedMetrics = 0
 	return s, nil
 }

sinks/influxSink.md

@@ -17,6 +17,8 @@ The `influxdb` sink uses the official [InfluxDB golang client](https://pkg.go.de
     "password" : "examplepw",
     "organization": "myorg",
     "ssl": true,
+    "flush_delay" : "1s",
+    "batch_size" : 100
   }
 }
 ```
@@ -29,4 +31,7 @@ The `influxdb` sink uses the official [InfluxDB golang client](https://pkg.go.de
 - `user`: Username for basic authentification
 - `password`: Password for basic authentification
 - `organization`: Organization in the InfluxDB
-- `ssl`: Use SSL connection
+- `ssl`: Use SSL connection
+- `flush_delay`: Group metrics coming in to a single batch
+- `batch_size`: Maximal batch size
+

sinks/libgangliaSink.go

@@ -73,6 +73,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 	"github.com/NVIDIA/go-nvml/pkg/dl"
 )
 
@@ -102,11 +103,12 @@ type LibgangliaSinkConfig struct {
 
 type LibgangliaSink struct {
 	sink
-	config         LibgangliaSinkConfig
-	global_context C.Ganglia_pool
-	gmond_config   C.Ganglia_gmond_config
-	send_channels  C.Ganglia_udp_send_channels
-	cstrCache      map[string]*C.char
+	config           LibgangliaSinkConfig
+	global_context   C.Ganglia_pool
+	gmond_config     C.Ganglia_gmond_config
+	send_channels    C.Ganglia_udp_send_channels
+	cstrCache        map[string]*C.char
+	statsSentMetrics int64
 }
 
 func (s *LibgangliaSink) Write(point lp.CCMetric) error {
@@ -124,24 +126,21 @@ func (s *LibgangliaSink) Write(point lp.CCMetric) error {
 		return s.cstrCache[key]
 	}
 
-	// Get metric name
-	metricname := GangliaMetricRename(point.Name())
-
 	conf := GetCommonGangliaConfig(point)
 	if len(conf.Type) == 0 {
 		conf = GetGangliaConfig(point)
 	}
 	if len(conf.Type) == 0 {
-		return fmt.Errorf("metric %s has no 'value' field", metricname)
+		return fmt.Errorf("metric %q (Ganglia name %q) has no 'value' field", point.Name(), conf.Name)
 	}
 
 	if s.config.AddTypeToName {
-		metricname = GangliaMetricName(point)
+		conf.Name = GangliaMetricName(point)
 	}
 
 	c_value = C.CString(conf.Value)
 	c_type = lookup(conf.Type)
-	c_name = lookup(metricname)
+	c_name = lookup(conf.Name)
 
 	// Add unit
 	unit := ""
@@ -205,6 +204,8 @@ func (s *LibgangliaSink) Write(point lp.CCMetric) error {
 	C.Ganglia_metric_destroy(gmetric)
 	// Free the value C string, the only one not stored in the cache
 	C.free(unsafe.Pointer(c_value))
+	s.statsSentMetrics++
+	stats.ComponentStatInt(s.name, "sent_metrics", s.statsSentMetrics)
 	return err
 }
 
@@ -250,7 +251,7 @@ func NewLibgangliaSink(name string, config json.RawMessage) (Sink, error) {
 	if err != nil {
 		return nil, fmt.Errorf("error opening %s: %v", s.config.GangliaLib, err)
 	}
-
+	s.statsSentMetrics = 0
 	// Set up cache for the C strings
 	s.cstrCache = make(map[string]*C.char)
 	// s.cstrCache["globals"] = C.CString("globals")

sinks/metricSink.go

@@ -5,13 +5,13 @@ import (
 )
 
 type defaultSinkConfig struct {
-	MetaAsTags bool   `json:"meta_as_tags,omitempty"`
-	Type       string `json:"type"`
+	MetaAsTags []string `json:"meta_as_tags,omitempty"`
+	Type       string   `json:"type"`
 }
 
 type sink struct {
-	meta_as_tags bool   // Use meta data tags as tags
-	name         string // Name of the sink
+	meta_as_tags map[string]bool // Use meta data tags as tags
+	name         string          // Name of the sink
 }
 
 type Sink interface {

sinks/natsSink.go

@@ -55,7 +55,7 @@ func (s *NatsSink) connect() error {
 
 func (s *NatsSink) Write(m lp.CCMetric) error {
 	if s.client != nil {
-		_, err := s.encoder.Encode(m.ToPoint(s.config.MetaAsTags))
+		_, err := s.encoder.Encode(m.ToPoint(s.meta_as_tags))
 		if err != nil {
 			cclog.ComponentError(s.name, "Write:", err.Error())
 			return err
@@ -97,6 +97,11 @@ func NewNatsSink(name string, config json.RawMessage) (Sink, error) {
 		len(s.config.Database) == 0 {
 		return nil, errors.New("not all configuration variables set required by NatsSink")
 	}
+	// Create lookup map to use meta infos as tags in the output metric
+	s.meta_as_tags = make(map[string]bool)
+	for _, k := range s.config.MetaAsTags {
+		s.meta_as_tags[k] = true
+	}
 	// Setup Influx line protocol
 	s.buffer = &bytes.Buffer{}
 	s.buffer.Grow(1025)
@@ -105,7 +110,7 @@ func NewNatsSink(name string, config json.RawMessage) (Sink, error) {
 	s.encoder.SetMaxLineBytes(1024)
 	// Setup infos for connection
 	if err := s.connect(); err != nil {
-		return nil, fmt.Errorf("Unable to connect: %v", err)
+		return nil, fmt.Errorf("unable to connect: %v", err)
 	}
 	return s, nil
 }

