Ocfs2 users - May 2013 - global_bitmap lock busy for a long time

I am often experiencing long delays on ocfs2 volumes when another 
process is writing to the fs.  The delays seem to coincide with a busy 
lock on //global_bitmap.  I can understand why a busy lock on the 
global_bitmap would cause delays (especially write delays), but it seems 
to block reads too?  Also, the lock seems to be held for quite a long 
time at times, probably over a minute.  This is on a 3TB fs that is 80% 
full.  This is a real problem for me because nfs mounts fail and then 
give stale NFS handle even after the volume is free again.  Is there a 
way to improve the length of time the global_bitmap is locked?

Thanks,

Andy

On 05/08/2013 12:07 PM, Andy wrote:
> 	I am often experiencing long delays on ocfs2 volumes when another
> process is writing to the fs.  The delays seem to coincide with a busy
> lock on //global_bitmap.  I can understand why a busy lock on the
> global_bitmap would cause delays (especially write delays), but it seems
> to block reads too?  Also, the lock seems to be held for quite a long
> time at times, probably over a minute.  This is on a 3TB fs that is 80%
> full.  This is a real problem for me because nfs mounts fail and then
> give stale NFS handle even after the volume is free again.  Is there a
> way to improve the length of time the global_bitmap is locked?
it's most likely because fs is fragmented. try mount -o localalloc=2 
<dev> and see if that helps.
>
> Thanks,
>
> Andy
>
>
>
>
>
> _______________________________________________
> Ocfs2-users mailing list
> Ocfs2-users at oss.oracle.com
> https://oss.oracle.com/mailman/listinfo/ocfs2-users