Hi- On Fri, 29 Aug 2008, Vitaly Shishakov wrote: > The feature i liked, is that when i switched off one of the two storage > nodes, all data was still accessible !!! > this is a major reason why i'd preffer this solution over others (like > lustre, coda and etc, which are much > more complex to set up, not speaking about gfs and ocfs) > with 2 nodes running i have read/write ratio about 20MB/sec (i believe > that was due to debugging checks, i built it with default options, from > the default GIT-branch) > i used btrfs for osd volumes.
There is a known performance issue when using btrfs for osd volumes that hasn't been looked at yet, so you may want to try using ebofs for now. The --debug_* options can slow things down dramatically. I think the compiler optimizations are on by default. There are a few other compile-time debug options that are enabled by default, but they shouldn't be terribly significant. > the only bug so far: when monitor server goes down, and ceph volume was > not unmounted, it stalls. nothing happens even when monitor server is up > again -- i get messages like: ceph: mds0 session probably timed out, > requesting mds map Yeah, this is also a known issue... there is no forced unmount. sage ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Ceph-devel mailing list Ceph-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ceph-devel
