similar to: GlusterFS as virtual machine storage

If your VMs use ext4 also check this: https://joejulian.name/blog/keeping-your-vms-from-going- read-only-when-encountering-a-ping-timeout-in-glusterfs/ I asked him what to do for VMs using XFS and he said he could not find a fix (setting to change) for those. HTH, Diego On Sep 8, 2017 6:19 AM, "Diego Remolina" <dijuremo at gmail.com> wrote: > The issue of I/O stopping may

This is exactly the problem, Systemctl stop glusterd does *not* kill the brick processes. On CentOS with gluster 3.10.x there is also a service, meant to only stop glusterfsd (brick processes). I think the reboot process may not be properly stopping glusterfsd or network or firewall may be stopped before glusterfsd and so the nodes go into the long timeout. Once again , in my case a simple

This is the qemu log of instance: [2017-09-08 09:31:48.381077] C [rpc-clnt-ping.c:160:rpc_clnt_ping_timer_expired] 0-gv_openstack_1-client-1: server 10.0.1.202:49152 has not responded in the last 1 seconds, disconnecting. [2017-09-08 09:31:48.382411] E [rpc-clnt.c:365:saved_frames_unwind] (--> /lib64/libglusterfs.so.0(_gf_log_callingfn+0x13b)[0x7fbcadb09e8b] (-->

Hi Diego, indeed glusterfsd processes are runnin and it is the reason I do server reboot instead of systemctl glusterd stop. Is killall different from reboot in a way glusterfsd processes are terminated in CentOS (init 1?)? However I will try this and let you know. -ps On Fri, Sep 8, 2017 at 12:19 PM, Diego Remolina <dijuremo at gmail.com> wrote: > The issue of I/O stopping may also

The issue of I/O stopping may also be with glusterfsd not being properly killed before rebooting the server. For example in RHEL 7.4 with official Gluster 3.8.4, the glusterd service does *not* stop glusterfsd when you run systemctl stop glusterd So give this a try on the nose you wish to reboot: 1. Stop glusterd 2. Check if glusterfsd processes are still running. If they are, use: killall

Oh, you really don't want to go below 30s, I was told. I'm using 30 seconds for the timeout, and indeed when a node goes down the VM freez for 30 seconds, but I've never seen them go read only for that. I _only_ use virtio though, maybe it's that. What are you using ? On Fri, Sep 08, 2017 at 11:41:13AM +0200, Pavel Szalbot wrote: > Back to replica 3 w/o arbiter. Two fio jobs

Hello, I'm having problems when write-behind is enabled on Gluster 3.8.4. I have 2 Gluster servers each with a single brick that is mirrored between them. The code causing these issues reads two data files each approx. 128G in size. It opens a third file, mmap()'s that file, and subsequently reads and writes to it. The third file, on sucessful runs (without write-behind enabled)

Hi Ravi, back to our client-cannot-reconnect-to-gluster-brick problem ... > Von: Ravishankar N [ravishankar at redhat.com] > Gesendet: Montag, 29. Mai 2017 06:34 > An: Markus Stockhausen; gluster-users at gluster.org > Betreff: Re: [Gluster-users] gluster heal entry reappears > > > On 05/28/2017 10:31 PM, Markus Stockhausen wrote: > > Hi, > > > > I'm

When executing the load test with the FIO tool, execute the following job from the client When executed, the load of 2 cores is high for the CPU. Up to 100%. At that time, if another client is performing NFS mounting, the df command I can not connect NFS without coming back. The log will continue to be output below. I believe that if the CPU utilization is distributed, the load will be eliminated.

