Merge branch 'dev' into review_logging

2025-03-31 01:25:55 +02:00 · 2025-02-28 17:19:00 +01:00 · 2025-02-28 17:19:00 +01:00 · c964f09a4f
commit c964f09a4f
parent bd0cc69668 d7aefe0cf0
14 changed files with 219 additions and 64 deletions
--- a/configs/default_metrics.json
+++ b/configs/default_metrics.json
@ -0,0 +1,12 @@
+{
+  "clusters": [
+    {
+      "name": "fritz",
+      "default_metrics": "cpu_load, flops_any, core_power, lustre_open, mem_used, mem_bw, net_bytes_in"
+    },
+    {
+      "name": "alex",
+      "default_metrics": "flops_any, mem_bw, mem_used, vectorization_ratio"
+    }
+  ]
+}
--- a/go.mod
+++ b/go.mod
@ -29,6 +29,7 @@ require (
 	golang.org/x/crypto v0.35.0
 	golang.org/x/exp v0.0.0-20250218142911-aa4b98e5adaa
 	golang.org/x/oauth2 v0.27.0
+	golang.org/x/time v0.5.0
 )

 require (
--- a/go.sum
+++ b/go.sum
@ -325,6 +325,8 @@ golang.org/x/text v0.15.0/go.mod h1:18ZOQIKpY8NJVqYksKHtTdi31H5itFRjB5/qKTNYzSU=
 golang.org/x/text v0.21.0/go.mod h1:4IBbMaMmOPCJ8SecivzSH54+73PCFmPWxNTLm+vZkEQ=
 golang.org/x/text v0.22.0 h1:bofq7m3/HAFvbF51jz3Q9wLg3jkvSPuiZu/pD1XwgtM=
 golang.org/x/text v0.22.0/go.mod h1:YRoo4H8PVmsu+E3Ou7cqLVH8oXWIHVoX0jqUWALQhfY=
+golang.org/x/time v0.5.0 h1:o7cqy6amK/52YcAKIPlM3a+Fpj35zvRj2TP+e1xFSfk=
+golang.org/x/time v0.5.0/go.mod h1:3BpzKBy/shNhVucY/MWOyx10tF3SFh9QdLuxbVysPQM=
 golang.org/x/tools v0.0.0-20180917221912-90fa682c2a6e/go.mod h1:n7NCudcB/nEzxVGmLbDWY5pfWTLqBcC2KZ6jyYvM4mQ=
 golang.org/x/tools v0.0.0-20191119224855-298f0cb1881e/go.mod h1:b+2E5dAYhXwXZwtnZ6UAqBI28+e2cm9otk0dWdXHAEo=
 golang.org/x/tools v0.1.12/go.mod h1:hNGJHUnrk76NpqgfD5Aqm5Crs+Hm0VOH/i9J2+nxYbc=
--- a/internal/auth/auth.go
+++ b/internal/auth/auth.go
@ -10,11 +10,14 @@ import (
 	"database/sql"
 	"encoding/base64"
 	"errors"
+	"net"
 	"net/http"
 	"os"
 	"sync"
 	"time"

+	"golang.org/x/time/rate"
+
 	"github.com/ClusterCockpit/cc-backend/internal/config"
 	"github.com/ClusterCockpit/cc-backend/internal/repository"
 	"github.com/ClusterCockpit/cc-backend/pkg/log"
@ -32,6 +35,19 @@ var (
 	authInstance *Authentication
 )

+var ipUserLimiters sync.Map
+
+func getIPUserLimiter(ip, username string) *rate.Limiter {
+	key := ip + ":" + username
+	limiter, ok := ipUserLimiters.Load(key)
+	if !ok {
+		newLimiter := rate.NewLimiter(rate.Every(time.Hour/10), 10)
+		ipUserLimiters.Store(key, newLimiter)
+		return newLimiter
+	}
+	return limiter.(*rate.Limiter)
+}
+
 type Authentication struct {
 	sessionStore   *sessions.CookieStore
 	LdapAuth       *LdapAuthenticator
@ -88,7 +104,7 @@ func Init() {
 			authInstance.sessionStore = sessions.NewCookieStore(bytes)
 		}

