mcvsubbu commented on issue #6516:
URL:
https://github.com/apache/incubator-pinot/issues/6516#issuecomment-771818395
I think Helix preserves the error stack/messages for a few days? State
stored in segmentDataManager can go away on restart. I think there is value in
retrieving helix messages (if they exist) as well.
On a recent issue in Linkedin, it turned out that the state transitions did
not succeed and caused a performance hit on pinot servers. The errors went away
upon restart. I can totally imagine an admin restarting the servers to get the
system up & running.
It may make sense to have an API gather all debug information available and
output it as a a blob of text. We can enhance this as we feel like, without
regarding to backward compat, etc.
curl controller:port/tables/{tableName}/debugState 1> saved 2>&1
can be done before any restart is attempted. This can also retrieve states
from each of the servers that have segments for the table.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org
For additional commands, e-mail: commits-h...@pinot.apache.org
