On Tue, 06 Aug 2013 07:24:26 +0200 Arne Jansen <[email protected]> wrote:
> On 05.08.2013 18:35, Tomasz Chmielewski wrote: > >> I am trying to use qgroups > >> functionality & with a basic random-write workload, it constantly > >> keeps leaking memory & within few minutes of IO, there is either > >> out-of-memory killer trying to kill some tasks or there are > >> page-allocation failures that btrfs or other kernel module > >> experiences. > > > > FYI, I just saw something similar with 3.10 on a server with 32 GB > > RAM: > > > > The result was a frozen server and a need to hard reset. > > > > What do I have to do to reproduce it here? How do you generate the > load? What is the disk setup, what the qgroups setup? Unfortunately I don't have a way to reproduce, as the issue happened to me only once. Server uptime was about 2 weeks, and there were millions of files with multiple hardlinks on disk (backuppc archive copied from ext4 to btrfs). Then, "rm" command was running for many days, about 5 times in parallel, and many other rsync commands, with snapshots being added and removed. When I saw the issue I looked up in btrfs list archives that there was a similar report in the past - but I doubt I will be able to find a reliable way to reproduce this. -- Tomasz Chmielewski http://wpkg.org -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
