You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Andrew Mashenkov (JIRA)" <ji...@apache.org> on 2019/01/30 16:04:00 UTC
[jira] [Updated] (IGNITE-11152) IgniteTxManager.idMap possible
memory leak
[ https://issues.apache.org/jira/browse/IGNITE-11152?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Andrew Mashenkov updated IGNITE-11152:
--------------------------------------
Ignite Flags: (was: Docs Required)
> IgniteTxManager.idMap possible memory leak
> ------------------------------------------
>
> Key: IGNITE-11152
> URL: https://issues.apache.org/jira/browse/IGNITE-11152
> Project: Ignite
> Issue Type: Bug
> Components: mvcc
> Reporter: Stepachev Maksim
> Priority: Major
> Labels: Faillover, Hanging, Transactions, mvcc_stabilization_stage_1
> Fix For: 2.8
>
>
> CacheContinuousQueryAsyncFailoverMvccTxSelfTest.testMultiThreadedFailover sometimes finished with OMM.
> Heapdump analyze showed that leak happened in IgniteTxManager.idMap, this map contains *2_097_152* instances of GridNearTxLocal with *ACTIVE state* and *without* finishFut *and prepFut.*
>
> {code:java}
> while (!updated) {
> try {
> prevVal = (Integer)qryClnCache.getAndPut(key, val);
> updated = true;
> }
> catch (CacheException e) {
> assertSame(atomicityMode(), CacheAtomicityMode.TRANSACTIONAL_SNAPSHOT);
> }
> }
> {code}
>
>
> Possible the CacheException is common and may hide wrong cases. Change it at specific (ignite-10976).
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)