Merge pull request #268 from ClusterCockpit/hotfix

Hotfix

cmd/cc-backend/main.go

@@ -76,6 +76,9 @@ const configString = `
         "kind": "file",
         "path": "./var/job-archive"
     },
+    "jwts": {
+        "max-age": "2000h"
+    },
     "clusters": [
         {
             "name": "name",
@@ -115,15 +118,15 @@ func initEnv() {
 		os.Exit(0)
 	}
 
-	if err := os.WriteFile("config.json", []byte(configString), 0666); err != nil {
+	if err := os.WriteFile("config.json", []byte(configString), 0o666); err != nil {
 		log.Fatalf("Writing config.json failed: %s", err.Error())
 	}
 
-	if err := os.WriteFile(".env", []byte(envString), 0666); err != nil {
+	if err := os.WriteFile(".env", []byte(envString), 0o666); err != nil {
 		log.Fatalf("Writing .env failed: %s", err.Error())
 	}
 
-	if err := os.Mkdir("var", 0777); err != nil {
+	if err := os.Mkdir("var", 0o777); err != nil {
 		log.Fatalf("Mkdir var failed: %s", err.Error())
 	}
 

internal/metricdata/metricdata.go

@@ -109,8 +109,8 @@ func LoadData(job *schema.Job,
 			jd, err = repo.LoadData(job, metrics, scopes, ctx)
 			if err != nil {
 				if len(jd) != 0 {
-					log.Errorf("partial error: %s", err.Error())
-					return err, 0, 0
+					log.Warnf("partial error: %s", err.Error())
+					// return err, 0, 0 // Reactivating will block archiving on one partial error
 				} else {
 					log.Error("Error while loading job data from metric repository")
 					return err, 0, 0

internal/repository/job.go

@@ -520,7 +520,7 @@ func (r *JobRepository) archivingWorker() {
 			// not using meta data, called to load JobMeta into Cache?
 			// will fail if job meta not in repository
 			if _, err := r.FetchMetadata(job); err != nil {
-				log.Errorf("archiving job (dbid: %d) failed: %s", job.ID, err.Error())
+				log.Errorf("archiving job (dbid: %d) failed at check metadata step: %s", job.ID, err.Error())
 				r.UpdateMonitoringStatus(job.ID, schema.MonitoringStatusArchivingFailed)
 				continue
 			}
@@ -529,14 +529,14 @@ func (r *JobRepository) archivingWorker() {
 			// TODO: Maybe use context with cancel/timeout here
 			jobMeta, err := metricdata.ArchiveJob(job, context.Background())
 			if err != nil {
-				log.Errorf("archiving job (dbid: %d) failed: %s", job.ID, err.Error())
+				log.Errorf("archiving job (dbid: %d) failed at archiving job step: %s", job.ID, err.Error())
 				r.UpdateMonitoringStatus(job.ID, schema.MonitoringStatusArchivingFailed)
 				continue
 			}
 
 			// Update the jobs database entry one last time:
 			if err := r.MarkArchived(job.ID, schema.MonitoringStatusArchivingSuccessful, jobMeta.Statistics); err != nil {
-				log.Errorf("archiving job (dbid: %d) failed: %s", job.ID, err.Error())
+				log.Errorf("archiving job (dbid: %d) failed at marking archived step: %s", job.ID, err.Error())
 				continue
 			}
 			log.Debugf("archiving job %d took %s", job.JobID, time.Since(start))

internal/repository/query.go

@@ -236,6 +236,9 @@ func buildStringCondition(field string, cond *model.StringInput, query sq.Select
 }
 
 func buildMetaJsonCondition(jsonField string, cond *model.StringInput, query sq.SelectBuilder) sq.SelectBuilder {
+	// Verify and Search Only in Valid Jsons
+	query = query.Where("JSON_VALID(meta_data)")
+	// add "AND" Sql query Block for field match
 	if cond.Eq != nil {
 		return query.Where("JSON_EXTRACT(meta_data, \"$."+jsonField+"\") = ?", *cond.Eq)
 	}