Good morning. We have a 6 node cluster. 3 nodes are participating in a replica 3 volume. Naming convention: xx01 - 3 nodes participating in ovirt_vol xx02 - 3 nodes NOT particpating in ovirt_vol Last week, restarted glusterd on each node in cluster to update (one at a time). The three xx01 nodes all show the following in glusterd.log: [2018-02-26 14:31:47.330670] E

On Mon, 3 Jul 2017 at 12:28, Pawan Alwandi <pawan at platform.sh> wrote: > Hello Atin, > > I've gotten around to this and was able to get upgrade done using 3.7.0 > before moving to 3.11. For some reason 3.7.9 wasn't working well. > > On 3.11 though I notice that gluster/nfs is really made optional and > nfs-ganesha is being recommended. We have plans to

Sorry for big attachment in previous mail...last 1000 lines of those logs attached now. On Mon, May 29, 2017 at 4:44 PM, Pawan Alwandi <pawan at platform.sh> wrote: > > > On Thu, May 25, 2017 at 9:54 PM, Atin Mukherjee <amukherj at redhat.com> > wrote: > >> >> On Thu, 25 May 2017 at 19:11, Pawan Alwandi <pawan at platform.sh> wrote: >>

Hello Atin, I've gotten around to this and was able to get upgrade done using 3.7.0 before moving to 3.11. For some reason 3.7.9 wasn't working well. On 3.11 though I notice that gluster/nfs is really made optional and nfs-ganesha is being recommended. We have plans to switch to nfs-ganesha on new clusters but would like to have glusterfs-gnfs on existing clusters so a seamless upgrade

Hi, I've got troubles after few minutes of glusterfs operations. I setup a 4-node replica 4 storage, with 2 bricks on every server: # gluster volume create vms replica 4 transport tcp 192.168.7.1:/srv/vol1 192.168.7.2:/srv/vol1 192.168.7.3:/srv/vol1 192.168.7.4:/srv/vol1 192.168.7.1:/srv/vol2 192.168.7.2:/srv/vol2 192.168.7.3:/srv/vol2 192.168.7.4:/srv/vol2 I started copying files with

I'm starting a new thread on this, because I have more concrete information than I did the first time around. The full rebalance log from the machine where I started the rebalance can be found at the following link. It is slightly redacted - one search/replace was made to replace an identifying word with REDACTED. https://dl.dropboxusercontent.com/u/97770508/mdfs-rebalance-redacted.zip

On 4/11/2018 11:54 AM, Alex K wrote: Hey Guy's, Returning to this topic, after disabling the the quorum: cluster.quorum-type: none cluster.server-quorum-type: none I've ran into a number of gluster errors (see below). I'm using gluster as the backend for my NFS storage. I have gluster running on two nodes, nfs01 and nfs02. It's mounted on /n on each host. The path /n is

I've run into a problem with Gluster stability with the RDMA transport. Below is a description of the environment, a simple script that can replicate the problem, and log files from my test system. I can work around the problem by using the TCP transport over IPoIB but would like some input onto what may be making the RDMA transport fail in this case. ===== Symptoms ===== - Error from test

Hi, I got the same problem as Juergen, My volume is a simple replicated volume with 2 host and GlusterFS 3.2.0 Volume Name: poolsave Type: Replicate Status: Started Number of Bricks: 2 Transport-type: tcp Bricks: Brick1: ylal2950:/soft/gluster-data Brick2: ylal2960:/soft/gluster-data Options Reconfigured: diagnostics.brick-log-level: DEBUG network.ping-timeout: 20 performance.cache-size: 512MB

Hi, I'm having a similar issue, were you able to solve it? Thanks. Hey all, I've got a strange problem going on here. I've installed glusterfs-server on ubuntu 16.04: glusterfs-client/xenial,now 3.7.6-1ubuntu1 amd64 [installed,automatic] glusterfs-common/xenial,now 3.7.6-1ubuntu1 amd64 [installed,automatic] glusterfs-server/xenial,now 3.7.6-1ubuntu1 amd64 [installed] I can

On Wed, Apr 11, 2018 at 4:35 AM, TomK <tomkcpr at mdevsys.com> wrote: > On 4/9/2018 2:45 AM, Alex K wrote: > Hey Alex, > > With two nodes, the setup works but both sides go down when one node is > missing. Still I set the below two params to none and that solved my issue: > > cluster.quorum-type: none > cluster.server-quorum-type: none > > yes this disables