Himanshu-g81 opened a new pull request, #2489:
URL: https://github.com/apache/phoenix/pull/2489

   On standby clusters, compaction can prematurely purge delete markers that 
have timestamps newer than the replication consistency point. This causes 
permanent stale data because when replay eventually catches up, the delete 
markers that should have been applied are already gone.
   
   This PR adds a compaction guard that floors `maxLookbackWindowStart` to the 
minimum consistency point across all HA groups, ensuring delete markers newer 
than the consistency point are retained until replay has processed them.
   
   The guard is controlled by `phoenix.replication.compaction.guard.enabled` 
(default `true`). It activates automatically when 
`phoenix.replication.replay.enabled=true` — no additional configuration needed 
on standby clusters. The separate flag allows operators to disable the guard 
independently if needed without turning off replay entirely.


-- 
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]

Reply via email to