[Gluster-devel] Too many open files

[Brent A Nelson]

I suspect you meant "ulimit -a"?
ulimit -a
core file size          (blocks, -c) 0
data seg size           (kbytes, -d) unlimited
max nice                        (-e) 20
file size               (blocks, -f) unlimited
pending signals                 (-i) unlimited
max locked memory       (kbytes, -l) unlimited
max memory size         (kbytes, -m) unlimited
open files                      (-n) 1024
pipe size            (512 bytes, -p) 8
POSIX message queues     (bytes, -q) unlimited
max rt priority                 (-r) unlimited
stack size              (kbytes, -s) 8192
cpu time               (seconds, -t) unlimited
max user processes              (-u) unlimited
virtual memory          (kbytes, -v) unlimited
file locks                      (-x) unlimited

Just in case you did mean uname -a:
uname -a
Linux jupiter01 2.6.15.7-ubuntu1.060913 #1 SMP Wed Sep 13 15:21:26 EDT 
2006 i686 GNU/Linux

The servers are Ubuntu 6.06.1, with a custom kernel for Lustre.  They have 
an old Fuse, not 2.6.3, but I suspect this is more of a glusterfsd issue 
rather than a glusterfs issue.  My third client does have FUSE 2.6.3, and 
it also reports "Too many open files."

I Googled a bit and found someone else had written their own Fuse 
filesystem and encountered the error.  Someone suggested ulimit to remove 
the limit, but he found it was a bug in his code where it was leaking open 
directories.

Even if I can increase the limit, I still suspect it's a bug, and the new 
limit would eventually be reached.

Thanks,

Brent

On Wed, 4 Apr 2007, Dale Dude wrote:

>
> Brent A Nelson wrote:
>> I already looked there, and my file-nr allocated is much lower than 
>> file-max.  It might be hitting some other limit, perhaps even a limit 
>> imposed by the bash that started glusterfsd.  Still it really seems like 
>> there shouldn't be so many allocated file descriptors lying around (no 
>> files should currently be open, and rsync wouldn't have left large numbers 
>> open at the same time, I assume), so I think it's a bug where GlusterFS is 
>> not closing all the file descriptors it opens or allocates.
>> 
>> Thanks,
>> 
>> Brent
>> 
>
> Paste the output of 'uname -a' from a shell that you would normally start 
> glusterfsd.  What distro are you using?
>
>
> _______________________________________________
> Gluster-devel mailing list
> Gluster-devel at nongnu.org
> http://lists.nongnu.org/mailman/listinfo/gluster-devel
>