Re: Rebalance + VM corruption - current status and request for feedback

Gandalf Corvotempesta <gandalf.corvotempesta@xxxxxxxxx> · Mon, 5 Jun 2017 13:36:30 +0200

Great, thanks! 

Il 5 giu 2017 6:49 AM, "Krutika Dhananjay" <kdhananj@xxxxxxxxxx> ha scritto:
The fixes are already available in 3.10.2, 3.8.12 and 3.11.0

-Krutika

On Sun, Jun 4, 2017 at 5:30 PM, Gandalf Corvotempesta <gandalf.corvotempesta@gmail.com> wrote:
Great news.Is this planned to be published in next release?

Il 29 mag 2017 3:27 PM, "Krutika Dhananjay" <kdhananj@xxxxxxxxxx> ha scritto:
Thanks for that update. Very happy to hear it ran fine without any issues. :)

Yeah so you can ignore those 'No such file or directory' errors. They represent a transient state where DHT in the client process is yet to figure out the new location of the file.

-Krutika

On Mon, May 29, 2017 at 6:51 PM, Mahdi Adnan <mahdi.adnan@xxxxxxxxxxx> wrote:

Hello,

Yes, i forgot to upgrade the client as well. 
I did the upgrade and created a new volume, same options as before, with one VM running and doing lots of IOs. i started the rebalance with force and after it completed the process i rebooted the VM, and it did start normally without issues.
I repeated the process and did another rebalance while the VM running and everything went fine.
But the logs in the client throwing lots of warning messages:

[2017-05-29 13:14:59.416382] W [MSGID: 114031] [client-rpc-fops.c:2928:client3_3_lookup_cbk] 2-gfs_vol2-client-2: remote operation failed. Path: /50294ed6-db7a-418d-965f-9b44c69a83fd/images/d59487fe-f3a9-4bad-a607-3a181c871711/aa01c3a0-5aa0-432d-82ad-d1f515f1d87f
 (93c403f5-c769-44b9-a087-dc51fc21412e) [No such file or directory]
[2017-05-29 13:14:59.416427] W [MSGID: 114031] [client-rpc-fops.c:2928:client3_3_lookup_cbk] 2-gfs_vol2-client-3: remote operation failed. Path: /50294ed6-db7a-418d-965f-9b44c69a83fd/images/d59487fe-f3a9-4bad-a607-3a181c871711/aa01c3a0-5aa0-432d-82ad-d1f515f1d87f
 (93c403f5-c769-44b9-a087-dc51fc21412e) [No such file or directory]
[2017-05-29 13:14:59.808251] W [MSGID: 114031] [client-rpc-fops.c:2928:client3_3_lookup_cbk] 2-gfs_vol2-client-2: remote operation failed. Path: /50294ed6-db7a-418d-965f-9b44c69a83fd/images/d59487fe-f3a9-4bad-a607-3a181c871711/aa01c3a0-5aa0-432d-82ad-d1f515f1d87f
 (93c403f5-c769-44b9-a087-dc51fc21412e) [No such file or directory]
[2017-05-29 13:14:59.808287] W [MSGID: 114031] [client-rpc-fops.c:2928:client3_3_lookup_cbk] 2-gfs_vol2-client-3: remote operation failed. Path: /50294ed6-db7a-418d-965f-9b44c69a83fd/images/d59487fe-f3a9-4bad-a607-3a181c871711/aa01c3a0-5aa0-432d-82ad-d1f515f1d87f
 (93c403f5-c769-44b9-a087-dc51fc21412e) [No such file or directory]

Although the process went smooth, i will run another extensive test tomorrow just to be sure.

-- 

Respectfully

Mahdi A. Mahdi

From: Krutika Dhananjay <kdhananj@xxxxxxxxxx>

Sent: Monday, May 29, 2017 9:20:29 AM

To: Mahdi Adnan

Cc: gluster-user; Gandalf Corvotempesta; Lindsay Mathieson; Kevin Lemonnier

Subject: Re: Rebalance + VM corruption - current status and request for feedback

Hi,

I took a look at your logs.

It very much seems like an issue that is caused by a mismatch in glusterfs client and server packages.