-		if d, err := time.ParseDuration(config.Keys.SessionMaxAge); err != nil {
+		if d, err := time.ParseDuration(config.Keys.SessionMaxAge); err == nil {
 			authInstance.SessionMaxAge = d
 		}

@ -208,9 +224,21 @@ func (auth *Authentication) Login(
 	onfailure func(rw http.ResponseWriter, r *http.Request, loginErr error),
 ) http.Handler {
 	return http.HandlerFunc(func(rw http.ResponseWriter, r *http.Request) {
-		username := r.FormValue("username")
-		var dbUser *schema.User
+		ip, _, err := net.SplitHostPort(r.RemoteAddr)
+		if err != nil {
+			ip = r.RemoteAddr
+		}

+		username := r.FormValue("username")
+
+		limiter := getIPUserLimiter(ip, username)
+		if !limiter.Allow() {
+				log.Warnf("AUTH/RATE > Too many login attempts for combination IP: %s, Username: %s", ip, username)
+				onfailure(rw, r, errors.New("Too many login attempts, try again in a few minutes."))
+				return
+		}
+
+		var dbUser *schema.User
 		if username != "" {
 			var err error
 			dbUser, err = repository.GetUserRepository().GetUser(username)
--- a/internal/config/default_metrics.go
+++ b/internal/config/default_metrics.go
@ -0,0 +1,44 @@
+package config
+
+import (
+	"encoding/json"
+	"os"
+	"strings"
+)
+
+type DefaultMetricsCluster struct {
+	Name           string `json:"name"`
+	DefaultMetrics string `json:"default_metrics"`
+}
+
+type DefaultMetricsConfig struct {
+	Clusters []DefaultMetricsCluster `json:"clusters"`
+}
+
+func LoadDefaultMetricsConfig() (*DefaultMetricsConfig, error) {
+	filePath := "default_metrics.json"
+	if _, err := os.Stat(filePath); os.IsNotExist(err) {
+		return nil, nil
+	}
+	data, err := os.ReadFile(filePath)
+	if err != nil {
+		return nil, err
+	}
+	var cfg DefaultMetricsConfig
+	if err := json.Unmarshal(data, &cfg); err != nil {
+		return nil, err
+	}
+	return &cfg, nil
+}
+
+func ParseMetricsString(s string) []string {
+	parts := strings.Split(s, ",")
+	var metrics []string
+	for _, p := range parts {
+		trimmed := strings.TrimSpace(p)
+		if trimmed != "" {
+			metrics = append(metrics, trimmed)
+		}
+	}
+	return metrics
+}
--- a/internal/repository/user.go
+++ b/internal/repository/user.go
@ -19,6 +19,7 @@ import (
 	sq "github.com/Masterminds/squirrel"
 	"github.com/jmoiron/sqlx"
 	"golang.org/x/crypto/bcrypt"
+	"github.com/ClusterCockpit/cc-backend/internal/config"
 )

 var (
@ -127,6 +128,30 @@ func (r *UserRepository) AddUser(user *schema.User) error {
 	}

 	log.Infof("new user %#v created (roles: %s, auth-source: %d, projects: %s)", user.Username, rolesJson, user.AuthSource, projectsJson)
+
+	defaultMetricsCfg, err := config.LoadDefaultMetricsConfig()
+	if err != nil {
+		log.Errorf("Error loading default metrics config: %v", err)
+	} else if defaultMetricsCfg != nil {
+		for _, cluster := range defaultMetricsCfg.Clusters {
+			metricsArray := config.ParseMetricsString(cluster.DefaultMetrics)
+			metricsJSON, err := json.Marshal(metricsArray)
+			if err != nil {
+				log.Errorf("Error marshaling default metrics for cluster %s: %v", cluster.Name, err)
+				continue
+			}
+			confKey := "job_view_selectedMetrics:" + cluster.Name
+			if _, err := sq.Insert("configuration").
+				Columns("username", "confkey", "value").
+				Values(user.Username, confKey, string(metricsJSON)).
+				RunWith(r.DB).Exec(); err != nil {
+				log.Errorf("Error inserting default job view metrics for user %s and cluster %s: %v", user.Username, cluster.Name, err)
+			} else {
+				log.Infof("Default job view metrics for user %s and cluster %s set to %s", user.Username, cluster.Name, string(metricsJSON))
+			}
+		}
+	}
+
 	return nil
 }

