You are viewing a plain text version of this content. The canonical link for it is here.
Posted to oak-issues@jackrabbit.apache.org by "Stefan Egli (Jira)" <ji...@apache.org> on 2024/11/06 13:11:00 UTC
[jira] [Updated] (OAK-11254) backgroundPurge interferes with fullgc tests
[ https://issues.apache.org/jira/browse/OAK-11254?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Stefan Egli updated OAK-11254:
------------------------------
Description:
DocumentNodeStore.backgroundPurge is started in the background and runs right after initialization. Amongst others this method cleans up unmerged branch commits on root, as noticed by [~baedke] offline. FullGC tests have seen flakyness with exactly "missing" unmerged branch commits on root.
We need a way to disable backgroundPurge for certain tests - to exclude such interference.
As a short-term fix however, the first idea is to add a sleep(1000) to tests that have shown flakyness.
was:
DocumentNodeStore.backgroundPurge is started in the background and runs right after initialization. Amongst others this method cleans up unmerged branch commits on root. FullGC tests have seen flakyness with exactly "missing" unmerged branch commits on root.
We need a way to disable backgroundPurge for certain tests - to exclude such interference.
As a short-term fix however, the first idea is to add a sleep(1000) to tests that have shown flakyness.
> backgroundPurge interferes with fullgc tests
> --------------------------------------------
>
> Key: OAK-11254
> URL: https://issues.apache.org/jira/browse/OAK-11254
> Project: Jackrabbit Oak
> Issue Type: Test
> Components: documentmk
> Reporter: Stefan Egli
> Assignee: Stefan Egli
> Priority: Critical
>
> DocumentNodeStore.backgroundPurge is started in the background and runs right after initialization. Amongst others this method cleans up unmerged branch commits on root, as noticed by [~baedke] offline. FullGC tests have seen flakyness with exactly "missing" unmerged branch commits on root.
> We need a way to disable backgroundPurge for certain tests - to exclude such interference.
> As a short-term fix however, the first idea is to add a sleep(1000) to tests that have shown flakyness.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)