I looked through those notes to no avail. Is there a way to "tag" a curator instance/change its thread pool, so that I can see which curator instance (if either) is leaking?
On Mon, Mar 31, 2014 at 5:39 PM, Jordan Zimmerman < [email protected]> wrote: > Here's the old change log, have a look: > > https://cwiki.apache.org/confluence/display/CURATOR/Old+Change+Log > > Also, check the Apache release notes: > > > https://issues.apache.org/jira/browse/CURATOR/?selectedTab=com.atlassian.jira.jira-projects-plugin:versions-panel > > > From: David Greenberg [email protected] > Reply: David Greenberg [email protected] > Date: March 31, 2014 at 4:37:30 PM > To: Jordan Zimmerman [email protected] > Cc: [email protected] [email protected] > Subject: Re: Noticing a curator memory/thread leak > > This is happening with Curator 1.2.3. > > > On Mon, Mar 31, 2014 at 4:55 PM, Jordan Zimmerman < > [email protected]> wrote: > >> Wow - that's really hard to say. What version of Curator? I seem to >> recall some threading bugs fixed over the years. >> >> -JZ >> >> >> From: David Greenberg [email protected] >> Reply: [email protected] [email protected] >> Date: March 31, 2014 at 3:53:48 PM >> To: [email protected] [email protected] >> Subject: Noticing a curator memory/thread leak >> >> I have been noticing that my application seems to be leaking curator >> threads and CuratorFrameworkImpls. >> >> These leak at around 10-200 per day. >> >> I create 2 curators in my application--one is from Datomic, and the other >> is used for a LeaderLatch. I can see that the LeaderLatch and Datomic >> instance each retain a single curator reference; however, I see hundreds of >> curators that are kept around along with their Executors, which ends up >> draining a lot of resources. >> >> I'm using an older version of Curator--was this or is this a problem >> before? Is it a Curator bug, or a bug in my code that uses it? >> >> Thank you, >> David >> >> >
