This release contains important fixes for SQL Server-based job storage to work better with sub-second polling (including
QueuePollingInterval) and properly send heartbeats for long-running jobs even if CLR’s Thread Pool is starved for long periods of time. It is also very likely that some problems related to high CPU usage and high number of fetching queries occurring after deploying new application version to IIS were also fixed (they are almost impossible to catch up, but same workaround already helped in the past).
Fixed – Reduce the number of attempts in
BackgroundJobStateChangerto avoid infinite loops.
- Fixed – Typos in German translation (by @saxx).
- Fixed – Translations in Turkish translation (by @can-zengin).
- Fixed – Possibly fixed CPU consumption problems and high amounts of fetching queries after deploys to IIS.
- Fixed – No more than a single long-polling query is allowed per storage instance when using sub-second polling.
- Fixed – Don’t depend on thread pool when sending heartbeats for active jobs to avoid problems when it’s starved.