Dev

.gitignore

@@ -17,3 +17,5 @@ var/job.db-wal
 
 dist/
 *.db
+internal/repository/testdata/job.db-shm
+internal/repository/testdata/job.db-wal

configs/config-mariadb.json

@@ -0,0 +1,69 @@
+{
+  "addr": "127.0.0.1:8080",
+  "short-running-jobs-duration": 300,
+  "archive": {
+    "kind": "file",
+    "path": "./var/job-archive"
+  },
+  "jwts": {
+    "max-age": "2000h"
+  },
+  "db-driver": "mysql",
+  "db": "clustercockpit:demo@tcp(127.0.0.1:3306)/clustercockpit",
+  "enable-resampling": {
+    "trigger": 30,
+    "resolutions": [
+      600,
+      300,
+      120,
+      60
+    ]
+  },
+  "emission-constant": 317,
+  "clusters": [
+    {
+      "name": "fritz",
+      "metricDataRepository": {
+        "kind": "cc-metric-store",
+        "url": "http://localhost:8082",
+        "token": ""
+      },
+      "filterRanges": {
+        "numNodes": {
+          "from": 1,
+          "to": 64
+        },
+        "duration": {
+          "from": 0,
+          "to": 86400
+        },
+        "startTime": {
+          "from": "2022-01-01T00:00:00Z",
+          "to": null
+        }
+      }
+    },
+    {
+      "name": "alex",
+      "metricDataRepository": {
+        "kind": "cc-metric-store",
+        "url": "http://localhost:8082",
+        "token": ""
+      },
+      "filterRanges": {
+        "numNodes": {
+          "from": 1,
+          "to": 64
+        },
+        "duration": {
+          "from": 0,
+          "to": 86400
+        },
+        "startTime": {
+          "from": "2022-01-01T00:00:00Z",
+          "to": null
+        }
+      }
+    }
+  ]
+}

internal/repository/job.go

@@ -51,7 +51,7 @@ func GetJobRepository() *JobRepository {
 }
 
 var jobColumns []string = []string{
-	"job.id", "job.job_id", "job.user", "job.project", "job.cluster", "job.subcluster", "job.start_time", "job.partition", "job.array_job_id",
+	"job.id", "job.job_id", "job.hpc_user", "job.project", "job.cluster", "job.subcluster", "job.start_time", "job.cluster_partition", "job.array_job_id",
 	"job.num_nodes", "job.num_hwthreads", "job.num_acc", "job.exclusive", "job.monitoring_status", "job.smt", "job.job_state",
 	"job.duration", "job.walltime", "job.resources", "job.footprint", "job.energy",
 }
@@ -314,7 +314,7 @@ func (r *JobRepository) FindUserOrProjectOrJobname(user *schema.User, searchterm
 				return "", uresult, "", ""
 			}
 			// Find username by name (like)
-			nresult, _ := r.FindColumnValue(user, searchterm, "user", "username", "name", true)
+			nresult, _ := r.FindColumnValue(user, searchterm, "hpc_user", "username", "name", true)
 			if nresult != "" {
 				return "", nresult, "", ""
 			}
@@ -400,7 +400,7 @@ func (r *JobRepository) Partitions(cluster string) ([]string, error) {
 	start := time.Now()
 	partitions := r.cache.Get("partitions:"+cluster, func() (interface{}, time.Duration, int) {
 		parts := []string{}
-		if err = r.DB.Select(&parts, `SELECT DISTINCT job.partition FROM job WHERE job.cluster = ?;`, cluster); err != nil {
+		if err = r.DB.Select(&parts, `SELECT DISTINCT job.cluster_partition FROM job WHERE job.cluster = ?;`, cluster); err != nil {
 			return nil, 0, 1000
 		}
 

internal/repository/jobCreate.go

@@ -14,10 +14,10 @@ import (
 )
 
 const NamedJobInsert string = `INSERT INTO job (
-	job_id, user, project, cluster, subcluster, ` + "`partition`" + `, array_job_id, num_nodes, num_hwthreads, num_acc,
+	job_id, hpc_user, project, cluster, subcluster, cluster_partition, array_job_id, num_nodes, num_hwthreads, num_acc,
 	exclusive, monitoring_status, smt, job_state, start_time, duration, walltime, footprint, energy, energy_footprint, resources, meta_data
 ) VALUES (
-	:job_id, :user, :project, :cluster, :subcluster, :partition, :array_job_id, :num_nodes, :num_hwthreads, :num_acc,
+	:job_id, :hpc_user, :project, :cluster, :subcluster, :cluster_partition, :array_job_id, :num_nodes, :num_hwthreads, :num_acc,
   :exclusive, :monitoring_status, :smt, :job_state, :start_time, :duration, :walltime, :footprint,  :energy, :energy_footprint, :resources, :meta_data
 );`
 

