LinuxCommandLibrary

slurmdbd

Slurm database accounting daemon

TLDR

Run in foreground mode with logging to stdout

$ slurmdbd -D
copy
Set the daemon's nice value
$ slurmdbd -n [value]
copy
Change working directory to LogFile path
$ slurmdbd -s
copy
Run with verbose output
$ slurmdbd -v
copy
Display version
$ slurmdbd -V
copy
Display help
$ slurmdbd -h
copy

SYNOPSIS

slurmdbd [OPTIONS...]

DESCRIPTION

slurmdbd (Slurm Database Daemon) provides a secure enterprise-wide database interface for Slurm workload manager. It is essential for archiving accounting records, tracking job history, usage statistics, and user/account associations across a cluster.
The daemon handles database connections, manages accounting data storage, and provides fair-share scheduling information to slurmctld. It typically runs on a dedicated host with access to the accounting database (MySQL/MariaDB).

PARAMETERS

-D

Run daemon in foreground with logging to stdout
-h
Display help and brief summary of options
-n value
Set daemon nice value (typically negative)
-s
Change working directory to LogFile path or /var/tmp
-u
Display database version and conversion status, then exit
-v
Verbose operation (up to 6 levels with multiple v's)
-V
Print version information and exit

CONFIGURATION

/etc/slurm/slurmdbd.conf

Database daemon configuration specifying database connection parameters, authentication settings, and logging options.
/etc/slurm/slurm.conf
Main Slurm configuration referenced for cluster-wide settings and AccountingStorageHost.

SIGNALS

SIGTERM, SIGINT, SIGQUIT

Graceful shutdown with rollup completion
SIGHUP
Reload configuration files
SIGUSR2
Reread log level and reopen log file (useful for logrotate)

CAVEATS

Requires proper configuration in slurmdbd.conf before starting. Database must be accessible and properly configured. Should run on a host separate from compute nodes for reliability. Uses significant memory for large clusters with extensive job history.

HISTORY

slurmdbd is part of Slurm (Simple Linux Utility for Resource Management), developed at Lawrence Livermore National Laboratory starting in 2002. The database daemon was added to support enterprise-level accounting and fair-share scheduling. Slurm is now maintained by SchedMD and is used on many of the world's largest supercomputers.

SEE ALSO

slurmctld(8), slurmd(8), sacct(1), sacctmgr(1)

> TERMINAL_GEAR

Curated for the Linux community

Copied to clipboard

> TERMINAL_GEAR

Curated for the Linux community