Orchestrator restart

Hello!
What is heppening with the data from queue or bucket if the Orchestrator is restrarted.?

Hi

Restarting orchestrator won’t impact queue and it’s items
Are you facing any challenge in that

@Andrei_Croitoru

No. But i am searching for any written documentation to prove that, and could not find any so i asked here. :slight_smile:

1 Like

Fine then @Andrei_Croitoru

I just wanted to add on that while Queue Items are stored in the database and Storage Buckets contents are stored where ever they are configured to be stored (File Systems, AWS, Azure, Min.io) you still run the risk of interrupting anything that is in processing which could cause jobs to fail, or missing triggers, queue items stuck in ‘in progress’, etc.

I would recommended to gracefully shutdown or remove the Orchestrator node from the cluster before simply restarting it.

  • Maintenance Mode is available, unfortunately it applies to the whole platform so if you have multiple Orchestrator nodes in a cluster, you cannot just place a single node into maintenance as it is done at the host/database level. The nice thing if you do use maintenance mode is it provides a audit log entry to indicate if any triggers were missed and their trigger properties so you can manually restart them if needed.
  • Remove specific Orchestrator node you wish to restart from load balance pool if you have a cluster, waiting until all connections have migrated to other nodes.
  • Deactivating Robots, not ideal if taking down the full platform but provides in-progress issues, but doesn’t address triggers, api requests, etc…
1 Like

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.