--- a/web/frontend/src/Analysis.root.svelte
+++ b/web/frontend/src/Analysis.root.svelte
@ -20,6 +20,7 @@
    Card,
    Table,
    Icon,
+    Tooltip
  } from "@sveltestrap/sveltestrap";
  import {
    init,
@ -70,6 +71,8 @@
    ...new Set([...metricsInHistograms, ...metricsInScatterplots.flat()]),
  ];

+  $: clusterName = cluster?.name ? cluster.name : cluster;
+
  const sortOptions = [
    { key: "totalWalltime", label: "Walltime" },
    { key: "totalNodeHours", label: "Node Hours" },
@ -159,6 +162,7 @@
          groupBy: $groupBy
        ) {
          id
+          name
          totalWalltime
          totalNodeHours
          totalCoreHours
@ -422,15 +426,22 @@
              <tr>
                <td><Icon name="circle-fill" style="color: {colors[i]};" /></td>
                {#if groupSelection.key == "user"}
-                  <th scope="col"
-                    ><a href="/monitoring/user/{te.id}?cluster={cluster}"
+                  <th scope="col" id="topName-{te.id}"
+                    ><a href="/monitoring/user/{te.id}?cluster={clusterName}"
                      >{te.id}</a
                    ></th
                  >
+                  {#if te?.name}
+                    <Tooltip
+                      target={`topName-${te.id}`}
+                      placement="left"
+                      >{te.name}</Tooltip
+                    >
+                  {/if}
                {:else}
                  <th scope="col"
                    ><a
-                      href="/monitoring/jobs/?cluster={cluster}&project={te.id}&projectMatch=eq"
+                      href="/monitoring/jobs/?cluster={clusterName}&project={te.id}&projectMatch=eq"
                      >{te.id}</a
                    ></th
                  >
--- a/web/frontend/src/Job.root.svelte
+++ b/web/frontend/src/Job.root.svelte
@ -58,7 +58,8 @@
  let plots = {},
    statsTable

-  let missingMetrics = [],
+  let availableMetrics = new Set(),
+    missingMetrics = [],
    missingHosts = [],
    somethingMissing = false;

@ -128,7 +129,12 @@

    const pendingMetrics = [
      ...(ccconfig[`job_view_selectedMetrics:${job.cluster}`] ||
-        ccconfig[`job_view_selectedMetrics`]
+      $initq.data.globalMetrics.reduce((names, gm) => {
+          if (gm.availability.find((av) => av.cluster === job.cluster)) {
+            names.push(gm.name);
+          }
+          return names;
+        }, [])
      ),
      ...(ccconfig[`job_view_nodestats_selectedMetrics:${job.cluster}`] ||
        ccconfig[`job_view_nodestats_selectedMetrics`]
@ -293,7 +299,7 @@
      {#if $initq.data}
        <Col xs="auto">
            <Button outline on:click={() => (isMetricsSelectionOpen = true)} color="primary">
-              Select Metrics
+              Select Metrics (Selected {selectedMetrics.length} of {availableMetrics.size} available)
            </Button>
        </Col>
      {/if}
@ -431,6 +437,7 @@
    configName="job_view_selectedMetrics"
    bind:metrics={selectedMetrics}
    bind:isOpen={isMetricsSelectionOpen}
+    bind:allMetrics={availableMetrics}
  />
 {/if}

