[
https://issues.apache.org/jira/browse/OFBIZ-2353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14114010#comment-14114010 ]
Jacques Le Roux commented on OFBIZ-2353:
----------------------------------------
[~jacopoc] I don't know Philippe's experience (though I know for which client it happened) which leads me to this fix. That's for sure resolved an issue for a client I worked with between 2011 and 2013. It was a clustered environment with 5 OFBIz servers running on a Postgres 9.1 DB cluster with 3 machines (1 master and 2 slaves with Pgpool-II). IIRW the DB cluster was able to handle a limit of maximum 1500 simultaneous connections.
The app servers used the default entityengine.xml parameters but this set of parameters
pool-maxsize="300"
idle-maxsize="40"
time-between-eviction-runs-millis="60000" (instead of 600000)
We were also using the default bank size.
With this setting we did not encounter deadlock situations. There can be situations were a thread wait for another to release the lock, but deadlock should not happen. Because the bank is only in one table, there can't be cross callings (
http://www.postgresql.org/docs/9.1/static/explicit-locking.html#LOCKING-DEADLOCKS)
Ha, our comments cross wired. I believe with the SELECT FOR UPDATE there should not be issues like described by Adrian:
bq. There is a similar issue in the way the entity value is updated. The UPDATE... WHERE... statement needs to include the previous sequence value - so you can guarantee you are updating the same value you read.
[~
[hidden email]]: did you encounter such issues with clustered= true?
> SequenceUtil may generate duplicate IDs in Load Balancing mode
> ---------------------------------------------------------------
>
> Key: OFBIZ-2353
> URL:
https://issues.apache.org/jira/browse/OFBIZ-2353> Project: OFBiz
> Issue Type: Bug
> Components: framework
> Affects Versions: Release Branch 4.0, Release Branch 09.04, Trunk
> Reporter: Philippe Mouawad
> Assignee: Jacopo Cappellato
> Priority: Critical
> Fix For: Release Branch 10.04, Release Branch 11.04, Trunk
>
> Attachments: OFBIZ-2353 SELECT FOR UPDATE solution.patch, OFBIZ-2353 SELECT FOR UPDATE solution.patch
>
>
> If Ofbiz is deploy on 2 servers in Load Balancing Mode
> SequenceUtil will generate duplicate IDs because synchronization is done at JVM level instead of doing it in DB.
> A good replacement implementation would be:
> org.hibernate.id.enhanced.TableGenerator
> But it would involve a dependency on Hibernate
> Philippe
> www.ubik-ingenierie.com
--
This message was sent by Atlassian JIRA
(v6.2#6252)
Locked
