From sysadmin-homelab
Sweep systemd unit state on a managed host — active, failed, flapping.
npx claudepluginhub danielrosehill/claude-code-plugins --plugin sysadmin-homelab[--host=<ssh-alias>] [unit-glob]# Sysadmin-Homelab — Check Services Inspect systemd service state. Prefix with `ssh <host>` for remote. ## Output - **Failed** — units in failed state, with last error from `journalctl -u <unit> -n 20`. - **Flapping** — units restarted more than twice in the last hour. - **Healthy** — count of active units. - **Recommended actions** — restart / mask / investigate per unit.
Share bugs, ideas, or general feedback.
Inspect systemd service state. Prefix with ssh <host> for remote.
# Failed units
systemctl --failed --no-pager
# All units matching glob (if provided)
GLOB="${ARGUMENTS:-*}"
systemctl list-units --type=service --all "$GLOB" --no-pager | head -60
# Recently restarted units (flapping detection)
journalctl --since "1 hour ago" | grep -E "Started|Stopped|Failed" | tail -40
# User units too
systemctl --user --failed --no-pager 2>/dev/null || true
journalctl -u <unit> -n 20.