internal/repository/jobFind.go

@@ -136,7 +136,7 @@ func (r *JobRepository) IsJobOwner(jobId int64, startTime int64, user string, cl
 	q := sq.Select("id").
 		From("job").
 		Where("job.job_id = ?", jobId).
-		Where("job.user = ?", user).
+		Where("job.hpc_user = ?", user).
 		Where("job.cluster = ?", cluster).
 		Where("job.start_time = ?", startTime)
 

internal/repository/jobQuery.go

@@ -121,13 +121,13 @@ func SecurityCheck(ctx context.Context, query sq.SelectBuilder) (sq.SelectBuilde
 		return query, nil
 	case user.HasRole(schema.RoleManager): // Manager : Add filter for managed projects' jobs only + personal jobs
 		if len(user.Projects) != 0 {
-			return query.Where(sq.Or{sq.Eq{"job.project": user.Projects}, sq.Eq{"job.user": user.Username}}), nil
+			return query.Where(sq.Or{sq.Eq{"job.project": user.Projects}, sq.Eq{"job.hpc_user": user.Username}}), nil
 		} else {
 			log.Debugf("Manager-User '%s' has no defined projects to lookup! Query only personal jobs ...", user.Username)
-			return query.Where("job.user = ?", user.Username), nil
+			return query.Where("job.hpc_user = ?", user.Username), nil
 		}
 	case user.HasRole(schema.RoleUser): // User : Only personal jobs
-		return query.Where("job.user = ?", user.Username), nil
+		return query.Where("job.hpc_user = ?", user.Username), nil
 	default: // No known Role, return error
 		var qnil sq.SelectBuilder
 		return qnil, fmt.Errorf("user has no or unknown roles")
@@ -147,7 +147,7 @@ func BuildWhereClause(filter *model.JobFilter, query sq.SelectBuilder) sq.Select
 		query = query.Where("job.array_job_id = ?", *filter.ArrayJobID)
 	}
 	if filter.User != nil {
-		query = buildStringCondition("job.user", filter.User, query)
+		query = buildStringCondition("job.hpc_user", filter.User, query)
 	}
 	if filter.Project != nil {
 		query = buildStringCondition("job.project", filter.Project, query)
@@ -159,7 +159,7 @@ func BuildWhereClause(filter *model.JobFilter, query sq.SelectBuilder) sq.Select
 		query = buildStringCondition("job.cluster", filter.Cluster, query)
 	}
 	if filter.Partition != nil {
-		query = buildStringCondition("job.partition", filter.Partition, query)
+		query = buildStringCondition("job.cluster_partition", filter.Partition, query)
 	}
 	if filter.StartTime != nil {
 		query = buildTimeCondition("job.start_time", filter.StartTime, query)

internal/repository/migration.go

@@ -114,6 +114,14 @@ func MigrateDB(backend string, db string) error {
 		return err
 	}
 
+	v, dirty, err := m.Version()
+
+	log.Infof("unsupported database version %d, need %d.\nPlease backup your database file and run cc-backend -migrate-db", v, Version)
+
+	if dirty {
+		return fmt.Errorf("last migration to version %d has failed, please fix the db manually and force version with -force-db flag", Version)
+	}
+
 	if err := m.Up(); err != nil {
 		if err == migrate.ErrNoChange {
 			log.Info("DB already up to date!")