sinks/prometheusSink.go

@@ -11,6 +11,7 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 	"github.com/gorilla/mux"
 	"github.com/prometheus/client_golang/prometheus"
 	"github.com/prometheus/client_golang/prometheus/promhttp"
@@ -29,11 +30,12 @@ type PrometheusSinkConfig struct {
 
 type PrometheusSink struct {
 	sink
-	config       PrometheusSinkConfig
-	labelMetrics map[string]*prometheus.GaugeVec
-	nodeMetrics  map[string]prometheus.Gauge
-	promWg       sync.WaitGroup
-	promServer   *http.Server
+	config           PrometheusSinkConfig
+	labelMetrics     map[string]*prometheus.GaugeVec
+	nodeMetrics      map[string]prometheus.Gauge
+	promWg           sync.WaitGroup
+	promServer       *http.Server
+	statsSentMetrics int64
 }
 
 func intToFloat64(input interface{}) (float64, error) {
@@ -113,6 +115,8 @@ func (s *PrometheusSink) newMetric(metric lp.CCMetric) error {
 		s.nodeMetrics[name] = new
 		prometheus.Register(new)
 	}
+	s.statsSentMetrics++
+	stats.ComponentStatInt(s.name, "sent_metrics", s.statsSentMetrics)
 	return nil
 }
 
@@ -146,6 +150,8 @@ func (s *PrometheusSink) updateMetric(metric lp.CCMetric) error {
 		}
 		s.nodeMetrics[name].Set(value)
 	}
+	s.statsSentMetrics++
+	stats.ComponentStatInt(s.name, "sent_metrics", s.statsSentMetrics)
 	return nil
 }
 

sinks/sampleSink.go

@@ -7,19 +7,23 @@ import (
 
 	cclog "github.com/ClusterCockpit/cc-metric-collector/internal/ccLogger"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 type SampleSinkConfig struct {
-	// defines JSON tags for 'type' and 'meta_as_tags'
+	// defines JSON tags for 'type' and 'meta_as_tags' (string list)
 	// See: metricSink.go
 	defaultSinkConfig
 	// Additional config options, for SampleSink
+
 }
 
 type SampleSink struct {
-	// declares elements 	'name' and 'meta_as_tags'
+	// declares elements 	'name' and 'meta_as_tags' (string to bool map!)
 	sink
 	config SampleSinkConfig // entry point to the SampleSinkConfig
+	// Stats counters
+	statsSentMetrics int64
 }
 
 // Implement functions required for Sink interface
@@ -28,7 +32,10 @@ type SampleSink struct {
 
 // Code to submit a single CCMetric to the sink
 func (s *SampleSink) Write(point lp.CCMetric) error {
+	// based on s.meta_as_tags use meta infos as tags
 	log.Print(point)
+	s.statsSentMetrics++
+	stats.ComponentStatInt(s.name, "sent_metrics", s.statsSentMetrics)
 	return nil
 }
 
@@ -62,6 +69,15 @@ func NewSampleSink(name string, config json.RawMessage) (Sink, error) {
 		}
 	}
 
+	// Initalize stats counters
+	s.statsSentMetrics = 0
+
+	// Create lookup map to use meta infos as tags in the output metric
+	s.meta_as_tags = make(map[string]bool)
+	for _, k := range s.config.MetaAsTags {
+		s.meta_as_tags[k] = true
+	}
+
 	// Check if all required fields in the config are set
 	// E.g. use 'len(s.config.Option) > 0' for string settings
 

sinks/sinkManager.go

@@ -102,13 +102,19 @@ func (sm *sinkManager) Start() {
 		}
 
 		toTheSinks := func(p lp.CCMetric) {
+			var wg sync.WaitGroup
 			// Send received metric to all outputs
 			cclog.ComponentDebug("SinkManager", "WRITE", p)
 			for _, s := range sm.sinks {
-				if err := s.Write(p); err != nil {
-					cclog.ComponentError("SinkManager", "WRITE", s.Name(), "write failed:", err.Error())
-				}
+				wg.Add(1)
+				go func(s Sink) {
+					if err := s.Write(p); err != nil {
+						cclog.ComponentError("SinkManager", "WRITE", s.Name(), "write failed:", err.Error())
+					}
+					wg.Done()
+				}(s)
 			}
+			wg.Wait()
 		}
 
 		for {

sinks/stdoutSink.go

@@ -8,6 +8,7 @@ import (
 
 	//	"time"
 	lp "github.com/ClusterCockpit/cc-metric-collector/internal/ccMetric"
+	stats "github.com/ClusterCockpit/cc-metric-collector/internal/metricRouter"
 )
 
 type StdoutSink struct {
@@ -17,6 +18,7 @@ type StdoutSink struct {
 		defaultSinkConfig
 		Output string `json:"output_file,omitempty"`
 	}
+	sentMetrics int64
 }
 
 func (s *StdoutSink) Write(m lp.CCMetric) error {
@@ -24,6 +26,8 @@ func (s *StdoutSink) Write(m lp.CCMetric) error {
 		s.output,
 		m.ToLineProtocol(s.meta_as_tags),
 	)
+	s.sentMetrics++
+	stats.ComponentStatInt(s.name, "sent_metrics", s.sentMetrics)
 	return nil
 }
 
@@ -63,7 +67,12 @@ func NewStdoutSink(name string, config json.RawMessage) (Sink, error) {
 			s.output = f
 		}
 	}
-	s.meta_as_tags = s.config.MetaAsTags
+	// Create lookup map to use meta infos as tags in the output metric
+	s.meta_as_tags = make(map[string]bool)
+	for _, k := range s.config.MetaAsTags {
+		s.meta_as_tags[k] = true
+	}
+	s.sentMetrics = 0
 
 	return s, nil
 }