Merge branch 'dev' into optimize-checkpoint-loading

ReleaseNotes.md

@@ -89,7 +89,7 @@ For release specific notes visit the [ClusterCockpit Documentation](https://clus
 - **Job tagger option**: Enable automatic job tagging via configuration flag
 - **Application detection**: Automatic detection of applications (MATLAB, GROMACS, etc.)
 - **Job classification**: Automatic detection of pathological jobs
-- **omitTagged flag**: Option to exclude tagged jobs from retention/cleanup operations
+- **omit-tagged**: Option to exclude tagged jobs from retention/cleanup operations (`none`, `all`, or `user`)
 - **Admin UI trigger**: Taggers can be run on-demand from the admin web interface
   without restarting the backend
 

api/swagger.json

@@ -389,8 +389,71 @@
                 ]
             }
         },
+        "/api/jobs/edit_meta/": {
+            "patch": {
+                "security": [
+                    {
+                        "ApiKeyAuth": []
+                    }
+                ],
+                "description": "Edit key value pairs in metadata json of job specified by jobID, StartTime and Cluster\nIf a key already exists its content will be overwritten",
+                "consumes": [
+                    "application/json"
+                ],
+                "produces": [
+                    "application/json"
+                ],
+                "tags": [
+                    "Job add and modify"
+                ],
+                "summary": "Edit meta-data json by request",
+                "parameters": [
+                    {
+                        "description": "Specifies job and payload to add or update",
+                        "name": "request",
+                        "in": "body",
+                        "required": true,
+                        "schema": {
+                            "$ref": "#/definitions/api.JobMetaRequest"
+                        }
+                    }
+                ],
+                "responses": {
+                    "200": {
+                        "description": "Updated job resource",
+                        "schema": {
+                            "$ref": "#/definitions/schema.Job"
+                        }
+                    },
+                    "400": {
+                        "description": "Bad Request",
+                        "schema": {
+                            "$ref": "#/definitions/api.ErrorResponse"
+                        }
+                    },
+                    "401": {
+                        "description": "Unauthorized",
+                        "schema": {
+                            "$ref": "#/definitions/api.ErrorResponse"
+                        }
+                    },
+                    "404": {
+                        "description": "Job does not exist",
+                        "schema": {
+                            "$ref": "#/definitions/api.ErrorResponse"
+                        }
+                    },
+                    "500": {
+                        "description": "Internal Server Error",
+                        "schema": {
+                            "$ref": "#/definitions/api.ErrorResponse"
+                        }
+                    }
+                }
+            }
+        },
         "/api/jobs/edit_meta/{id}": {
-            "post": {
+            "patch": {
                 "description": "Edit key value pairs in job metadata json\nIf a key already exists its content will be overwritten",
                 "consumes": [
                     "application/json"

api/swagger.yaml

@@ -102,6 +102,27 @@ definitions:
         description: Page id returned
         type: integer
     type: object
+  api.JobMetaRequest:
+    properties:
+      cluster:
+        description: Cluster of job
+        example: fritz
+        type: string
+      jobId:
+        description: Cluster Job ID of job
+        example: 123000
+        type: integer
+      payload:
+        allOf:
+        - $ref: '#/definitions/api.EditMetaRequest'
+        description: Content to Add to Job Meta_Data
+      startTime:
+        description: Start Time of job as epoch
+        example: 1649723812
+        type: integer
+    required:
+    - jobId
+    type: object
   api.JobMetricWithName:
     properties:
       metric:
@@ -1091,8 +1112,50 @@ paths:
       summary: Remove a job from the sql database
       tags:
       - Job remove
+  /api/jobs/edit_meta/:
+    patch:
+      consumes:
+      - application/json
+      description: |-
+        Edit key value pairs in metadata json of job specified by jobID, StartTime and Cluster
+        If a key already exists its content will be overwritten
+      parameters:
+      - description: Specifies job and payload to add or update
+        in: body
+        name: request
+        required: true
+        schema:
+          $ref: '#/definitions/api.JobMetaRequest'
+      produces:
+      - application/json
+      responses:
+        "200":
+          description: Updated job resource
+          schema:
+            $ref: '#/definitions/schema.Job'
+        "400":
+          description: Bad Request
+          schema:
+            $ref: '#/definitions/api.ErrorResponse'
+        "401":
+          description: Unauthorized
+          schema:
+            $ref: '#/definitions/api.ErrorResponse'
+        "404":
+          description: Job does not exist
+          schema:
+            $ref: '#/definitions/api.ErrorResponse'
+        "500":
+          description: Internal Server Error
+          schema:
+            $ref: '#/definitions/api.ErrorResponse'
+      security:
+      - ApiKeyAuth: []
+      summary: Edit meta-data json by request
+      tags:
+      - Job add and modify
   /api/jobs/edit_meta/{id}:
-    post:
+    patch:
       consumes:
       - application/json
       description: |-

