feat: support ulimits service key (final per-container Bucket A key); reclassify sysctls as pod-level#26
Merged
Merged
Conversation
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 final per-container Bucket A key, plus a docs-only
sysctlsreclassification.ulimits->--ulimit name=soft:hard(mapping form) /--ulimit name=value(scalar; podman sets soft = hard). Dedicated_ulimit_args/_add_ulimit_flagsemission and a dedicated_validate_ulimits(kept out of_validate_service_formsto avoid the ruffC901limit). A mapping value must have exactlysoftandhard, each an int/string — a non-scalarsoft/hardis rejected at parse time rather than mis-emitted as a garbage flag. Args are${VAR}-interpolated.sysctlsreclassified out of Bucket A (no code change — it keeps raising as unsupported). Only namespaced sysctls (net.*, the IPCkernel.*/fs.mqueue.*set) are settable, and a pod owns net + ipc, so they belong onpodman pod create --sysctl(pod-wide), not a per-container run. Deferred alongsidedns. A guard test locks the non-support.Design and rationale (single home):
planning/changes/2026-07-09.07-ulimits-service-key.mdandplanning/decisions/2026-07-09-sysctls-pod-level.md. Behavior promoted intoarchitecture/supported-subset.md; the coverage audit now marks per-container Bucket A complete.Tests: 195 passing at 100% line coverage;
just lint-ciclean.🤖 Generated with Claude Code