Repository navigation
Node daemon image ghcr.io/ai-computing/rm-accel; chart 0.5.0 uses it - #8
Merged
Merged
Conversation
- Dockerfile: python:3.11-slim + rm-abstract (no torch) + rm-core v0.1.1 + kubernetes + numpy. Vendor tools stay on the host; /host/usr/bin is on the image's PATH. - Image workflow: build and smoke-test on PRs, publish :master and :X.Y.Z / :X.Y from v* tags with GITHUB_TOKEN (packages: write). - Chart 0.5.0 (appVersion 0.4.2): default image ghcr.io/ai-computing/rm-accel, new value hostBin (default /usr/bin) mounting the host's vendor tools, so the init-container workaround is no longer needed. hostDev and the writable /tmp stay. CI lints the chart. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
gspark-etri
added a commit
that referenced
this pull request
Oct 20, 2026
Node daemon image ghcr.io/ai-computing/rm-accel; chart 0.5.0 uses it
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Roadmap W2 (ai-rm
docs/roadmap.md): reproducible deployment of the L1 node daemon.What
Dockerfile:python:3.11-slim+ rm-abstract (installed--no-deps, so no torch) + rm-core v0.1.1 +kubernetes+numpy. Vendor CLI tools are not baked in;/host/usr/binis on the image'sPATH. The build runs--provider sim --once --dry-runas a self-check..github/workflows/image.yml: on PRs build + smoke-test (sim provider dry-run, PATH check); onmasterpushghcr.io/ai-computing/rm-accel:master(+sha-…), onvX.Y.Ztags push:X.Y.Zand:X.Y.GITHUB_TOKEN,packages: write.appVersion0.4.2): default imageghcr.io/ai-computing/rm-accel:<appVersion>; new valuehostBin(default/usr/bin) mounts the host's vendor tools read-only at/host/usr/bin. With that, sites no longer need the init-container +PYTHONPATHworkaround.hostDevand the writable/tmp(rbln-stat) are unchanged; the extension points stay for sites with their own image.Helm chart lint(default and sim values). Version 0.4.2, CHANGELOG, README (EN/KO),ai-rm.yamldeploy.image.After merge
v0.4.2soghcr.io/ai-computing/rm-accel:0.4.2(the chart default) exists;:masteris published by the merge itself.rm-accelpackage to public (org Packages settings) so nodes can pull without a secret.umbrella-chart) pins this chart through the submodule; re-bump it to the merge commit / tag.Verified
python -m rm_abstract.airm.daemon --provider sim --once --dry-runprints 2 AcceleratorProfiles + node patch;--provider rm-abstractwithout tools prints an empty list. No torch needed.helm lint/helm template(v3.16.4) pass for default and sim values; rendered DaemonSet has/tmpemptyDir,/devhostPath,/usr/bin→/host/usr/binread-only.python tools/check_sensitive.py .: clean.Imagejob; ai-rm's sim e2e builds this Dockerfile and runs it on kind)./usr/binon RNGD/ATOM nodes (same mount the site already uses via the workaround).🤖 Generated with Claude Code