sacctmgr_user

Overview

Collects user account and association data from SLURM using sacctmgr and publishes it at regular intervals. Provides daily snapshots of user account associations, default accounts and QoS per user, account-specific QoS access, user permissions and access control, and account hierarchy.

Data Type: DataType.LOG, Schema: SacctmgrUserPayload

Execution Scope

Single node in the cluster.

Output Schema

SacctmgrUserPayload

Published with DataType.LOG:

{
    "ds": str,                    # Collection date (YYYY-MM-DD in Pacific time)
    "cluster": str,               # Cluster identifier
    "derived_cluster": str,       # Sub-cluster (same as cluster if not `--heterogeneous-cluster-v1`)
    "sacctmgr_user": {            # Dictionary of user attributes
        "User": str,              # Username
        "DefaultAccount": str,    # User's default account
        "Account": str,           # Account for this association
        "DefaultQOS": str,        # User's default QoS
        "QOS": str,               # Comma-separated list of available QoS
        # Additional fields depending on SLURM version
    }
}

Important Notes:

1. Each user-account association creates a separate record
2. A user with 3 accounts will generate 3 separate payloads
3. The QOS field is a comma-separated string of available QoS levels
4. Fields depend on SLURM version; collector dynamically parses headers

Data Collection Commands

The collector executes:

Stage 1: Get User List

sacctmgr show user format=User -nP

Stage 2: Get User Details (per user)

sacctmgr show user <username> withassoc format=User,DefaultAccount,Account,DefaultQOS,QOS -P

Command-Line Options

Option	Type	Default	Description
--cluster	String	Auto-detected	Cluster name for metadata enrichment
--sink	String	Required	Sink destination, see Exporters
--sink-opts	Multiple	-	Sink-specific options
--log-level	Choice	INFO	DEBUG, INFO, WARNING, ERROR, CRITICAL
--log-folder	String	/var/log/fb-monitoring	Log directory
--stdout	Flag	False	Display metrics to stdout in addition to logs
--heterogeneous-cluster-v1	Flag	False	Enable per-partition metrics for heterogeneous clusters
--interval	Integer	300	Seconds between collection cycles (5 minutes)
--once	Flag	False	Run once and exit (no continuous monitoring)
--retries	Integer	Shared default	Retry attempts on sink failures
--dry-run	Flag	False	Print to stdout instead of publishing to sink
--chunk-size	Integer	Shared default	The maximum size in bytes of each chunk when writing data to sink.

Usage Examples

Basic Daily Collection

gcm sacctmgr_user --sink otel --sink-opts "log_resource_attributes={'attr_1': 'value1'}"

One-Time Snapshot

gcm sacctmgr_user --once --sink stdout

Debug Mode with Local File Output

gcm sacctmgr_user \
  --once \
  --log-level DEBUG \
  --stdout \
  --sink file --sink-opts filepath=/tmp/user_data.jsonl