[
https://issues.apache.org/jira/browse/IGNITE-17875?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17643891#comment-17643891
]
Maksim Timonin commented on IGNITE-17875:
-----------------------------------------
TX recovery finishes before exchange, after that all counters are finalized.
Looks like "Failed to update the partition counter..." shows cases when a
partition copy have gaps (received LVM < local HWM).
The open question what if received LWM > local HWM > local LWM. Looks like such
cases will be fixed with rebalance. Then no need in logging here.
Close this ticket as "Won't fix", counters must be refactored.
> TX counters with gaps must be logged on every PME
> -------------------------------------------------
>
> Key: IGNITE-17875
> URL: https://issues.apache.org/jira/browse/IGNITE-17875
> Project: Ignite
> Issue Type: Sub-task
> Reporter: Anton Vinogradov
> Assignee: Maksim Timonin
> Priority: Major
> Labels: ise
>
> TX counters on PME (when all operations are finished) must contain no gaps.
> Such partitions should be listed (with counters) with a special warning.
> Currently, we already have a "Failed to update the counter ..." warning which
> is logged when the primary tries to set the backup's update counter to its
> LWM which is less than the backup's HWM.
> But, the better case is to list all broken partitions during the full message
> calculation during the PME at the coordinator.
> All counters must be listed for every partition which has at least one
> counter with gaps.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
