[
https://issues.apache.org/jira/browse/OAK-1886?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14029184#comment-14029184
]
Marcel Reutegger commented on OAK-1886:
---------------------------------------
My heap dump shows that the five oak instances in the test cluster use quite a
lot of memory. Each oak instance with a MongoMK may use up to 256 MB of memory
for its cache. Multiplied by five is obviously more than the 512 MB we specify
as the maximum heap size (the heap size is configured in the
oak-parent/pom.xml).
I re-ran the test with a lower cache size for the MongoMK instances and this
time no OOME occurred.
> OOM in
> org.apache.jackrabbit.oak.jcr.OrderedIndexConcurrentClusterIT::deleteConcurrently()
> ------------------------------------------------------------------------------------------
>
> Key: OAK-1886
> URL: https://issues.apache.org/jira/browse/OAK-1886
> Project: Jackrabbit Oak
> Issue Type: Bug
> Components: core
> Reporter: Davide Giannella
> Assignee: Marcel Reutegger
>
> When running the IT profile it fails with OOM when executing the
> {{org.apache.jackrabbit.oak.jcr.OrderedIndexConcurrentClusterIT::deleteConcurrently()}}
> test.
> Either increase the heap size to 2GB or find a way to lower the memory
> consumption.
> On my local by pushing the heap it doesn't happen.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
