Transient error while executing action
I was testing something which involved recursive action execution (run an action, then run it again with
scheduler.runAfter(0, ...), until the base case is reached), and I ran into a failure which was logged as Transient error while executing action.
What is this? Should I expect it to happen regularly? Do I need to orchestrate some state management/retry scaffolding in order to anticipate the possibility of this occurring?4 Replies
If you have the request ID we could look up what the exception was, but yes actions can fail due to transient errors like machine restarts, network failures, backend deploys. If your action can be made idempotent then you could use https://stack.convex.dev/retry-actions
Automatically Retry Actions
Learn how to automatically retry actions in Convex while also learning a little about scheduling, system tables, and function references.
No need to look it up if the cause is something mundane and unsurprising, it just looked from my searching on Discord like in past instances this error indicated an unexpected problem. I can in fact make my action idempotent, so I'll look into that! Thanks @ian 🙂
ed2b9807188ee143 If you feel its worth investigating... I just had it with a fairly standard agent tool call with streaming. Unfortunately it ended the message chain under a success state. Which was sorta successful with the tool call, but no summary text letting the user know it was done.
I think for the Agent it should append a pending message until it's done, which is Agent-specific.