Hi Sam, Thanks for sending the tool. After some testing and debugging I found that the "option page-count 32" in your client spec is the culprit for the memory leak observed at server. Please remove that one line from you specs and you will have a smoothly running system. I am still investigating the root cause for the problem, when using that option. Will mail back once I find it. Till then please use the default option of page-count 2. That will solve the issue. Let me know if it works for you. Regards, Amar On Dec 18, 2007 4:06 PM, Amar S. Tumballi <amar@xxxxxxxxxxxxx> wrote: > > > The attached benchmark program can be used to cause the problem > > rewrite_test --blocksize 8192 --count 8192 /mnt/foobar > > should do it. Bonnie will also cause the problem when it does its > > rewrite test. > > > > Sam Douglas > > > > > > Sam, > can you send me the "rewrite_test" program.. (it was not attached). Using > your spec files, i couldn't get memory leak by running 'dbench', kernel > compilation.. etc. > > Regards, > Amar > -- Amar Tumballi Gluster/GlusterFS Hacker [bulde on #gluster/irc.gnu.org] http://www.zresearch.com - Commoditizing Supercomputing and Superstorage!