internal/api/docs.go

@@ -396,8 +396,71 @@ const docTemplate = `{
                 ]
             }
         },
+        "/api/jobs/edit_meta/": {
+            "patch": {
+                "security": [
+                    {
+                        "ApiKeyAuth": []
+                    }
+                ],
+                "description": "Edit key value pairs in metadata json of job specified by jobID, StartTime and Cluster\nIf a key already exists its content will be overwritten",
+                "consumes": [
+                    "application/json"
+                ],
+                "produces": [
+                    "application/json"
+                ],
+                "tags": [
+                    "Job add and modify"
+                ],
+                "summary": "Edit meta-data json by request",
+                "parameters": [
+                    {
+                        "description": "Specifies job and payload to add or update",
+                        "name": "request",
+                        "in": "body",
+                        "required": true,
+                        "schema": {
+                            "$ref": "#/definitions/api.JobMetaRequest"
+                        }
+                    }
+                ],
+                "responses": {
+                    "200": {
+                        "description": "Updated job resource",
+                        "schema": {
+                            "$ref": "#/definitions/schema.Job"
+                        }
+                    },
+                    "400": {
+                        "description": "Bad Request",
+                        "schema": {
+                            "$ref": "#/definitions/api.ErrorResponse"
+                        }
+                    },
+                    "401": {
+                        "description": "Unauthorized",
+                        "schema": {
+                            "$ref": "#/definitions/api.ErrorResponse"
+                        }
+                    },
+                    "404": {
+                        "description": "Job does not exist",
+                        "schema": {
+                            "$ref": "#/definitions/api.ErrorResponse"
+                        }
+                    },
+                    "500": {
+                        "description": "Internal Server Error",
+                        "schema": {
+                            "$ref": "#/definitions/api.ErrorResponse"
+                        }
+                    }
+                }
+            }
+        },
         "/api/jobs/edit_meta/{id}": {
-            "post": {
+            "patch": {
                 "description": "Edit key value pairs in job metadata json\nIf a key already exists its content will be overwritten",
                 "consumes": [
                     "application/json"

internal/api/job.go

@@ -72,6 +72,14 @@ type EditMetaRequest struct {
 	Value string `json:"value" example:"bash script"`
 }
 
+// JobMetaRequest model
+type JobMetaRequest struct {
+	JobId     *int64          `json:"jobId" validate:"required" example:"123000"` // Cluster Job ID of job
+	Cluster   *string         `json:"cluster" example:"fritz"`                    // Cluster of job
+	StartTime *int64          `json:"startTime" example:"1649723812"`             // Start Time of job as epoch
+	Payload   EditMetaRequest `json:"payload"`                                    // Content to Add to Job Meta_Data
+}
+
 type TagJobAPIRequest []*APITag
 
 type GetJobAPIRequest []string
@@ -423,21 +431,21 @@ func (api *RestAPI) getJobByID(rw http.ResponseWriter, r *http.Request) {
 }
 
 // editMeta godoc
-// @summary    Edit meta-data json
+// @summary    Edit meta-data json of job identified by database id
 // @tags Job add and modify
-// @description Edit key value pairs in job metadata json
+// @description Edit key value pairs in job metadata json of job specified by database id
 // @description If a key already exists its content will be overwritten
 // @accept      json
 // @produce     json
 // @param       id      path     int                  true "Job Database ID"
-// @param       request body     api.EditMetaRequest  true "Kay value pair to add"
+// @param       request body     api.EditMetaRequest  true "Metadata Key value pair to add or update"
 // @success     200     {object} schema.Job                "Updated job resource"
 // @failure     400     {object} api.ErrorResponse         "Bad Request"
 // @failure     401     {object} api.ErrorResponse         "Unauthorized"
 // @failure     404     {object} api.ErrorResponse         "Job does not exist"
 // @failure     500     {object} api.ErrorResponse         "Internal Server Error"
 // @security    ApiKeyAuth
-// @router      /api/jobs/edit_meta/{id} [post]
+// @router      /api/jobs/edit_meta/{id} [patch]
 func (api *RestAPI) editMeta(rw http.ResponseWriter, r *http.Request) {
 	id, err := strconv.ParseInt(chi.URLParam(r, "id"), 10, 64)
 	if err != nil {
@@ -469,6 +477,54 @@ func (api *RestAPI) editMeta(rw http.ResponseWriter, r *http.Request) {
 	}
 }
 
