Re: [Gluster-users] How to maintain HA using NFS clients if the NFS daemon process gets killed on a gluster node?

On Thu, Jan 28, 2016 at 11:08 AM, Kris Laib wrote: > Soumya, > > CTDB failover works great if the server crashes or the NIC is pulled, but > I don't believe there's anything in the CTDB setup that would cause it to > realize there is a problem if only the glusterfs process responsible for > servi

Re: [Gluster-users] How to maintain HA using NFS clients if the NFS daemon process gets killed on a gluster node?

Kris, You can achieve what you want with Corosync-Pacemaker, Corosync is a heartbeat and Pacemaker is a cluster manager. You can create a pacemaker cluster using the hosts used for the Gluster cluster, then configure a virtual IP resource and Gluster monitoring resources with the count of the n

Re: [Gluster-users] How to maintain HA using NFS clients if the NFS daemon process gets killed on a gluster node?

On 01/28/2016 11:08 AM, Kris Laib wrote: Soumya, CTDB failover works great if the server crashes or the NIC is pulled, but I don't believe there's anything in the CTDB setup that would cause it to realize there is a problem if only the glusterfs process responsible for serving NFS is killed

Re: [Gluster-users] How to maintain HA using NFS clients if the NFS daemon process gets killed on a gluster node?

Soumya, CTDB failover works great if the server crashes or the NIC is pulled, but I don't believe there's anything in the CTDB setup that would cause it to realize there is a problem if only the glusterfs process responsible for serving NFS is killed but network connectivity with other CTDB nod

Re: [Gluster-users] Determining Connected Client Version

On 01/28/2016 02:29 AM, Vijay Bellur wrote: > On 01/27/2016 11:43 AM, Atin Mukherjee wrote: >> >> >> On 01/27/2016 09:43 PM, Raghavendra Bhat wrote: >>> But, why should client version be a problem when brick-log-level is >>> being changed? >> Because the check is there in any volume set operation

Re: [Gluster-users] How to maintain HA using NFS clients if the NFS daemon process gets killed on a gluster node?

On 01/27/2016 09:39 PM, Kris Laib wrote: Hi all, We're getting ready to roll out Gluster using standard NFS from the clients, and CTDB and RRDNS to help facilitate HA. I thought we were good to know, but recently had an issue where there wasn't enough memory on one of the gluster nodes in a

Re: [Gluster-users] Installing 3.7.6 in Debian Wheezy

Kaleb Keithley wrote on 26/01/2016 16:35: > From: "Benjamin Wilson" >> >> Do you have any feel for whether or not getting the Wheezy package >> working will be a quick fix? I was planning deploying the client >> to a number of our Wheezy servers, but if it’s going to be a while >> I might look a

Re: [Gluster-users] Determining Connected Client Version

On 01/27/2016 11:43 AM, Atin Mukherjee wrote: On 01/27/2016 09:43 PM, Raghavendra Bhat wrote: But, why should client version be a problem when brick-log-level is being changed? Because the check is there in any volume set operation. Do we not use OPT_FLAG_CLIENT_OPT for differentiating a cl

[Gluster-users] Write speed issues with 16MB files and using Gluster fuse mount vs NFS

Hi all, We were initially planning on using NFS mounts for our gluster deployment to reduce the amount of client-side changes we had to make to swap out our existing NFS solution. I ran into an HA issue with NFS (see my other post from this morning), so started looking into the Fuse client as

Re: [Gluster-users] Quota issue after upgrading from 3.6.4 to 3.6.6

Please share the glusterd log files along with cmd history log file from all the nodes. Thanks, Atin On 01/27/2016 08:33 PM, Jorick Astrego wrote: > Hi, > > After upgrading one of a 3 node glusterfs cluster to 3.6.6, I cannot get > the HDD-1 volume to start properly on this server. The other ser

Re: [Gluster-users] Determining Connected Client Version

On 01/27/2016 09:43 PM, Raghavendra Bhat wrote: > But, why should client version be a problem when brick-log-level is > being changed? Because the check is there in any volume set operation. > > > Regards, > Raghavendra > > On Wed, Jan 27, 2016 at 9:57 AM, Atin Mukherjee > mailto:atin.mukherje

[Gluster-users] How to maintain HA using NFS clients if the NFS daemon process gets killed on a gluster node?

