[
http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-222?pag...
]
Sanne Grinovero commented on HSEARCH-222:
-----------------------------------------
I need some clues about how to fix this.
The nice thing is that if we use the same resource to insert than we used for
delete no lock is needed. So the other issues you say it depends on are
HSEARCH-226 and the move to Lucene 2.4?
Entities not found during concurrent update
-------------------------------------------
Key: HSEARCH-222
URL:
http://opensource.atlassian.com/projects/hibernate/browse/HSEARCH-222
Project: Hibernate Search
Issue Type: Bug
Components: directory provider, engine
Reporter: Sanne Grinovero
Fix For: 3.1.0.Beta2
As we update an entity by deleting it from index by using an IndexReader, and afterwards
we reinsert it
using an IndexWriter, there is a point in time in the middle were no DPlock is held on
the Directory,
At this moment a ReaderProvider could (re)open the index and
all searches performed on this IndexReader will not find the entity at all.
We should use a Read/Write Lock on DP instead of the current ReentrantLock to use a
write
lock during index updates and avoid creating a bottleneck for different threads needing a
searcher.
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://opensource.atlassian.com/projects/hibernate/secure/Administrators....
-
For more information on JIRA, see:
http://www.atlassian.com/software/jira