Re: [Gluster-devel] handling open fds and graph switches

On 08/07/2013 05:11 AM, Raghavendra G wrote: On Wed, Aug 7, 2013 at 12:21 AM, Raghavendra Bhat > wrote: On 08/06/2013 05:22 PM, Raghavendra Gowdappa wrote: - Original Message - From: "Raghavendra Bhat" mailto:rab...@redhat.com>>

Re: [Gluster-devel] Snapshot design for glusterfs volumes

- Original Message - > From: "Brian Foster" > To: "Shishir Gowda" > Cc: gluster-devel@nongnu.org > Sent: Wednesday, August 7, 2013 3:11:30 AM > Subject: Re: [Gluster-devel] Snapshot design for glusterfs volumes > On 08/06/2013 12:16 AM, Shishir Gowda wrote: > > Hi Brian, > > > > - A bar

Re: [Gluster-devel] handling open fds and graph switches

On Wed, Aug 7, 2013 at 12:21 AM, Raghavendra Bhat wrote: > On 08/06/2013 05:22 PM, Raghavendra Gowdappa wrote: > >> >> - Original Message - >> >>> From: "Raghavendra Bhat" >>> To: gluster-devel@nongnu.org >>> Sent: Tuesday, August 6, 2013 1:52:40 PM >>> Subject: [Gluster-devel] handling

Re: [Gluster-devel] Snapshot design for glusterfs volumes

On 08/06/2013 12:16 AM, Shishir Gowda wrote: > Hi Brian, > > - A barrier is similar to a throttling mechanism. All it does is queue up the > call_backs at the server xlator. > Once barrier'ing is done, it just starts unwinding, so that clients can now > get the response. > The idea is that

Re: [Gluster-devel] handling open fds and graph switches

On 08/06/2013 05:22 PM, Raghavendra Gowdappa wrote: - Original Message - From: "Raghavendra Bhat" To: gluster-devel@nongnu.org Sent: Tuesday, August 6, 2013 1:52:40 PM Subject: [Gluster-devel] handling open fds and graph switches Hi, As of now, there is a problem when following set

Re: [Gluster-devel] [Gluster-users] Status of the SAMBA module

On 08/06/2013 05:09 PM, haiwei.xie-soulinfo wrote: > Hi, > > Unfortunately, add the two params into smb.conf, nfs client OS will core dump > with dd testing. > I reconfigure gluster vol, without any new params in smb.conf, performance > looks good. Maybe > former low performance result from erro

Re: [Gluster-devel] new glusterfs logging framework

On Tue, Jul 30, 2013 at 05:29:43AM -0400, Balamurugan Arumugam wrote: > > Hi All, > > Recently new logging framework was introduced [1][2][3] in glusterfs > master branch. You could read more about this on doc/logging.txt. In > brief, current log target is moved to syslog and user has an opti

Re: [Gluster-devel] handling open fds and graph switches

- Original Message - > From: "Raghavendra Bhat" > To: gluster-devel@nongnu.org > Sent: Tuesday, August 6, 2013 1:52:40 PM > Subject: [Gluster-devel] handling open fds and graph switches > > > Hi, > > As of now, there is a problem when following set of operations are > performed on a f

Re: [Gluster-devel] [Gluster-users] Status of the SAMBA module

Hi, Unfortunately, add the two params into smb.conf, nfs client OS will core dump with dd testing. I reconfigure gluster vol, without any new params in smb.conf, performance looks good. Maybe former low performance result from error config. [root@infi131 cifs]# dd if=/dev/zero of=t1 bs=64k cou

Re: [Gluster-devel] Access to git repo for minor glusterfs versions such as 3.3.0.7 and 3.3.0.11 so I can git diff them

On 08/05/2013 10:41 PM, Diane Feddema wrote: Hello All, I would like to do a git diff on glusterfs versions 3.3.0.7 and 3.3.0.11. How can I gain access to the git repository for these versions? These are the RPM "versions" — they're all glusterfs-3.3.0 and there is no diff to the glusterfs s

Re: [Gluster-devel] handling open fds and graph switches

On 08/06/2013 04:22 AM, Raghavendra Bhat wrote: > > Hi, > > As of now, there is a problem when following set of operations are > performed on a file. > > open () => unlink () => do a graph change (not reconfigure) => fop on > the opened fd (may be write) > > In the above set of operations, the

Re: [Gluster-devel] [Gluster-users] Status of the SAMBA module

hi, All samba vfs low performance, compare with non-vfs export dir. nonvfs: [root@infi131 cifs]# dd if=/dev/zero of=test1 bs=64k ^C记录了55548+1 的读入 记录了55548+0 的写出 3640393728字节(3.6 GB)已复制，17.192 秒，212 MB/秒 vfs: [root@infi131 cifs]# cd /mnt/cifs [root@infi131 cifs]# dd if=/dev/zero of=test1 bs=64k

Re: [Gluster-devel] [Gluster-users] Status of the SAMBA module

hi, my testing: # glusterd -V glusterfs 3.4.0 built on Jul 25 2013 04:12:26 # smbd -V Version 3.6.9-151.el6 samba-gluster-vfs: git://forge.gluster.org/samba-glusterfs/samba-glusterfs-vfs.git smb.conf: [gvol] comment = For samba export of volume test vfs objects = glusterfs

Re: [Gluster-devel] [Gluster-users] Status of the SAMBA module

Hi, 1. Please add in global section in smb.conf the following lines: kernel oplocks = no stat cache = no 2. service smb restart Please try the same test and let us know if that helps to increase the throughput. Raghavendra Talur On Tue, Aug 6, 2013 at 1:00 PM, haiwei.xie-soulinfo

[Gluster-devel] handling open fds and graph switches

Hi, As of now, there is a problem when following set of operations are performed on a file. open () => unlink () => do a graph change (not reconfigure) => fop on the opened fd (may be write) In the above set of operations, the fop performed on the fd after the graph switch fails with EBAD