[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [Linux-cluster] rm -r on gfs2 filesystem is very slow

We found with GFS2 on RHEL5.3 the same issue, and although we did not resolve it (and are migrating to NFS+SAN instead), we were able to partially mitigate it with several suggestions from either RedHat's support or from the cluster list:
1. disable LS color output (http://kbase.redhat.com/faq/docs/DOC-6533)
2. the "nodiratime,noatime" mount flags
3. Unlocking the plock rate to allow maximum locks generated and purged per second with the "plock_rate_limit=0" option in cluster.conf

That said though, we found that the GFS2 was only about as fast as NFS4 on GigE networks, and not the best solution for our setup.

On Wed, Jul 8, 2009 at 4:58 PM, Peter Schobel <pschobel 1iopen net> wrote:
I am trying to set up a four node cluster but am getting very poor
performance when removing large directories. A directory approximately
1.6G  in size takes around 5 mins to remove from the gfs2 filesystem
but removes in around 10 seconds from the local disk.

I am using CentOS 5.3 with kernel 2.6.18-128.1.16.el5PAE.

The filesystem was formatted in the following manner: mkfs.gfs2 -t
wtl_build:dev_home00 -p lock_dlm -j 10
/dev/mapper/VolGroupGFS-LogVolDevHome00 and is being mounted with the
following options: _netdev,noatime,defaults.

If anyone knows what could be causing this please let me know. I'm
happy to provide any other information.


Peter Schobel

Linux-cluster mailing list
Linux-cluster redhat com

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]