So your client (mount) seems to be still running 3.7.20, as confirmed by the occurrence of the following log message:

[2017-05-26 08:58:23.647458] I [MSGID: 100030] [glusterfsd.c:2338:main] 0-/usr/sbin/glusterfs: Started running /usr/sbin/glusterfs version 3.7.20 (args: /usr/sbin/glusterfs --volfile-server=s1 --volfile-server=s2 --volfile-server=s3 --volfile-server=s4 --volfile-id=/testvol
 /rhev/data-center/mnt/glusterSD/s1:_testvol)

[2017-05-26 08:58:40.901204] I [MSGID: 100030] [glusterfsd.c:2338:main] 0-/usr/sbin/glusterfs: Started running /usr/sbin/glusterfs version 3.7.20 (args: /usr/sbin/glusterfs --volfile-server=s1 --volfile-server=s2 --volfile-server=s3 --volfile-server=s4 --volfile-id=/testvol
 /rhev/data-center/mnt/glusterSD/s1:_testvol)

[2017-05-26 08:58:48.923452] I [MSGID: 100030] [glusterfsd.c:2338:main] 0-/usr/sbin/glusterfs: Started running /usr/sbin/glusterfs version 3.7.20 (args: /usr/sbin/glusterfs --volfile-server=s1 --volfile-server=s2 --volfile-server=s3 --volfile-server=s4 --volfile-id=/testvol
 /rhev/data-center/mnt/glusterSD/s1:_testvol)

whereas the servers have rightly been upgraded to 3.10.2, as seen in rebalance log:

[2017-05-26 09:36:36.075940] I [MSGID: 100030] [glusterfsd.c:2475:main] 0-/usr/sbin/glusterfs: Started running /usr/sbin/glusterfs version 3.10.2 (args: /usr/sbin/glusterfs -s localhost --volfile-id rebalance/testvol --xlator-option *dht.use-readdirp=yes --xlator-option
 *dht.lookup-unhashed=yes --xlator-option *dht.assert-no-child-down=yes --xlator-option *replicate*.data-self-heal=off --xlator-option *replicate*.metadata-self-heal=off --xlator-option *replicate*.entry-self-heal=off --xlator-option *dht.readdir-optimize=on
 --xlator-option *dht.rebalance-cmd=5 --xlator-option *dht.node-uuid=7c0bf49e-1ede-47b1-b9a5-bfde6e60f07b --xlator-option *dht.commit-hash=3376396580 --socket-file /var/run/gluster/gluster-rebalance-801faefa-a583-46b4-8eef-e0ec160da9ea.sock --pid-file /var/lib/glusterd/vols/testvol/rebalance/7c0bf49e-1ede-47b1-b9a5-bfde6e60f07b.pid
 -l /var/log/glusterfs/testvol-rebalance.log)

Could you upgrade all packages to 3.10.2 and try again?

-Krutika

On Fri, May 26, 2017 at 4:46 PM, Mahdi Adnan 
<mahdi.adnan@xxxxxxxxxxx> wrote:

Hi,

Attached are the logs for both the rebalance and the mount.

-- 

Respectfully

Mahdi A. Mahdi

From: Krutika Dhananjay <kdhananj@xxxxxxxxxx>

Sent: Friday, May 26, 2017 1:12:28 PM

To: Mahdi Adnan

Cc: gluster-user; Gandalf Corvotempesta; Lindsay Mathieson; Kevin Lemonnier

Subject: Re: Rebalance + VM corruption - current status and request for feedback

Could you provide the rebalance and mount logs?

-Krutika

On Fri, May 26, 2017 at 3:17 PM, Mahdi Adnan 
<mahdi.adnan@xxxxxxxxxxx> wrote:

Good morning,

So i have tested the new Gluster 3.10.2, and after starting rebalance two VMs were paused due to storage error and third one was not responding.
After rebalance completed i started the VMs and it did not boot, and throw an XFS wrong inode error into the screen.

My setup:
4 nodes running CentOS7.3 with Gluster 3.10.2
4 bricks in distributed replica with group set to virt.
I added the volume to ovirt and created three VMs, i ran a loop to create 5GB file inside the VMs.
Added new 4 bricks to the existing nodes. 
Started rebalane "with force to bypass the warning message"
VMs started to fail after rebalancing.

-- 

Respectfully

Mahdi A. Mahdi

From: Krutika Dhananjay <kdhananj@xxxxxxxxxx>

Sent: Wednesday, May 17, 2017 6:59:20 AM

To: gluster-user

Cc: Gandalf Corvotempesta; Lindsay Mathieson; Kevin Lemonnier; Mahdi Adnan

Subject: Rebalance + VM corruption - current status and request for feedback

Hi,

In the past couple of weeks, we've sent the following fixes concerning VM corruption upon doing rebalance -

https://review.gluster.org/#/q/status:merged+project:glusterfs+branch:master+topic:bug-1440051

These fixes are very much part of the latest 3.10.2 release.

Satheesaran within Red Hat also verified that they work and he's not seeing corruption issues anymore.

I'd like to hear feedback from the users themselves on these fixes (on your test environments to begin with) before even changing the status of the bug to CLOSED.

Although 3.10.2 has a patch that prevents rebalance sub-commands from being executed on sharded volumes, you can override the check by using the 'force' option.

For example,

# gluster volume rebalance myvol start force

Very much looking forward to hearing from you all.

Thanks,

Krutika

_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
http://lists.gluster.org/mailman/listinfo/gluster-users