Hi all, We're getting ready to roll out Gluster using standard NFS from the clients, and CTDB and RRDNS to help facilitate HA. I thought we were good to know, but recently had an issue where there wasn't enough memory on one of the gluster nodes in a test cluster, and OOM killer took out the

Re: [Gluster-users] Determining Connected Client Version

But, why should client version be a problem when brick-log-level is being changed? Regards, Raghavendra On Wed, Jan 27, 2016 at 9:57 AM, Atin Mukherjee wrote: > gluster volume status clients can give you the list of clients > connected. By that would be able to scan through all clients and se

[Gluster-users] Quota issue after upgrading from 3.6.4 to 3.6.6

Hi, After upgrading one of a 3 node glusterfs cluster to 3.6.6, I cannot get the HDD-1 volume to start properly on this server. The other servers are still running 3.6.4 because I don't know if they will all fail after the upgrade. There appears to be an issue with the quota fixed in 3.6.6 (ht

Re: [Gluster-users] Determining Connected Client Version

gluster volume status clients can give you the list of clients connected. By that would be able to scan through all clients and see who is the culprit by checking its op-version. -Atin Sent from one plus one On Jan 27, 2016 8:13 PM, "Marc Eisenbarth" wrote: > Unfortunately, I don't have any of

Re: [Gluster-users] Determining Connected Client Version

Unfortunately, I don't have any of these messages and I'm stuck in a bit of a loop here: $ sudo gluster volume get backup diagnostics.brick-log-level Option Value -- - diagnostics.brick-log-level WARNING $

Re: [Gluster-users] RPM-GPG-KEY-CentOS-SIG-Storage not valid for current packages?

On Wed, Jan 27, 2016 at 01:20:15PM +0100, Jorick Astrego wrote: > > > On 01/27/2016 01:14 PM, Jorick Astrego wrote: > > > > > > On 01/27/2016 12:59 PM, Jorick Astrego wrote: > >> Hi, > >> > >> Currently I'm trying to upgrade our ovirt/glusterfs/katello cluster > >> but I'm having troubles with t

[Gluster-users] Gluster community weekly meeting MoM

Minutes: http://meetbot.fedoraproject.org/gluster-meeting/2016-01-27/gluster_community_weekly_meeting.2016-01-27-12.00.html Minutes (text): http://meetbot.fedoraproject.org/gluster-meeting/2016-01-27/gluster_community_weekly_meeting.2016-01-27-12.00.txt Log: http://meetbot.fedoraproject.org/gluster

Re: [Gluster-users] RPM-GPG-KEY-CentOS-SIG-Storage not valid for current packages?

On 01/27/2016 01:14 PM, Jorick Astrego wrote: > > > On 01/27/2016 12:59 PM, Jorick Astrego wrote: >> Hi, >> >> Currently I'm trying to upgrade our ovirt/glusterfs/katello cluster >> but I'm having troubles with the gpg key. >> >> In Katello I put the GPG key >> (/etc/pki/rpm-gpg/RPM-GPG-KEY-Cen

Re: [Gluster-users] RPM-GPG-KEY-CentOS-SIG-Storage not valid for current packages?

On 01/27/2016 12:59 PM, Jorick Astrego wrote: > Hi, > > Currently I'm trying to upgrade our ovirt/glusterfs/katello cluster > but I'm having troubles with the gpg key. > > In Katello I put the GPG key > (/etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-SIG-Storage) that gets install by > "centos-release-gl

[Gluster-users] RPM-GPG-KEY-CentOS-SIG-Storage not valid for current packages?

Hi, Currently I'm trying to upgrade our ovirt/glusterfs/katello cluster but I'm having troubles with the gpg key. In Katello I put the GPG key (/etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-SIG-Storage) that gets install by "centos-release-gluster36-1.0-3.el7.centos.noarch" But I'm getting: Retrie

Re: [Gluster-users] broken doc link

Yes, as you said, this http://gluster.readthedocs.org/en/latest/Administrator%20Guide/README/ is the right link as we are rendering the admin guide from readthedocs now. --Humble On Wed, Jan 27, 2016 at 3:54 PM, Humble D

Re: [Gluster-users] broken doc link

Hi Mike, Afaict, we are no longer maintaining Mediawiki and in process of porting the contents from it. May be Amye can give an update on this. --Humble On Wed, Jan 27, 2016 at 12:42 PM, Pranith Kumar Karampuri < pkara...@redhat.com> wrote: > > +Humble who maintains this > > On 01/20/2016 05:0

Re: [Gluster-users] Gluster FS in a docker container

Last I looked as far as RHEL is concerned btrfs tech preview.What that means is there may be stability issues on RHEL and or their support team isn't trained in it yet.  So you're experience may not be a good one

Re: [Gluster-users] [Gluster-devel] Gluster Monthly Newsletter, January 2015 Edition

On 22 January 2016 at 22:59, Niels de Vos wrote: > On Mon, Jan 18, 2016 at 07:46:16PM -0800, Amye Scavarda wrote: > > We're kicking off an updated Monthly Newsletter, coming out mid-month. > > We'll highlight special posts, news and noteworthy threads from the > > mailing lists, events, and other

Re: [Gluster-users] [Gluster-devel] GlusterFS FUSE client hangs on rsyncing lots of file

Hi, If the hang appears on enabling client side io-threads then it could be because of some race that is seen when io-threads is enabled on the client side. 2 things will help us debug this issue: 1) thread apply all bt inside gdb (with debuginfo rpms/debs installed ) 2) Complete statedum