+// editMetaByRequest godoc
+// @summary    Edit meta-data json of job identified by request
+// @tags Job add and modify
+// @description Edit key value pairs in metadata json of job specified by jobID, StartTime and Cluster
+// @description If a key already exists its content will be overwritten
+// @accept      json
+// @produce     json
+// @param       request body     api.JobMetaRequest   true "Specifies job and payload to add or update"
+// @success     200     {object} schema.Job                "Updated job resource"
+// @failure     400     {object} api.ErrorResponse         "Bad Request"
+// @failure     401     {object} api.ErrorResponse         "Unauthorized"
+// @failure     404     {object} api.ErrorResponse         "Job does not exist"
+// @failure     500     {object} api.ErrorResponse         "Internal Server Error"
+// @security    ApiKeyAuth
+// @router      /api/jobs/edit_meta/ [patch]
+func (api *RestAPI) editMetaByRequest(rw http.ResponseWriter, r *http.Request) {
+	// Parse request body
+	req := JobMetaRequest{}
+	if err := decode(r.Body, &req); err != nil {
+		handleError(fmt.Errorf("parsing request body failed: %w", err), http.StatusBadRequest, rw)
+		return
+	}
+
+	// Fetch job (that will have its meta_data edited) from db
+	var job *schema.Job
+	var err error
+	if req.JobId == nil {
+		handleError(errors.New("the field 'jobId' is required"), http.StatusBadRequest, rw)
+		return
+	}
+
+	// log.Printf("loading db job for editMetaByRequest... : JobMetaRequest=%v", req)
+	job, err = api.JobRepository.Find(req.JobId, req.Cluster, req.StartTime)
+	if err != nil {
+		handleError(fmt.Errorf("finding job failed: %w", err), http.StatusUnprocessableEntity, rw)
+		return
+	}
+
+	if err := api.JobRepository.UpdateMetadata(job, req.Payload.Key, req.Payload.Value); err != nil {
+		http.Error(rw, err.Error(), http.StatusInternalServerError)
+		return
+	}
+
+	rw.Header().Add("Content-Type", "application/json")
+	rw.WriteHeader(http.StatusOK)
+	json.NewEncoder(rw).Encode(job)
+}
+
 // tagJob godoc
 // @summary     Adds one or more tags to a job
 // @tags Job add and modify

internal/api/rest.go

@@ -96,8 +96,8 @@ func (api *RestAPI) MountAPIRoutes(r chi.Router) {
 	r.Post("/jobs/tag_job/{id}", api.tagJob)
 	r.Patch("/jobs/tag_job/{id}", api.tagJob)
 	r.Delete("/jobs/tag_job/{id}", api.removeTagJob)
-	r.Post("/jobs/edit_meta/{id}", api.editMeta)
 	r.Patch("/jobs/edit_meta/{id}", api.editMeta)
+	r.Patch("/jobs/edit_meta/", api.editMetaByRequest)
 	r.Get("/jobs/metrics/{id}", api.getJobMetrics)
 	r.Delete("/jobs/delete_job/", api.deleteJobByRequest)
 	r.Delete("/jobs/delete_job/{id}", api.deleteJobByID)

web/frontend/src/status/dashdetails/HealthDash.svelte

@@ -32,15 +32,6 @@
 
   /* Const Init */
   const client = getContextClient();
-  const stateOptions = [
-    "all",
-    "allocated",
-    "idle",
-    "down",
-    "mixed",
-    "reserved",
-    "unknown",
-  ];
   const healthOptions = [
     "all",
     "full",
@@ -52,12 +43,10 @@
   let pieWidth = $state(0);
   let querySorting = $state({ field: "startTime", type: "col", order: "DESC" })
   let tableHostFilter = $state("");
-  let tableStateFilter = $state(stateOptions[0]);
   let tableHealthFilter = $state(healthOptions[0]);
   let healthTableSorting = $state(
     {
-      schedulerState: { dir: "down", active: true },
+      healthState:  { dir: "up", active: true },
-      healthState:  { dir: "down", active: false },
       hostname:  { dir: "down", active: false },
     }
   );
@@ -79,9 +68,7 @@
             hostname
             cluster
             subCluster
-            schedulerState
             healthState
-            metaData
             healthData
           }
         }
@@ -102,7 +89,7 @@
   let healthTableData = $derived.by(() => {
     if ($statusQuery?.data) {
       return [...$statusQuery.data.nodes.items].sort((n1, n2) => {
-        return n1['schedulerState'].localeCompare(n2['schedulerState'])
+        return n1['healthState'].localeCompare(n2['healthState'])
       });
     } else {
       return [];
@@ -114,21 +101,12 @@
     if (tableHostFilter != "") {
           pendingTableData = pendingTableData.filter((e) => e.hostname.includes(tableHostFilter))
     }
-    if (tableStateFilter != "all") {
-          pendingTableData = pendingTableData.filter((e) => e.schedulerState.includes(tableStateFilter))
-    }
     if (tableHealthFilter != "all") {
           pendingTableData = pendingTableData.filter((e) => e.healthState.includes(tableHealthFilter))
     }
     return pendingTableData
   });
 