internal/routerConfig/routes.go

@@ -302,11 +302,19 @@ func HandleSearchBar(rw http.ResponseWriter, r *http.Request, buildInfo web.Buil
 			case "jobId":
 				http.Redirect(rw, r, "/monitoring/jobs/?jobId="+url.QueryEscape(strings.Trim(splitSearch[1], " ")), http.StatusFound) // All Users: Redirect to Tablequery
 			case "jobName":
-				http.Redirect(rw, r, "/monitoring/jobs/?jobName="+url.QueryEscape(strings.Trim(splitSearch[1], " ")), http.StatusFound) // All Users: Redirect to Tablequery
+				// Add Last 30 Days to migitate timeouts
+				untilTime := strconv.FormatInt(time.Now().Unix(), 10)
+				fromTime := strconv.FormatInt((time.Now().Unix() - int64(30*24*3600)), 10)
+
+				http.Redirect(rw, r, "/monitoring/jobs/?startTime="+fromTime+"-"+untilTime+"&jobName="+url.QueryEscape(strings.Trim(splitSearch[1], " ")), http.StatusFound) // All Users: Redirect to Tablequery
 			case "projectId":
 				http.Redirect(rw, r, "/monitoring/jobs/?projectMatch=eq&project="+url.QueryEscape(strings.Trim(splitSearch[1], " ")), http.StatusFound) // All Users: Redirect to Tablequery
 			case "arrayJobId":
-				http.Redirect(rw, r, "/monitoring/jobs/?arrayJobId="+url.QueryEscape(strings.Trim(splitSearch[1], " ")), http.StatusFound) // All Users: Redirect to Tablequery
+				// Add Last 30 Days to migitate timeouts
+				untilTime := strconv.FormatInt(time.Now().Unix(), 10)
+				fromTime := strconv.FormatInt((time.Now().Unix() - int64(30*24*3600)), 10)
+
+				http.Redirect(rw, r, "/monitoring/jobs/?startTime="+fromTime+"-"+untilTime+"&arrayJobId="+url.QueryEscape(strings.Trim(splitSearch[1], " ")), http.StatusFound) // All Users: Redirect to Tablequery
 			case "username":
 				if user.HasAnyRole([]schema.Role{schema.RoleAdmin, schema.RoleSupport, schema.RoleManager}) {
 					http.Redirect(rw, r, "/monitoring/users/?user="+url.QueryEscape(strings.Trim(splitSearch[1], " ")), http.StatusFound)
@@ -339,7 +347,11 @@ func HandleSearchBar(rw http.ResponseWriter, r *http.Request, buildInfo web.Buil
 			} else if project != "" {
 				http.Redirect(rw, r, "/monitoring/jobs/?projectMatch=eq&project="+url.QueryEscape(project), http.StatusFound) // projectId (equal)
 			} else if jobname != "" {
-				http.Redirect(rw, r, "/monitoring/jobs/?jobName="+url.QueryEscape(jobname), http.StatusFound) // JobName (contains)
+				// Add Last 30 Days to migitate timeouts
+				untilTime := strconv.FormatInt(time.Now().Unix(), 10)
+				fromTime := strconv.FormatInt((time.Now().Unix() - int64(30*24*3600)), 10)
+
+				http.Redirect(rw, r, "/monitoring/jobs/?startTime="+fromTime+"-"+untilTime+"&jobName="+url.QueryEscape(jobname), http.StatusFound) // 30D Fitler + JobName (contains)
 			} else {
 				web.RenderTemplate(rw, "message.tmpl", &web.Page{Title: "Info", MsgType: "alert-info", Message: "Search without result", User: *user, Roles: availableRoles, Build: buildInfo})
 			}

web/frontend/src/Job.root.svelte

@@ -165,10 +165,11 @@
         .find((c) => c.name == job.cluster)
         .metricConfig.map((mc) => mc.name);
 
-    // Metric not found in JobMetrics && Metric not explicitly disabled: Was expected, but is Missing
+    // Metric not found in JobMetrics && Metric not explicitly disabled in config or deselected: Was expected, but is Missing
     missingMetrics = metricNames.filter(
       (metric) =>
         !metrics.some((jm) => jm.name == metric) &&
+        selectedMetrics.includes(metric) && 
         !checkMetricDisabled(
           metric,
           $initq.data.job.cluster,
@@ -306,9 +307,6 @@
       </Button>
     {/if}
   </Col>
-  <!--     <Col xs="auto">
-        <Zoom timeseriesPlots={plots} />
-    </Col> -->
 </Row>
 <Row>
   <Col>

web/frontend/src/Jobs.root.svelte

@@ -14,7 +14,7 @@
   import Refresher from "./joblist/Refresher.svelte";
   import Sorting from "./joblist/SortSelection.svelte";
   import MetricSelection from "./MetricSelection.svelte";
-  import UserOrProject from "./filters/UserOrProject.svelte";
+  import TextFilter from "./filters/TextFilter.svelte";
 
   const { query: initq } = init();
 
@@ -38,6 +38,7 @@
     ? !!ccconfig[`plot_list_showFootprint:${filterPresets.cluster}`]
     : !!ccconfig.plot_list_showFootprint;
   let selectedCluster = filterPresets?.cluster ? filterPresets.cluster : null;
+  let presetProject = filterPresets?.project ? filterPresets.project : ""
 
