Bernard Li wrote: > Hi Daniel: > > On Tue, Jul 7, 2009 at 8:43 AM, Daniel Pocock<dan...@pocock.com.au> wrote: > > >> I've noticed on a large gmetad install (400 clusters) that some of the >> data threads get stuck in the CLOSE_WAIT state, and cease polling the >> gmonds. Other data threads are running normally. Over time, more and >> more of the threads get stuck. >> >> Looking at one of the threads, I found that it is stuck in hash_insert, >> waiting for a reader to release a lock so that the writer can get the >> exclusive lock. >> >> Specifically, the hash_insert where it sticks is in process_xml.c: >> >> rdatum = hash_insert( &hashkey, &hashval, xmldata->root); >> >> Has anyone else seen anything like this? >> > > Not me personally, but if you don't hear from anybody else, I would > suggest you file this as a bug. > > BTW, I am assuming you are running the latest version (i.e. 3.1.2)? > I found this in gmetad 3.1.0, I'm going to test 3.1.2 as well. However, looking at a diff between the two versions, it is not immediately obvious if any code related to this has changed.
I'll open the bug report once I have more detail and/or a solution. ------------------------------------------------------------------------------ Enter the BlackBerry Developer Challenge This is your chance to win up to $100,000 in prizes! For a limited time, vendors submitting new applications to BlackBerry App World(TM) will have the opportunity to enter the BlackBerry Developer Challenge. See full prize details at: http://p.sf.net/sfu/Challenge _______________________________________________ Ganglia-developers mailing list Ganglia-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ganglia-developers