internal/repository/migrations/mysql/08_add-footprint.down.sql

@@ -0,0 +1,83 @@
+ALTER TABLE job DROP energy;
+ALTER TABLE job DROP energy_footprint;
+ALTER TABLE job ADD COLUMN flops_any_avg;
+ALTER TABLE job ADD COLUMN mem_bw_avg;
+ALTER TABLE job ADD COLUMN mem_used_max;
+ALTER TABLE job ADD COLUMN load_avg;
+ALTER TABLE job ADD COLUMN net_bw_avg;
+ALTER TABLE job ADD COLUMN net_data_vol_total;
+ALTER TABLE job ADD COLUMN file_bw_avg;
+ALTER TABLE job ADD COLUMN file_data_vol_total;
+
+UPDATE job SET flops_any_avg = json_extract(footprint, '$.flops_any_avg');
+UPDATE job SET mem_bw_avg = json_extract(footprint, '$.mem_bw_avg');
+UPDATE job SET mem_used_max = json_extract(footprint, '$.mem_used_max');
+UPDATE job SET load_avg = json_extract(footprint, '$.cpu_load_avg');
+UPDATE job SET net_bw_avg = json_extract(footprint, '$.net_bw_avg');
+UPDATE job SET net_data_vol_total = json_extract(footprint, '$.net_data_vol_total');
+UPDATE job SET file_bw_avg = json_extract(footprint, '$.file_bw_avg');
+UPDATE job SET file_data_vol_total = json_extract(footprint, '$.file_data_vol_total');
+
+ALTER TABLE job DROP footprint;
+-- Do not use reserved keywords anymore
+RENAME TABLE hpc_user TO `user`;
+ALTER TABLE job RENAME COLUMN hpc_user TO `user`;
+ALTER TABLE job RENAME COLUMN cluster_partition TO `partition`;
+
+DROP INDEX IF EXISTS jobs_cluster;
+DROP INDEX IF EXISTS jobs_cluster_user;
+DROP INDEX IF EXISTS jobs_cluster_project;
+DROP INDEX IF EXISTS jobs_cluster_subcluster;
+DROP INDEX IF EXISTS jobs_cluster_starttime;
+DROP INDEX IF EXISTS jobs_cluster_duration;
+DROP INDEX IF EXISTS jobs_cluster_numnodes;
+
+DROP INDEX IF EXISTS jobs_cluster_partition;
+DROP INDEX IF EXISTS jobs_cluster_partition_starttime;
+DROP INDEX IF EXISTS jobs_cluster_partition_duration;
+DROP INDEX IF EXISTS jobs_cluster_partition_numnodes;
+
+DROP INDEX IF EXISTS jobs_cluster_partition_jobstate;
+DROP INDEX IF EXISTS jobs_cluster_partition_jobstate_user;
+DROP INDEX IF EXISTS jobs_cluster_partition_jobstate_project;
+DROP INDEX IF EXISTS jobs_cluster_partition_jobstate_starttime;
+DROP INDEX IF EXISTS jobs_cluster_partition_jobstate_duration;
+DROP INDEX IF EXISTS jobs_cluster_partition_jobstate_numnodes;
+
+DROP INDEX IF EXISTS jobs_cluster_jobstate;
+DROP INDEX IF EXISTS jobs_cluster_jobstate_user;
+DROP INDEX IF EXISTS jobs_cluster_jobstate_project;
+
+DROP INDEX IF EXISTS jobs_cluster_jobstate_starttime;
+DROP INDEX IF EXISTS jobs_cluster_jobstate_duration;
+DROP INDEX IF EXISTS jobs_cluster_jobstate_numnodes;
+
+DROP INDEX IF EXISTS jobs_user;
+DROP INDEX IF EXISTS jobs_user_starttime;
+DROP INDEX IF EXISTS jobs_user_duration;
+DROP INDEX IF EXISTS jobs_user_numnodes;
+
+DROP INDEX IF EXISTS jobs_project;
+DROP INDEX IF EXISTS jobs_project_user;
+DROP INDEX IF EXISTS jobs_project_starttime;
+DROP INDEX IF EXISTS jobs_project_duration;
+DROP INDEX IF EXISTS jobs_project_numnodes;
+
+DROP INDEX IF EXISTS jobs_jobstate;
+DROP INDEX IF EXISTS jobs_jobstate_user;
+DROP INDEX IF EXISTS jobs_jobstate_project;
+DROP INDEX IF EXISTS jobs_jobstate_starttime;
+DROP INDEX IF EXISTS jobs_jobstate_duration;
+DROP INDEX IF EXISTS jobs_jobstate_numnodes;
+
+DROP INDEX IF EXISTS jobs_arrayjobid_starttime;
+DROP INDEX IF EXISTS jobs_cluster_arrayjobid_starttime;
+
+DROP INDEX IF EXISTS jobs_starttime;
+DROP INDEX IF EXISTS jobs_duration;
+DROP INDEX IF EXISTS jobs_numnodes;
+
+DROP INDEX IF EXISTS jobs_duration_starttime;
+DROP INDEX IF EXISTS jobs_numnodes_starttime;
+DROP INDEX IF EXISTS jobs_numacc_starttime;
+DROP INDEX IF EXISTS jobs_energy_starttime;

