[Feature Request] Can we have more descriptive worker side error log?
Is your feature request related to a problem? Please describe.
While we changed the workflow/activity code and redeployed it, we expected the running workflow to use the new code. But instead, we faced a burst of the below message
2022-09-16T04:51:36.050908Z ERROR temporal_sdk_core::worker::workflow::managed_run: Error in run machines, error: RunUpdateErr { source: Nondeterminism("Complete workflow machine does not handle this event: HistoryEvent(id: 66, Some(ActivityTaskScheduled))"), complete_resp_chan: Some(Sender { inner: Some(Inner { state: State { is_complete: false, is_closed: false, is_rx_task_set: true, is_tx_task_set: false } }) }) }
I wanted to understand the log and make the following actions but we didn't have any additional log nor any hint from above log line about what to check. I wish the description of the error log from the worker side is more verbose.
Describe the solution you'd like
More descriptive error logs from worker side
Yes, this is a non-determinism issue. See https://docs.temporal.io/workflows (specifically https://docs.temporal.io/workflows/#code-changes-can-cause-non-deterministic-behavior).
I wish the description of the error log from the worker side is more verbose.
Any specific suggestions on what you would like to see here? Would a link to those docs be sufficient?
This error is written from our core engine, so I am transferring the issue there for some more discussion.
To clarify here, we need contextual state on our Core logs. We need workflow ID, run ID, workflow type, etc.
Will be fixed by #410
There's more info available by default now.