--- a/web/frontend/src/Status.root.svelte
+++ b/web/frontend/src/Status.root.svelte
@ -19,6 +19,7 @@
    Progress,
    Icon,
    Button,
+    Tooltip
  } from "@sveltestrap/sveltestrap";
  import {
    queryStore,
@ -75,9 +76,9 @@
    );

  let isHistogramSelectionOpen = false;
-  $: metricsInHistograms = cluster
-    ? ccconfig[`user_view_histogramMetrics:${cluster}`] || []
-    : ccconfig.user_view_histogramMetrics || [];
+  $: selectedHistograms = cluster
+    ? ccconfig[`user_view_histogramMetrics:${cluster}`] || ( ccconfig['user_view_histogramMetrics'] || [] )
+    : ccconfig['user_view_histogramMetrics'] || [];

  const client = getContextClient();
  // Note: nodeMetrics are requested on configured $timestep resolution
@ -90,7 +91,7 @@
        $metrics: [String!]
        $from: Time!
        $to: Time!
-        $metricsInHistograms: [String!]
+        $selectedHistograms: [String!]
      ) {
        nodeMetrics(
          cluster: $cluster
@ -116,7 +117,7 @@
          }
        }

-        stats: jobsStatistics(filter: $filter, metrics: $metricsInHistograms) {
+        stats: jobsStatistics(filter: $filter, metrics: $selectedHistograms) {
          histDuration {
            count
            value
@ -157,7 +158,7 @@
      from: from.toISOString(),
      to: to.toISOString(),
      filter: [{ state: ["running"] }, { cluster: { eq: cluster } }],
-      metricsInHistograms: metricsInHistograms,
+      selectedHistograms: selectedHistograms,
    },
  });

@ -177,6 +178,7 @@
          groupBy: USER
        ) {
          id
+          name
          totalJobs
          totalNodes
          totalCores
@ -515,12 +517,19 @@
            {#each $topUserQuery.data.topUser as tu, i}
              <tr>
                <td><Icon name="circle-fill" style="color: {colors[i]};" /></td>
-                <th scope="col"
+                <th scope="col" id="topName-{tu.id}"
                  ><a
                    href="/monitoring/user/{tu.id}?cluster={cluster}&state=running"
                    >{tu.id}</a
                  ></th
                >
+                {#if tu?.name}
+                  <Tooltip
+                    target={`topName-${tu.id}`}
+                    placement="left"
+                    >{tu.name}</Tooltip
+                  >
+                {/if}
                <td>{tu[topUserSelection.key]}</td>
              </tr>
            {/each}
@ -652,7 +661,7 @@

  <!-- Selectable Stats as Histograms : Average Values of Running Jobs -->

-  {#if metricsInHistograms}
+  {#if selectedHistograms}
    {#key $mainQuery.data.stats[0].histMetrics}
      <PlotGrid
        let:item
@ -675,6 +684,6 @@

 <HistogramSelection
  bind:cluster
-  bind:metricsInHistograms
+  bind:selectedHistograms
  bind:isOpen={isHistogramSelectionOpen}
 />
--- a/web/frontend/src/User.root.svelte
+++ b/web/frontend/src/User.root.svelte
@ -68,16 +68,16 @@
  let durationBinOptions = ["1m","10m","1h","6h","12h"];
  let metricBinOptions = [10, 20, 50, 100];

-  $: metricsInHistograms = selectedCluster
-    ? ccconfig[`user_view_histogramMetrics:${selectedCluster}`] || []
-    : ccconfig.user_view_histogramMetrics || [];
+  $: selectedHistograms = selectedCluster
+    ? ccconfig[`user_view_histogramMetrics:${selectedCluster}`] || ( ccconfig['user_view_histogramMetrics'] || [] )
+    : ccconfig['user_view_histogramMetrics'] || [];

  const client = getContextClient();
  $: stats = queryStore({
    client: client,
    query: gql`
-      query ($jobFilters: [JobFilter!]!, $metricsInHistograms: [String!], $numDurationBins: String, $numMetricBins: Int) {
-        jobsStatistics(filter: $jobFilters, metrics: $metricsInHistograms, numDurationBins: $numDurationBins , numMetricBins: $numMetricBins ) {
+      query ($jobFilters: [JobFilter!]!, $selectedHistograms: [String!], $numDurationBins: String, $numMetricBins: Int) {
+        jobsStatistics(filter: $jobFilters, metrics: $selectedHistograms, numDurationBins: $numDurationBins , numMetricBins: $numMetricBins ) {
          totalJobs
          shortJobs
          totalWalltime
@ -104,7 +104,7 @@
        }
      }
    `,
-    variables: { jobFilters, metricsInHistograms, numDurationBins, numMetricBins },
+    variables: { jobFilters, selectedHistograms, numDurationBins, numMetricBins },
  });

  onMount(() => filterComponent.updateFilters());
@ -290,7 +290,7 @@
    </InputGroup>
  </Col>
 </Row>
-{#if metricsInHistograms?.length > 0}
+{#if selectedHistograms?.length > 0}
  {#if $stats.error}
    <Row>
      <Col>
@ -357,6 +357,6 @@