   // The filterPresets are handled by the Filters component,
   // so we need to wait for it to be ready before we can start a query.
@@ -86,7 +87,8 @@
   </Col>
 
   <Col xs="3" style="margin-left: auto;">
-    <UserOrProject
+    <TextFilter
+      {presetProject}
       bind:authlevel
       bind:roles
       on:update={({ detail }) => filterComponent.update(detail)}

web/frontend/src/Metric.svelte

@@ -33,8 +33,17 @@
     error = null;
   let selectedScope = minScope(scopes);
 
+  let statsPattern = /(.*)-stats$/
+  let statsSeries = rawData.map((data) => data?.statisticsSeries ? data.statisticsSeries : null)
+  let selectedScopeIndex
+
   $: availableScopes = scopes;
-  $: selectedScopeIndex = scopes.findIndex((s) => s == selectedScope);
+  $: patternMatches = statsPattern.exec(selectedScope)
+  $: if (!patternMatches) {
+      selectedScopeIndex = scopes.findIndex((s) => s == selectedScope);
+    } else {
+      selectedScopeIndex = scopes.findIndex((s) => s == patternMatches[1]);
+    }
   $: data = rawData[selectedScopeIndex];
   $: series = data?.series.filter(
     (series) => selectedHost == null || series.hostname == selectedHost,
@@ -62,6 +71,7 @@
       if (jm.scope != "node") {
         scopes = [...scopes, jm.scope];
         rawData.push(jm.metric);
+        statsSeries = rawData.map((data) => data?.statisticsSeries ? data.statisticsSeries : null)
         selectedScope = jm.scope;
         selectedScopeIndex = scopes.findIndex((s) => s == jm.scope);
         dispatch("more-loaded", jm);
@@ -79,15 +89,18 @@
       : "") + (metricConfig?.unit?.base ? metricConfig.unit.base : "")})
   </InputGroupText>
   <select class="form-select" bind:value={selectedScope}>
-    {#each availableScopes as scope}
+    {#each availableScopes as scope, index}
       <option value={scope}>{scope}</option>
+      {#if statsSeries[index]}
+        <option value={scope + '-stats'}>stats series ({scope})</option>
+      {/if}
     {/each}
     {#if availableScopes.length == 1 && metricConfig?.scope != "node"}
       <option value={"load-more"}>Load more...</option>
     {/if}
   </select>
   {#if job.resources.length > 1}
-    <select class="form-select" bind:value={selectedHost}>
+    <select class="form-select" bind:value={selectedHost} disabled={patternMatches}>
       <option value={null}>All Hosts</option>
       {#each job.resources as { hostname }}
         <option value={hostname}>{hostname}</option>
@@ -100,7 +113,21 @@
     <Spinner />
   {:else if error != null}
     <Card body color="danger">{error.message}</Card>
-  {:else if series != null}
+  {:else if series != null && !patternMatches}
+    <Timeseries
+      bind:this={plot}
+      {width}
+      height={300}
+      {cluster}
+      {subCluster}
+      timestep={data.timestep}
+      scope={selectedScope}
+      metric={metricName}
+      {series}
+      {isShared}
+      resources={job.resources}
+    />
+  {:else if statsSeries[selectedScopeIndex] != null && patternMatches}
     <Timeseries
       bind:this={plot}
       {width}
@@ -113,6 +140,8 @@
       {series}
       {isShared}
       resources={job.resources}
+      statisticsSeries={statsSeries[selectedScopeIndex]}
+      useStatsSeries={!!statsSeries[selectedScopeIndex]}
     />
   {/if}
 {/key}

web/frontend/src/Status.root.svelte

@@ -315,20 +315,11 @@
 
 <!-- Loading indicator & Refresh -->
 
-<Row cols={3}>
-  <Col xs="auto" style="align-self: flex-end;">
+<Row cols={{ lg: 3, md: 3, sm: 1 }}>
+  <Col style="">
     <h4 class="mb-0">Current utilization of cluster "{cluster}"</h4>
   </Col>
-  <Col xs="auto" style="margin-left: 0.25rem;">
-    {#if $initq.fetching || $mainQuery.fetching}
-      <Spinner />
-    {:else if $initq.error}
-      <Card body color="danger">{$initq.error.message}</Card>
-    {:else}
-      <!-- ... -->
-    {/if}
-  </Col>
-  <Col xs="auto" style="margin-left: auto;">
+  <Col class="mt-2 mt-md-0 text-md-end">
     <Button
       outline
       color="secondary"
@@ -337,7 +328,7 @@
       <Icon name="bar-chart-line" /> Select Histograms
     </Button>
   </Col>
-  <Col xs="auto" style="margin-left: 0.25rem;">
+  <Col class="mt-2 mt-md-0">
     <Refresher
       initially={120}
       on:reload={() => {
@@ -347,6 +338,17 @@
     />
   </Col>
 </Row>
+<Row cols={1} class="text-center mt-3">
+  <Col>
+    {#if $initq.fetching || $mainQuery.fetching}
+      <Spinner />
+    {:else if $initq.error}
+      <Card body color="danger">{$initq.error.message}</Card>
+    {:else}
+      <!-- ... -->
+    {/if}
+  </Col>
+</Row>
 {#if $mainQuery.error}
   <Row cols={1}>
     <Col>
@@ -361,8 +363,8 @@
 
