oomd - Userspace Out-Of-Memory (OOM) killer for Linux systems
oomd leverages PSI and cgroupv2 to monitor a system
holistically. oomd then takes corrective action in userspace before
an OOM occurs in kernel space. Corrective action is configured via a
flexible plugin system, in which custom code can be written. By default,
this involves killing offending processes. This enables an unparalleled
level of flexibility where each workload can have custom protection rules.
Furthermore, time spent livedlocked in kernelspace is minimized.
- --help, -h
- Show this help message and exit
- --version,
-v
- Print version and exit
- --config, -C
CONFIG
- Config file (default: /etc/oomd/oomd.json)
- --interval, -i
INTERVAL
- Event loop polling interval (default: 5)
- --cgroup-fs, -f
FS
- Cgroup2 filesystem mount point (default: /sys/fs/cgroup)
- --check-config, -c
CONFIG
- Check config file (default: /etc/oomd/oomd.json)
- --list-plugins,
-l
- List all available plugins
- --drop-in-dir, -w
DIR
- Directory to watch for drop in configs
- --socket-path, -s
PATH
- Specify stats socket path (default: /run/oomd/oomd-stats.socket)
- --dump-stats,
-d
- Dump accumulated stats
- --reset-stats,
-r
- Reset stats collection
- --device
DEVS
- Comma separated <major>:<minor> pairs for IO
cost calculation (default: none)
- --ssd-coeffs
COEFFS
- Comma separated values for SSD IO cost calculation (default: see doc)
- --hdd-coeffs
COEFFS
- Comma separated values for HDD IO cost calculation (default: see doc)
https://github.com/facebookincubator/oomd#configuration
oomd is written by
Facebook, Inc.
This manual page was written by
Yangfl for the Debian
Project (and may be used by others).