OmniRoute/docs/ops/RUNNER_BOX.md

title

title
Self-Hosted Runner Box Operations

Self-Hosted Runner Box Operations (.113 pool)

The self-hosted pool (self-hosted, omni-release on all eight runners; omni-build on two) runs on the .113 box. Measured 2026-08-28 (v3.8.50 postmortem, Parte III):

resource	value	what it means for scheduling
RAM / CPU	31 GB / 32 cores (was 16 GB when this doc was first written)	one next-build peaks at ~14 GB → 2 concurrent heavy builds saturate the box, 3 take it down (2026-08-28 06:42Z: load 56, two jobs lost)
swap	15 GB	it swapped its way through the v3.8.50 publish; pressure shows in /proc/pressure/memory
/tmp	12 GB tmpfs = RAM	anything parked there is memory; leftovers are swept after 3 h
disk	188 GB	_work checkouts of 8 runners reach ~70 GB with no cap
runners	10 listeners: 8 OmniRoute + OmniHeuris + OmniMind	all share the memory above

Install the janitor (one-time, on the box)

scp scripts/ops/runner-janitor.sh root@192.168.0.113:/opt/omniroute-ops/runner-janitor.sh
ssh root@192.168.0.113 'chmod +x /opt/omniroute-ops/runner-janitor.sh; apt-get install -y lsof'
# cron (root): every 30 min, log to /var/log/runner-janitor.log
*/30 * * * * MAX_ACTIVE_RUNNERS=8 /opt/omniroute-ops/runner-janitor.sh >> /var/log/runner-janitor.log 2>&1

lsof is required: the janitor proves a path is idle with one snapshot of open files before removing it, and without the tool it removes nothing and says so (exit 1). Try any change with --dry-run first — it prints exactly what it would do and touches nothing.

What it does every run: sweeps our own leftovers (runner-*, omniroute-*, next-build*, e2e-build.tar.gz) after 3 h on tmpfs and 24 h on disk _work/_temp; kills a next-build older than 75 min (no job runs that long — on 2026-08-27 one ran 70 min after GitHub had declared its job lost); prunes 48 h-old checkouts of runners whose unit is stopped; alerts on disk ≥ 85 %, memory PSI full/avg60 ≥ 10 %, and more listeners than MAX_ACTIVE_RUNNERS (with an omniroute/other breakdown). Exit 1 = attention needed; read the log.

Runner units: KillMode

The runner's default KillMode=process leaves Runner.Worker → npm → next-build alive when a unit is stopped or restarted — an orphan build keeps eating RAM and CPU with no job attached. Every OmniRoute unit carries a drop-in (/etc/systemd/system/actions.runner.diegosouzapw-OmniRoute.<name>.service.d/10-killmode.conf) with KillMode=mixed: SIGTERM to the listener first, SIGKILL to the whole cgroup at TimeoutStop. It takes effect on the unit's next restart — restart one runner at a time, only when idle, with the idle check and the restart in the same command.

Operating rules

• Heavy-build ceiling: 2 at a time — enforced by label. Every job that runs a next build (ci.yml build, npm-publish.yml publish, both nightly-release-green validations) targets [self-hosted, omni-build], and only two runners carry that label (omniroute-113-5, omniroute-113-6, added through the runners API — no re-registration). The other six keep omni-release and take nothing heavy; GitHub queues a third build instead of the kernel killing one. Pair with the heavy-build-* concurrency lanes in ci.yml. To add capacity, label another runner — never raise the count past what 31 GB holds (one next-build ≈ 14–16 GB).
• Never clean /tmp or _work by hand while any runner is busy. A check-then-delete with a gap between the two is how a live Build job lost its _work on 2026-08-27. The janitor does the check and the removal in one step; let it.
• Stopping a runner mid-job cancels the job (observed live): systemctl stop only when its listener has no Runner.Worker child — and do it in one command.
• Workflows must not park artefacts in /tmp (it is RAM). Download to $RUNNER_TEMP (on disk, per runner) — the 1.3 GB next-build artefact took 27–32 minutes to land on the tmpfs and 2 minutes to upload from disk.
• The .15 VPS is homologation-only — never runs CI runners.