 <HistogramSelection
  bind:cluster={selectedCluster}
-  bind:metricsInHistograms
+  bind:selectedHistograms
  bind:isOpen={isHistogramSelectionOpen}
 />
--- a/web/frontend/src/generic/filters/Cluster.svelte
+++ b/web/frontend/src/generic/filters/Cluster.svelte
@ -43,26 +43,31 @@
  <ModalBody>
    {#if $initialized}
      <h4>Cluster</h4>
-      <ListGroup>
-        <ListGroupItem
-          disabled={disableClusterSelection}
-          active={pendingCluster == null}
-          on:click={() => ((pendingCluster = null), (pendingPartition = null))}
-        >
-          Any Cluster
-        </ListGroupItem>
-        {#each clusters as cluster}
+      {#if disableClusterSelection}
+        <Button color="info" class="w-100 mb-2" disabled><b>Info: Cluster Selection Disabled in This View</b></Button>
+        <Button outline color="primary" class="w-100 mb-2" disabled><b>Selected Cluster: {cluster}</b></Button>
+      {:else}
+        <ListGroup>
          <ListGroupItem
            disabled={disableClusterSelection}
-            active={pendingCluster == cluster.name}
-            on:click={() => (
-              (pendingCluster = cluster.name), (pendingPartition = null)
-            )}
+            active={pendingCluster == null}
+            on:click={() => ((pendingCluster = null), (pendingPartition = null))}
          >
-            {cluster.name}
+            Any Cluster
          </ListGroupItem>
-        {/each}
-      </ListGroup>
+          {#each clusters as cluster}
+            <ListGroupItem
+              disabled={disableClusterSelection}
+              active={pendingCluster == cluster.name}
+              on:click={() => (
+                (pendingCluster = cluster.name), (pendingPartition = null)
+              )}
+            >
+              {cluster.name}
+            </ListGroupItem>
+          {/each}
+        </ListGroup>
+      {/if}
    {/if}
    {#if $initialized && pendingCluster != null}
      <br />
--- a/web/frontend/src/generic/plots/Roofline.svelte
+++ b/web/frontend/src/generic/plots/Roofline.svelte
@ -179,7 +179,7 @@
  function render(plotData) {
    if (plotData) {
      const opts = {
-        title: "",
+        title: "CPU Roofline Diagram",
        mode: 2,
        width: width,
        height: height,
--- a/web/frontend/src/generic/select/HistogramSelection.svelte
+++ b/web/frontend/src/generic/select/HistogramSelection.svelte
@ -3,7 +3,7 @@

    Properties:
    - `cluster String`: Currently selected cluster
-    - `metricsInHistograms [String]`: The currently selected metrics to display as histogram
+    - `selectedHistograms [String]`: The currently selected metrics to display as histogram
    - ìsOpen Bool`: Is selection opened
 -->

@ -21,22 +21,27 @@
  import { gql, getContextClient, mutationStore } from "@urql/svelte";

  export let cluster;
-  export let metricsInHistograms;
+  export let selectedHistograms;
  export let isOpen;

  const client = getContextClient();
  const initialized = getContext("initialized");

-  let availableMetrics = []
+  function loadHistoMetrics(isInitialized, thisCluster) {
+    if (!isInitialized) return [];

