Fixed docker-entrypoint.sh scripts

.gitignore

@@ -3,6 +3,10 @@ data/job-archive/**
 data/influxdb
 data/sqldata
 data/cc-metric-store
+data/cc-metric-store-source
+data/ldap
+data/mariadb
+data/slurm
 cc-backend
 cc-backend/**
 .vscode

docker-compose.yml

@@ -104,6 +104,8 @@ services:
     volumes:
       - ${DATADIR}/slurm/home:/home
       - ${DATADIR}/slurm/secret:/.secret
+      - ./slurm/worker/cgroup.conf:/home/config/cgroup.conf
+      - ./slurm/controller/slurm.conf:/home/config/slurm.conf
     ports:
       - "6818:6818"
 

setupDev.sh

@@ -1,4 +1,20 @@
 #!/bin/bash
+echo ""
+echo "-----------------------------------------------------------------"
+echo "Welcome to cc-docker automatic deployment script."
+echo "Make sure you have sudo rights to run docker services"
+echo "This script assumes that docker command is added to sudo group"
+echo "This means that docker commands do not explicitly require"
+echo "'sudo' keyword to run. You can use this following command:"
+echo ""
+echo "sudo groupadd docker"
+echo "sudo usermod -aG docker $USER"
+echo ""
+echo "This will add docker to the sudo usergroup and all the docker"
+echo "command will run as sudo by default without requiring"
+echo "'sudo' keyword."  
+echo "-----------------------------------------------------------------"
+echo ""
 
 # Check cc-backend, touch job.db if exists
 if [ ! -d cc-backend ]; then
@@ -7,8 +23,6 @@ if [ ! -d cc-backend ]; then
     exit
 else
     cd cc-backend
-    make
-
     if [ ! -d var ]; then
         wget https://hpc-mover.rrze.uni-erlangen.de/HPC-Data/0x7b58aefb/eig7ahyo6fo2bais0ephuf2aitohv1ai/job-archive-demo.tar
         tar xf job-archive-demo.tar
@@ -17,6 +31,8 @@ else
         cp ./configs/env-template.txt .env
         cp ./configs/config-demo.json config.json
 
+        make
+
         ./cc-backend -migrate-db
         ./cc-backend --init-db --add-user demo:admin:AdminDev
         cd ..
@@ -28,6 +44,8 @@ else
     fi
 fi
 
+mkdir -m777 data
+
 # Download unedited checkpoint files to ./data/cc-metric-store-source/checkpoints
 if [ ! -d data/cc-metric-store-source ]; then
     mkdir -p data/cc-metric-store-source/checkpoints
@@ -80,6 +98,5 @@ docker-compose up -d
 
 echo ""
 echo "Setup complete, containers are up by default: Shut down with 'docker-compose down'."
-echo "Use './cc-backend/cc-backend' to start cc-backend."
+echo "Use './cc-backend/cc-backend -server' to start cc-backend."
 echo "Use scripts in /scripts to load data into influx or mariadb."
-# ./cc-backend/cc-backend

slurm/controller/docker-entrypoint.sh

@@ -55,7 +55,7 @@ _munge_start() {
     /usr/sbin/create-munge-key -r -f
     sh -c "dd if=/dev/urandom bs=1 count=1024 > /etc/munge/munge.key"
     chown munge: /etc/munge/munge.key
-    chmod 400 /etc/munge/munge.key
+    chmod 600 /etc/munge/munge.key
     sudo -u munge /sbin/munged
     munge -n
     munge -n | unmunge
@@ -64,6 +64,10 @@ _munge_start() {
 
 # copy secrets to /.secret directory for other nodes
 _copy_secrets() {
+    while [ ! -f /home/worker/worker-secret.tar.gz ]; do
+        echo -n "."
+        sleep 1
+    done
     cp /home/worker/worker-secret.tar.gz /.secret/worker-secret.tar.gz
     cp /home/worker/setup-worker-ssh.sh /.secret/setup-worker-ssh.sh
     cp /etc/munge/munge.key /.secret/munge.key

slurm/controller/slurm.conf

@@ -21,7 +21,7 @@ SwitchType=switch/none
 MpiDefault=none
 SlurmctldPidFile=/var/run/slurmd/slurmctld.pid
 SlurmdPidFile=/var/run/slurmd/slurmd.pid
-ProctrackType=proctrack/linuxproc
+# ProctrackType=proctrack/linuxproc
 #PluginDir=
 #CacheGroups=0
 #FirstJobId=
@@ -58,7 +58,7 @@ SchedulerType=sched/backfill
 #SchedulerRootFilter=
 SelectType=select/cons_res
 SelectTypeParameters=CR_CPU_Memory
-FastSchedule=1
+# FastSchedule=1
 #PriorityType=priority/multifactor
 #PriorityDecayHalfLife=14-0
 #PriorityUsageResetPeriod=14-0

slurm/database/docker-entrypoint.sh

@@ -68,7 +68,7 @@ _slurmdbd() {
   fi
   echo "Starting slurmdbd"
   cp /etc/slurm/slurmdbd.conf /.secret/slurmdbd.conf
-  /usr/sbin/slurmdbd
+  /usr/sbin/slurmdbd -Dvv
   echo "Started slurmdbd"
 }
 

slurm/database/slurmdbd.conf

@@ -24,7 +24,7 @@ SlurmUser=slurm
 DebugLevel=4
 #DefaultQOS=normal,standby
 LogFile=/var/log/slurm/slurmdbd.log
-PidFile=/var/run/slurmdbd/slurmdbd.pid
+# PidFile=/var/run/slurmdbd/slurmdbd.pid
 #PluginDir=/usr/lib/slurm
 #PrivateData=accounts,users,usage,jobs
 #TrackWCKey=yes

slurm/worker/cgroup.conf

@@ -0,0 +1,5 @@
+CgroupPlugin=cgroup/v1
+ConstrainCores=yes
+ConstrainDevices=no
+ConstrainRAMSpace=yes
+ConstrainSwapSpace=yes