[ 
https://issues.apache.org/jira/browse/LUCENE-9304?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17083468#comment-17083468
 ] 

ASF subversion and git services commented on LUCENE-9304:
---------------------------------------------------------

Commit 47ed21324c41d4414cd259ceaba15f5d45863ac6 in lucene-solr's branch 
refs/heads/branch_8x from Simon Willnauer
[ https://gitbox.apache.org/repos/asf?p=lucene-solr.git;h=47ed213 ]

LUCENE-9304: Fix IW#getMaxCompletedSequenceNumber()  (#1427)

After recent refactoring on LUCENE-9304 `IW#getMaxCompletedSequenceNumber()` 
might
return values that belong to non-completed operations if a full flush is 
running, a new delete
queue is already in place but not all DWPTs that participate in the full flush 
have finished it's in
flight operation. This caused rare failures in
`TestControlledRealTimeReopenThread#testControlledRealTimeReopenThread` where
documents are not actually visible given the max completed seqNo. This change 
streamlines
the delete queue advance, adds a dedicated testcase and ensures that a delete 
queues
sequence Id space is never exhausted.

> Clean up DWPTPool 
> ------------------
>
>                 Key: LUCENE-9304
>                 URL: https://issues.apache.org/jira/browse/LUCENE-9304
>             Project: Lucene - Core
>          Issue Type: Improvement
>    Affects Versions: master (9.0), 8.6
>            Reporter: Simon Willnauer
>            Assignee: Simon Willnauer
>            Priority: Major
>             Fix For: master (9.0), 8.6
>
>          Time Spent: 11h 10m
>  Remaining Estimate: 0h
>
> DWPTPool currently uses an indirection called ThreadState to hold DWPT 
> instances. This class holds several information that belongs in other places, 
> inherits from ReentrantLock and has a mutable nature. Instead we could pool 
> the DWPT directly and remove other indirections  inside DWPTFlushControl if 
> we move some of the ThreadState properties to DWPT directly. The threadpool 
> also has a problem that is grows it's ThreadStates to the number of 
> concurrently indexing threads but never shrinks it if they are reduced. With 
> pooling DWPT directly this limitation could be removed. 
> In summary, this component has seen quite some refactoring and requires some 
> cleanups and docs changes in order to stay the test of time.  



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org

Reply via email to