 {#if $initq.data && $mainQuery.data}
   {#each $initq.data.clusters.find((c) => c.name == cluster).subClusters as subCluster, i}
-    <Row cols={2} class="mb-3 justify-content-center">
-      <Col md="4" class="px-3">
+    <Row cols={{ lg: 2, md: 1 , sm: 1}} class="mb-3 justify-content-center">
+      <Col class="px-3">
         <Card class="h-auto mt-1">
           <CardHeader>
             <CardTitle class="mb-0">SubCluster "{subCluster.name}"</CardTitle>
@@ -433,7 +435,7 @@
           </CardBody>
         </Card>
       </Col>
-      <Col class="px-3">
+      <Col class="px-3 mt-2 mt-lg-0">
         <div bind:clientWidth={plotWidths[i]}>
           {#key $mainQuery.data.nodeMetrics}
             <Roofline
@@ -457,7 +459,7 @@
 
   <!-- Usage Stats as Histograms -->
 
-  <Row cols={4}>
+  <Row cols={{ lg: 4, md: 2, sm: 1 }}>
     <Col class="p-2">
       <div bind:clientWidth={colWidth1}>
         <h4 class="text-center">
@@ -580,7 +582,7 @@
     </Col>
   </Row>
   <hr class="my-2" />
-  <Row cols={2}>
+  <Row cols={{ lg: 2, md: 1 }}>
     <Col class="p-2">
       <div bind:clientWidth={colWidth2}>
         {#key $mainQuery.data.stats}
@@ -610,7 +612,7 @@
       {/key}
     </Col>
   </Row>
-  <Row cols={2}>
+  <Row cols={{ lg: 2, md: 1 }}>
     <Col class="p-2">
       <div bind:clientWidth={colWidth2}>
         {#key $mainQuery.data.stats}
@@ -642,15 +644,15 @@
   </Row>
   <hr class="my-2" />
   {#if metricsInHistograms}
-    <Row>
+    <Row cols={1}>
       <Col>
         {#key $mainQuery.data.stats[0].histMetrics}
           <PlotTable
             let:item
             let:width
             renderFor="user"
             items={$mainQuery.data.stats[0].histMetrics}
-            itemsPerRow={3}
+            itemsPerRow={2}
           >
             <Histogram
               data={convert2uplot(item.data)}

web/frontend/src/User.root.svelte

@@ -12,6 +12,7 @@
   } from "@sveltestrap/sveltestrap";
   import { queryStore, gql, getContextClient } from "@urql/svelte";
   import Filters from "./filters/Filters.svelte";
+  import TextFilter from "./filters/TextFilter.svelte"
   import JobList from "./joblist/JobList.svelte";
   import Sorting from "./joblist/SortSelection.svelte";
   import Refresher from "./joblist/Refresher.svelte";
@@ -132,6 +133,11 @@
     />
   </Col>
   <Col xs="auto" style="margin-left: auto;">
+    <TextFilter
+      on:update={({ detail }) => filterComponent.update(detail)}
+    />
+  </Col>
+  <Col xs="auto">
     <Refresher on:reload={() => jobList.refresh()} />
   </Col>
 </Row>

web/frontend/src/config/admin/ShowUsersRow.svelte

@@ -18,7 +18,7 @@
 
 <td>{user.username}</td>
 <td>{user.name}</td>
-<td>{user.projects}</td>
+<td style="max-width: 200px;">{user.projects}</td>
 <td>{user.email}</td>
 <td><code>{user?.roles ? user.roles.join(", ") : "No Roles"}</code></td>
 <td>

web/frontend/src/filters/Filters.svelte

@@ -193,7 +193,8 @@
       opts.push(`userMatch=${filters.userMatch}`);
     if (filters.project) opts.push(`project=${filters.project}`);
     if (filters.jobName) opts.push(`jobName=${filters.jobName}`);
-    if (filters.projectMatch != "contains")
+    if (filters.arrayJobId) opts.push(`arrayJobId=${filters.arrayJobId}`);
+    if (filters.project && filters.projectMatch != "contains")
       opts.push(`projectMatch=${filters.projectMatch}`);
 
     if (opts.length == 0 && window.location.search.length <= 1) return;