This repository has been archived by the owner on Aug 21, 2024. It is now read-only.
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Fixes for instanceserver connection bugs (#10766)
A rare situation was cropping up where an instance record was getting removed from the database after an instance provision. This would cause initializeInstance to return false, and loadEngine to not run, but instanceStarted would remain true. That instance wouldn't start, but the server would still remain in the pool of Ready servers. Later clients assigned to that pod would then just endlessly wait for InstanceServerState.ready to be true, which would never happen. setting instanceStarted back to false if initializeInstance returns false should fix this. Increased the timeout on assigned but not-connected instances to 60 seconds from 30. This may be what was causing those instance records to be removed, if a client on a slow network was getting the provision but then taking more than 30 seconds to make the websocket connection.
- Loading branch information