Repository navigation
Recover ordinary model containers after sampled idleness - #47
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Add a separate recovery observer for pinned ordinary-model containers. It records running intent, waits for a fresh two-hour sampled idle window after an eligible exit or host reboot, checks device holders and container identity before starting, and preserves observed manual stops. The existing DeepInfer policy remains unchanged.
Recovery is best effort: sampled idleness is not a resource lease, exit codes do not always reveal operator intent, and the observer refuses to connect while the Docker socket-activation unit is active. System-process exemptions require exact service, process, executable, command-line and environment fingerprints.
Validation:
python -m unittest discover -s tests -v(275 tests, 8 skipped). Fault injection and full two-hour recovery on a live serving machine were not performed.