-  function loadHistoMetrics(isInitialized) {
-    if (!isInitialized) return;
-    const rawAvailableMetrics = getContext("globalMetrics").filter((gm) => gm?.footprint).map((fgm) => { return fgm.name })
-    availableMetrics = [...rawAvailableMetrics]
+    if (!thisCluster) {
+      return getContext("globalMetrics")
+      .filter((gm) => gm?.footprint)
+      .map((fgm) => { return fgm.name })
+    } else {
+      return getContext("globalMetrics")
+      .filter((gm) => gm?.availability.find((av) => av.cluster == thisCluster))
+      .filter((agm) => agm?.footprint)
+      .map((afgm) => { return afgm.name })
+    }
  }

-  let pendingMetrics = [...metricsInHistograms]; // Copy
-
  const updateConfigurationMutation = ({ name, value }) => {
    return mutationStore({
      client: client,
@ -61,17 +66,16 @@
  }

  function closeAndApply() {
-    metricsInHistograms = [...pendingMetrics]; // Set for parent
    isOpen = !isOpen;
    updateConfiguration({
      name: cluster
        ? `user_view_histogramMetrics:${cluster}`
        : "user_view_histogramMetrics",
-      value: metricsInHistograms,
+      value: selectedHistograms,
    });
  }

-  $: loadHistoMetrics($initialized);
+  $: availableMetrics = loadHistoMetrics($initialized, cluster);

 </script>

@ -81,7 +85,7 @@
    <ListGroup>
      {#each availableMetrics as metric (metric)}
        <ListGroupItem>
-          <input type="checkbox" bind:group={pendingMetrics} value={metric} />
+          <input type="checkbox" bind:group={selectedHistograms} value={metric} />
          {metric}
        </ListGroupItem>
      {/each}
--- a/web/frontend/src/job/StatsTable.svelte
+++ b/web/frontend/src/job/StatsTable.svelte
@ -18,6 +18,8 @@
    InputGroup,
    InputGroupText,
    Icon,
+    Row,
+    Col
  } from "@sveltestrap/sveltestrap";
  import { maxScope } from "../generic/utils.js";
  import StatsTableEntry from "./StatsTableEntry.svelte";
@ -26,7 +28,7 @@
  export let job;
  export let jobMetrics;

-  const allMetrics = [...new Set(jobMetrics.map((m) => m.name))].sort()
+  const sortedJobMetrics = [...new Set(jobMetrics.map((m) => m.name))].sort()
  const scopesForMetric = (metric) =>
      jobMetrics.filter((jm) => jm.name == metric).map((jm) => jm.scope);

@ -34,11 +36,12 @@
    selectedScopes = {},
    sorting = {},
    isMetricSelectionOpen = false,
+    availableMetrics = new Set(),
    selectedMetrics =
      getContext("cc-config")[`job_view_nodestats_selectedMetrics:${job.cluster}`] ||
      getContext("cc-config")["job_view_nodestats_selectedMetrics"];

-  for (let metric of allMetrics) {
+  for (let metric of sortedJobMetrics) {
    // Not Exclusive or Multi-Node: get maxScope directly (mostly: node)
    //   -> Else: Load smallest available granularity as default as per availability
    const availableScopes = scopesForMetric(metric);
@ -95,15 +98,19 @@
  };
 </script>

+<Row>
+  <Col class="m-2">
+    <Button outline on:click={() => (isMetricSelectionOpen = true)} class="w-auto px-2" color="primary">
+      Select Metrics (Selected {selectedMetrics.length} of {availableMetrics.size} available)
+    </Button>
+  </Col>
+</Row>
+<hr class="mb-1 mt-1"/>
 <Table class="mb-0">
  <thead>
    <!-- Header Row 1: Selectors -->
    <tr>
-      <th>
-        <Button outline on:click={() => (isMetricSelectionOpen = true)} class="w-100 px-2" color="primary">
-          Select Metrics
-        </Button>
-      </th>
+      <th/>
      {#each selectedMetrics as metric}
        <!-- To Match Row-2 Header Field Count-->
        <th colspan={selectedScopes[metric] == "node" ? 3 : 4}>
@ -163,7 +170,7 @@
 <MetricSelection
  cluster={job.cluster}
  configName="job_view_nodestats_selectedMetrics"
-  allMetrics={new Set(allMetrics)}
+  bind:allMetrics={availableMetrics}
  bind:metrics={selectedMetrics}
  bind:isOpen={isMetricSelectionOpen}
 />