fix: recovery after agent kill and rejoin #325
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
raresgaia123
changed the title
raresgaia123
force-pushed
the
recover_agent_failure
branch
2 times, most recently
from
a28aed2 to
06ea259
Compare
In the case when an agent that deployed workers is destroyed, the current code implementation would skip the agent that previously deployed failed workers if it wasn't found during the first iteration of recovery, continuing with remaining agents. We had to refactor the code to remove this limitation. So when an agent fails and is restarted, it will be included in the recovery process. Added changes for when an agent joins controller, to also update all existing job contexts with the details about this agent. So whenever an agent joins, all job contexts will be able to use its resources.
raresgaia123
force-pushed
the
recover_agent_failure
branch
from
06ea259 to
772cdc7
Compare
myungjin
approved these changes
Dec 16, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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.
Description
In the case when an agent that deployed workers is destroyed, the current code implementation would skip the agent that previously deployed failed workers if it wasn't found during the first iteration of recovery, continuing with remaining agents. We had to refactor the code to remove this limitation. So when an agent fails and is restarted, it will be included in the recovery process. Added changes for when an agent joins controller, to also update all existing job contexts with the details about this agent. So whenever an agent joins, all job contexts will be able to use its resources.
Type of Change
Checklist