FIX: Prevent job stuck at Starting status when AgentWatchdog replaces failed agents #445
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
Zakaria-Kofiro
changed the title
added 7 commits
December 18, 2025 21:43
2 tasks
Zakaria-Kofiro
changed the title
added 7 commits
January 6, 2026 14:16
…ove verbose logging
kevin-mcgoldrick
approved these changes
Jan 9, 2026
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Outdated suggestions cannot be applied.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.
Introduce VMStatus.replaced for AgentWatchdog-replaced agents
When the AgentWatchdog replaced failed agents, they were previously marked as terminated and treated the same as intentionally stopped agents. This caused broken bulk Kill/Stop commands (attempting to reach non-existent agents), incorrect job status calculations (jobs stuck in "Starting" or prematurely marked "Completed"), and inflated agent counts in the UI (e.g., "50/54" instead of "50/50").
The new VMStatus.replaced state keeps failed agents visible for debugging while explicitly excluding them from active job counts and command execution targets.
This PR also fixes a bug where replacement agents were initialized with VMStatus.pending instead of VMStatus.starting, causing the watchdog to wait indefinitely for them to report. Replacement agents now correctly start in starting status and transition normally when they call /v2/agent/ready.
Additionally, public/private IP addresses are now captured and logged during instance creation to aid network connectivity debugging when agents fail to come up.
Please make sure these check boxes are checked before submitting
mvn clean test -P default** PR review process **