vchag commented on PR #64040:
URL: https://github.com/apache/doris/pull/64040#issuecomment-4763492902
> I am reviewing it, but it may takes sometimes. Could you provide a way to
reproduce this bug?
Sure thing, here you go!
Cluster setup:
- 20 BE nodes, apache/doris:be-4.0.4 (brps 1.4.0, unpatched)
Table shape:
- Duplicate Key (event_time. event_id)
- Partition by range(event_time) with dynamic hourly partitions
- 30+columns with at lease 2-3 inverted indexes (wide table)
Load:
- Synthetics row generator streaming via group_commit=sync_mode Stream Load.
- 15-20K EPS/BE sustained for at least 20-30 minutes (~300-400k EPS total
across 20BEs)
Expected Result:
- Greater than 1 BE crashed with SIGSEGV in
`bvar::Reducer::SeriesSampler::take_sample` within 30 minutes
- Kubernetes signal: RESTART counter incrementing on BE Pods
- Bare-METAL signal: grep SIGSEGV /path/to/be/log/bo.out
Let me know if you need additional details.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