-  const refinedStateData = $derived.by(() => {
-    return $statusQuery?.data?.nodeStates.
-      filter((e) => ['allocated', 'reserved', 'idle', 'mixed','down', 'unknown'].includes(e.state)).
-      sort((a, b) => b.count - a.count)
-  });
-
   const refinedHealthData = $derived.by(() => {
     return $statusQuery?.data?.nodeStates.
       filter((e) => ['full', 'partial', 'failed'].includes(e.state)).
@@ -296,7 +274,7 @@
           <thead>
             <!-- Header Row 1: Titles and Sorting -->
             <tr>
-              <th style="width: 9%; min-width: 100px; max-width:10%;" onclick={() => sortBy('hostname')}>
+              <th style="width: 10%; min-width: 100px; max-width:12%;" onclick={() => sortBy('hostname')}>
                 Hosts ({filteredTableData.length})
                 <Icon
                   name="caret-{healthTableSorting['hostname'].dir}{healthTableSorting['hostname']
@@ -305,16 +283,7 @@
                     : ''}"
                 />
               </th>
-              <th style="width: 9%; min-width: 100px; max-width:10%;" onclick={() => sortBy('schedulerState')}>
+              <th style="width: 10%; min-width: 100px; max-width:12%;" onclick={() => sortBy('healthState')}>
-                Scheduler State
-                <Icon
-                  name="caret-{healthTableSorting['schedulerState'].dir}{healthTableSorting['schedulerState']
-                    .active
-                    ? '-fill'
-                    : ''}"
-                />
-              </th>
-              <th style="width: 9%; min-width: 100px; max-width:10%;" onclick={() => sortBy('healthState')}>
                 Health State
                 <Icon
                   name="caret-{healthTableSorting['healthState'].dir}{healthTableSorting['healthState']
@@ -324,7 +293,6 @@
                 />
               </th>
               <th>Metric Availability</th>
-              <th>Meta Information</th>
             </tr>
             <!-- Header Row 2: Filters -->
             <tr>
@@ -337,53 +305,27 @@
                 </InputGroup>
               </th>
               <th>
-                <InputGroup size="sm">
+                  <Input size="sm" type="select" bind:value={tableHealthFilter}>
-                  <Input type="select" bind:value={tableStateFilter}>
-                    {#each stateOptions as so}
-                      <option value={so}>{so}</option>
-                    {/each}
-                  </Input>
-                  <InputGroupText>
-                    <Icon name="search"></Icon>
-                  </InputGroupText>
-                </InputGroup>
-              </th>
-              <th>
-                <InputGroup size="sm">
-                  <Input type="select" bind:value={tableHealthFilter}>
                     {#each healthOptions as ho}
                       <option value={ho}>{ho}</option>
                     {/each}
                   </Input>
-                  <InputGroupText>
-                    <Icon name="search"></Icon>
-                  </InputGroupText>
-                </InputGroup>
               </th>
               <th></th>
-              <th></th>
             </tr>
           </thead>
           <tbody>
             {#each filteredTableData as host (host.hostname)}
               <tr>
                 <th scope="row"><b><a href="/monitoring/node/{cluster}/{host.hostname}" target="_blank">{host.hostname}</a></b></th>
-                <td>{host.schedulerState}</td>
                 <td>{host.healthState}</td>
-                <td style="max-width: 250px;">
+                <td style="max-width: 76%;">
                   {#each Object.keys(host.healthData) as hkey}
                     <p>
                       <b>{hkey}</b>: {host.healthData[hkey]}
                     </p>
                   {/each}
                 </td>
-                <td style="max-width: 250px;">
-                  {#each Object.keys(host.metaData) as mkey}
-                    <p>
-                      <b>{mkey}</b>: {host.metaData[mkey]}
-                    </p>
-                  {/each}
-                </td>
               </tr>
             {/each}
           </tbody>