Repository navigation
CI: self-hosted jobs run on the fleet-ci pool (beelink, minis, x1) - #1012
Merged
Merged
Conversation
minis and x1 now serve this repo too (mkreyman/infra decisions/0021). Each carries the same CI-only Postgres on 127.0.0.1:5433 (the same pgvector digest) that every self-hosted job here already targets, and a developer's databases on 5432, which none of them touch. The GitHub-hosted jobs are unchanged.
Review fixes: scale-test's tmpfs note says its memory is a developer's RAM on minis and x1, outside ci.slice and unseen by the CPU-only gate; scale-nightly legs can run at once across hosts, and a leftover database is on the host the job page names; CLAUDE.md and the scale runbook stop saying one runner.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The seven self-hosted jobs in
ci.yml(lint, test, retrieval-eval, security, dialyzer, scale-test, scale-nightly) move from[self-hosted, beelink]to[self-hosted, fleet-ci]. Mark wants minis and x1 to be full CI runners (mkreyman/infra decisions/0021, infra#325), and since 2026-10-21 both serve loopctl. beelink still takes a job first; minis and x1 take it after 45 s queued and only while their measured CPU leaves room.Each host runs the same CI-only Postgres on 127.0.0.1:5433 with the same pgvector digest, and every self-hosted job here already targets 5433, never the developer's 5432. The GitHub-hosted jobs are unchanged. The static runner
beelink-loopctlnow carriesfleet-citoo.Testing
postgres-cirunspgvector/pgvector@sha256:40d828ed…, the digest scale-test pins.Risk
scale-test's tmpfs Postgres can hold up to 4 GB of a dev box's RAM while it runs; the comment on it says so. A job on a dev box shares CPU with the developer inside ci.slice.
How to revert
Revert this PR; every self-hosted job goes back to
[self-hosted, beelink].Review findings