internal/repository/migrations/mysql/08_add-footprint.up.sql

@@ -0,0 +1,123 @@
+DROP INDEX IF EXISTS job_stats ON job;
+DROP INDEX IF EXISTS job_by_user ON job;
+DROP INDEX IF EXISTS job_by_starttime ON job;
+DROP INDEX IF EXISTS job_by_job_id ON job;
+DROP INDEX IF EXISTS job_list ON job;
+DROP INDEX IF EXISTS job_list_user ON job;
+DROP INDEX IF EXISTS job_list_users ON job;
+DROP INDEX IF EXISTS job_list_users_start ON job;
+
+ALTER TABLE job ADD COLUMN energy REAL NOT NULL DEFAULT 0.0;
+ALTER TABLE job ADD COLUMN energy_footprint JSON;
+
+ALTER TABLE job ADD COLUMN footprint JSON;
+ALTER TABLE tag ADD COLUMN tag_scope TEXT NOT NULL DEFAULT 'global';
+
+-- Do not use reserved keywords anymore
+RENAME TABLE `user` TO hpc_user;
+ALTER TABLE job RENAME COLUMN `user` TO hpc_user;
+ALTER TABLE job RENAME COLUMN `partition` TO cluster_partition;
+
+ALTER TABLE job MODIFY COLUMN cluster VARCHAR(50);
+ALTER TABLE job MODIFY COLUMN hpc_user VARCHAR(50);
+ALTER TABLE job MODIFY COLUMN subcluster VARCHAR(50);
+ALTER TABLE job MODIFY COLUMN project VARCHAR(50);
+ALTER TABLE job MODIFY COLUMN cluster_partition VARCHAR(50);
+ALTER TABLE job MODIFY COLUMN job_state VARCHAR(25);
+
+UPDATE job SET footprint = '{"flops_any_avg": 0.0}';
+UPDATE job SET footprint = json_replace(footprint, '$.flops_any_avg', job.flops_any_avg);
+UPDATE job SET footprint = json_insert(footprint, '$.mem_bw_avg', job.mem_bw_avg);
+UPDATE job SET footprint = json_insert(footprint, '$.mem_used_max', job.mem_used_max);
+UPDATE job SET footprint = json_insert(footprint, '$.cpu_load_avg', job.load_avg);
+UPDATE job SET footprint = json_insert(footprint, '$.net_bw_avg', job.net_bw_avg) WHERE job.net_bw_avg != 0;
+UPDATE job SET footprint = json_insert(footprint, '$.net_data_vol_total', job.net_data_vol_total) WHERE job.net_data_vol_total != 0;
+UPDATE job SET footprint = json_insert(footprint, '$.file_bw_avg', job.file_bw_avg) WHERE job.file_bw_avg != 0;
+UPDATE job SET footprint = json_insert(footprint, '$.file_data_vol_total', job.file_data_vol_total) WHERE job.file_data_vol_total != 0;
+
+ALTER TABLE job DROP flops_any_avg;
+ALTER TABLE job DROP mem_bw_avg;
+ALTER TABLE job DROP mem_used_max;
+ALTER TABLE job DROP load_avg;
+ALTER TABLE job DROP net_bw_avg;
+ALTER TABLE job DROP net_data_vol_total;
+ALTER TABLE job DROP file_bw_avg;
+ALTER TABLE job DROP file_data_vol_total;
+
+-- Indices for: Single filters, combined filters, sorting, sorting with filters
+-- Cluster Filter
+CREATE INDEX IF NOT EXISTS jobs_cluster ON job (cluster);
+CREATE INDEX IF NOT EXISTS jobs_cluster_user ON job (cluster, hpc_user);
+CREATE INDEX IF NOT EXISTS jobs_cluster_project ON job (cluster, project);
+CREATE INDEX IF NOT EXISTS jobs_cluster_subcluster ON job (cluster, subcluster);
+-- Cluster Filter Sorting
+CREATE INDEX IF NOT EXISTS jobs_cluster_starttime ON job (cluster, start_time);
+CREATE INDEX IF NOT EXISTS jobs_cluster_duration ON job (cluster, duration);
+CREATE INDEX IF NOT EXISTS jobs_cluster_numnodes ON job (cluster, num_nodes);
+
+-- Cluster+Partition Filter
+CREATE INDEX IF NOT EXISTS jobs_cluster_partition ON job (cluster, cluster_partition);
+-- Cluster+Partition Filter Sorting
+CREATE INDEX IF NOT EXISTS jobs_cluster_partition_starttime ON job (cluster, cluster_partition, start_time);
+CREATE INDEX IF NOT EXISTS jobs_cluster_partition_duration ON job (cluster, cluster_partition, duration);
+CREATE INDEX IF NOT EXISTS jobs_cluster_partition_numnodes ON job (cluster, cluster_partition, num_nodes);
+
+-- Cluster+Partition+Jobstate Filter
+CREATE INDEX IF NOT EXISTS jobs_cluster_partition_jobstate ON job (cluster, cluster_partition, job_state);
+CREATE INDEX IF NOT EXISTS jobs_cluster_partition_jobstate_user ON job (cluster, cluster_partition, job_state, hpc_user);
+CREATE INDEX IF NOT EXISTS jobs_cluster_partition_jobstate_project ON job (cluster, cluster_partition, job_state, project);
+-- Cluster+Partition+Jobstate Filter Sorting
+CREATE INDEX IF NOT EXISTS jobs_cluster_partition_jobstate_starttime ON job (cluster, cluster_partition, job_state, start_time);
+CREATE INDEX IF NOT EXISTS jobs_cluster_partition_jobstate_duration ON job (cluster, cluster_partition, job_state, duration);
+CREATE INDEX IF NOT EXISTS jobs_cluster_partition_jobstate_numnodes ON job (cluster, cluster_partition, job_state, num_nodes);
+
+-- Cluster+JobState Filter
+CREATE INDEX IF NOT EXISTS jobs_cluster_jobstate ON job (cluster, job_state);
+CREATE INDEX IF NOT EXISTS jobs_cluster_jobstate_user ON job (cluster, job_state, hpc_user);
+CREATE INDEX IF NOT EXISTS jobs_cluster_jobstate_project ON job (cluster, job_state, project);
+-- Cluster+JobState Filter Sorting
+CREATE INDEX IF NOT EXISTS jobs_cluster_jobstate_starttime ON job (cluster, job_state, start_time);
+CREATE INDEX IF NOT EXISTS jobs_cluster_jobstate_duration ON job (cluster, job_state, duration);
+CREATE INDEX IF NOT EXISTS jobs_cluster_jobstate_numnodes ON job (cluster, job_state, num_nodes);
+
+-- User Filter
+CREATE INDEX IF NOT EXISTS jobs_user ON job (hpc_user);
+-- User Filter Sorting
+CREATE INDEX IF NOT EXISTS jobs_user_starttime ON job (hpc_user, start_time);
+CREATE INDEX IF NOT EXISTS jobs_user_duration ON job (hpc_user, duration);
+CREATE INDEX IF NOT EXISTS jobs_user_numnodes ON job (hpc_user, num_nodes);
+
+-- Project Filter
+CREATE INDEX IF NOT EXISTS jobs_project ON job (project);
+CREATE INDEX IF NOT EXISTS jobs_project_user ON job (project, hpc_user);
+-- Project Filter Sorting
+CREATE INDEX IF NOT EXISTS jobs_project_starttime ON job (project, start_time);
+CREATE INDEX IF NOT EXISTS jobs_project_duration ON job (project, duration);
+CREATE INDEX IF NOT EXISTS jobs_project_numnodes ON job (project, num_nodes);
+
+-- JobState Filter
+CREATE INDEX IF NOT EXISTS jobs_jobstate ON job (job_state);
+CREATE INDEX IF NOT EXISTS jobs_jobstate_user ON job (job_state, hpc_user);
+CREATE INDEX IF NOT EXISTS jobs_jobstate_project ON job (job_state, project);
+CREATE INDEX IF NOT EXISTS jobs_jobstate_cluster ON job (job_state, cluster);
+-- JobState Filter Sorting
+CREATE INDEX IF NOT EXISTS jobs_jobstate_starttime ON job (job_state, start_time);
+CREATE INDEX IF NOT EXISTS jobs_jobstate_duration ON job (job_state, duration);
+CREATE INDEX IF NOT EXISTS jobs_jobstate_numnodes ON job (job_state, num_nodes);
+
+-- ArrayJob Filter
+CREATE INDEX IF NOT EXISTS jobs_arrayjobid_starttime ON job (array_job_id, start_time);
+CREATE INDEX IF NOT EXISTS jobs_cluster_arrayjobid_starttime ON job (cluster, array_job_id, start_time);
+
+-- Sorting without active filters
+CREATE INDEX IF NOT EXISTS jobs_starttime ON job (start_time);
+CREATE INDEX IF NOT EXISTS jobs_duration ON job (duration);
+CREATE INDEX IF NOT EXISTS jobs_numnodes ON job (num_nodes);
+
+-- Single filters with default starttime sorting
+CREATE INDEX IF NOT EXISTS jobs_duration_starttime ON job (duration, start_time);
+CREATE INDEX IF NOT EXISTS jobs_numnodes_starttime ON job (num_nodes, start_time);
+CREATE INDEX IF NOT EXISTS jobs_numacc_starttime ON job (num_acc, start_time);
+CREATE INDEX IF NOT EXISTS jobs_energy_starttime ON job (energy, start_time);
+
+-- Optimize DB index usage