Re: Replicated striped data lose

[Mahdi Adnan <mahdi.adnan@xxxxxxxxxxxxxxxxx>]

and here's the log of the ESXi;

2016-03-15T12:50:17.982Z cpu41:33260)WARNING: NFS: 4566: Short read for object b00f 60 6b5d5087 c5d851fc 4c474f3a 8efd48b3 9d4617b1 a556c0bc 30c9df6b 16c12f10 250219b3f4146a7 431899d1 0 431200000000 offset: 0x0 requested: 0x200 read: 0x95
2016-03-15T12:50:17.983Z cpu41:33260)WARNING: NFS: 4566: Short read for object b00f 60 6b5d5087 c5d851fc 4c474f3a 8efd48b3 9d4617b1 a556c0bc 30c9df6b 16c12f10 250219b3f4146a7 431899d1 0 431200000000 offset: 0x0 requested: 0x200 read: 0x95
2016-03-15T12:50:17.984Z cpu41:33260)WARNING: NFS: 4566: Short read for object b00f 60 6b5d5087 c5d851fc 4c474f3a 8efd48b3 9d4617b1 a556c0bc 30c9df6b 16c12f10 250219b3f4146a7 431899d1 0 431200000000 offset: 0x0 requested: 0x200 read: 0x95
2016-03-15T12:50:17.995Z cpu41:35990)WARNING: NFS: 4566: Short read for object b00f 60 6b5d5087 c5d851fc 4c474f3a 8efd48b3 9d4617b1 a556c0bc 30c9df6b 16c12f10 250219b3f4146a7 431899d1 0 431200000000 offset: 0x0 requested: 0x200 read: 0x95
2016-03-15T12:50:18.000Z cpu41:35990)WARNING: NFS: 4566: Short read for object b00f 60 6b5d5087 c5d851fc 4c474f3a 8efd48b3 9d4617b1 a556c0bc 30c9df6b 16c12f10 250219b3f4146a7 431899d1 0 431200000000 offset: 0x0 requested: 0x200 read: 0x95
2016-03-15T12:50:18.031Z cpu41:35990)WARNING: NFS: 4566: Short read for object b00f 60 6b5d5087 c5d851fc 4c474f3a 8efd48b3 9d4617b1 a556c0bc 30c9df6b 16c12f10 250219b3f4146a7 431899d1 0 431200000000 offset: 0x0 requested: 0x200 read: 0x95
2016-03-15T12:50:18.032Z cpu41:35990)WARNING: NFS: 4566: Short read for object b00f 60 6b5d5087 c5d851fc 4c474f3a 8efd48b3 9d4617b1 a556c0bc 30c9df6b 16c12f10 250219b3f4146a7 431899d1 0 431200000000 offset: 0x0 requested: 0x200 read: 0x95
2016-03-15T12:50:18.032Z cpu41:35990)WARNING: NFS: 4566: Short read for object b00f 60 6b5d5087 c5d851fc 4c474f3a 8efd48b3 9d4617b1 a556c0bc 30c9df6b 16c12f10 250219b3f4146a7 431899d1 0 431200000000 offset: 0x0 requested: 0x200 read: 0x95
2016-03-15T12:50:18.043Z cpu41:35990)WARNING: NFS: 4566: Short read for object b00f 60 6b5d5087 c5d851fc 4c474f3a 8efd48b3 9d4617b1 a556c0bc 30c9df6b 16c12f10 250219b3f4146a7 431899d1 0 431200000000 offset: 0x0 requested: 0x200 read: 0x95
2016-03-15T12:50:18.048Z cpu41:35990)WARNING: NFS: 4566: Short read for object b00f 60 6b5d5087 c5d851fc 4c474f3a 8efd48b3 9d4617b1 a556c0bc 30c9df6b 16c12f10 250219b3f4146a7 431899d1 0 431200000000 offset: 0x0 requested: 0x200 read: 0x95

Respectfully
Mahdi A. Mahdi

Skype: mahdi.adnan@xxxxxxxxxxx

On 03/15/2016 03:06 PM, Mahdi Adnan wrote:

[2016-03-15 14:12:01.421615] I [MSGID: 109036] [dht-common.c:8043:dht_log_new_layout_for_dir_selfheal] 0-v-dht: Setting layout of /New Virtual Machine_2 with [Subvol_name: v-replicate-0, Err: -1 , Start: 0 , Stop: 1431655764 , Hash: 1 ], [Subvol_name: v-replicate-1, Err: -1 , Start: 1431655765 , Stop: 2863311529 , Hash: 1 ], [Subvol_name: v-replicate-2, Err: -1 , Start: 2863311530 , Stop: 4294967295 , Hash: 1 ],
[2016-03-15 14:12:02.001167] I [MSGID: 109066] [dht-rename.c:1413:dht_rename] 0-v-dht: renaming /New Virtual Machine_2/New Virtual Machine.vmdk~ (hash=v-replicate-2/cache=v-replicate-2) => /New Virtual Machine_2/New Virtual Machine.vmdk (hash=v-replicate-2/cache=v-replicate-2)
[2016-03-15 14:12:02.248164] W [MSGID: 112032] [nfs3.c:3622:nfs3svc_rmdir_cbk] 0-nfs: 3fed7d9f: /New Virtual Machine_2 => -1 (Directory not empty) [Directory not empty]
[2016-03-15 14:12:02.259015] W [MSGID: 112032] [nfs3.c:3622:nfs3svc_rmdir_cbk] 0-nfs: 3fed7da3: /New Virtual Machine_2 => -1 (Directory not empty) [Directory not empty]

Respectfully
Mahdi A. Mahdi

On 03/15/2016 03:03 PM, Krutika Dhananjay wrote:

Hmm ok. Could you share the nfs.log content?

-Krutika

On Tue, Mar 15, 2016 at 1:45 PM, Mahdi Adnan <mahdi.adnan@xxxxxxxxxxxxxxxxx> wrote:

Okay, here's what i did;

Volume Name: v
Type: Distributed-Replicate
Volume ID: b348fd8e-b117-469d-bcc0-56a56bdfc930
Status: Started
Number of Bricks: 3 x 2 = 6
Transport-type: tcp
Bricks:
Brick1: gfs001:/bricks/b001/v
Brick2: gfs001:/bricks/b002/v
Brick3: gfs001:/bricks/b003/v
Brick4: gfs002:/bricks/b004/v
Brick5: gfs002:/bricks/b005/v
Brick6: gfs002:/bricks/b006/v
Options Reconfigured:
features.shard-block-size: 128MB
features.shard: enable
cluster.server-quorum-type: server
cluster.quorum-type: auto
network.remote-dio: enable
cluster.eager-lock: enable
performance.stat-prefetch: off
performance.io-cache: off
performance.read-ahead: off
performance.quick-read: off
performance.readdir-ahead: on


same error.
and still mounting using glusterfs will work just fine.

Respectfully
Mahdi A. Mahdi

On 03/15/2016 11:04 AM, Krutika Dhananjay wrote:

OK but what if you use it with replication? Do you still see the error? I think not.

Could you give it a try and tell me what you find?

-Krutika

On Tue, Mar 15, 2016 at 1:23 PM, Mahdi Adnan <mahdi.adnan@xxxxxxxxxxxxxxxxx> wrote:

Hi,

I have created the following volume;

Volume Name: v
Type: Distribute
Volume ID: 90de6430-7f83-4eda-a98f-ad1fabcf1043
Status: Started
Number of Bricks: 3
Transport-type: tcp
Bricks:
Brick1: gfs001:/bricks/b001/v
Brick2: gfs001:/bricks/b002/v
Brick3: gfs001:/bricks/b003/v
Options Reconfigured:
features.shard-block-size: 128MB
features.shard: enable
cluster.server-quorum-type: server
cluster.quorum-type: auto
network.remote-dio: enable
cluster.eager-lock: enable
performance.stat-prefetch: off
performance.io-cache: off
performance.read-ahead: off
performance.quick-read: off
performance.readdir-ahead: on

and after mounting it in ESXi and trying to clone a VM to it, i got the same error.

Respectfully
Mahdi A. Mahdi

On 03/15/2016 10:44 AM, Krutika Dhananjay wrote:

Hi,

Do not use sharding and stripe together in the same volume because

a) It is not recommended and there is no point in using both. Using sharding alone on your volume should work fine.

b) Nobody tested it.

c) Like Niels said, stripe feature is virtually deprecated.

I would suggest that you create an nx3 volume where n is the number of distribute subvols you prefer, enable group virt options on it, and enable sharding on it,
set the shard-block-size that you feel appropriate and then just start off with VM image creation etc.

If you run into any issues even after you do this, let us know and we'll help you out.

-Krutika 

On Tue, Mar 15, 2016 at 1:07 PM, Mahdi Adnan <mahdi.adnan@xxxxxxxxxxxxxxxxx> wrote:

Thanks Krutika,

I have deleted the volume and created a new one.
I found that it may be an issue with the NFS itself, i have created a new striped volume and enabled sharding and mounted it via glusterfs and it worked just fine, if i mount it with nfs it will fail and gives me the same errors.

Respectfully
Mahdi A. Mahdi

On 03/15/2016 06:24 AM, Krutika Dhananjay wrote:

Hi,

So could you share the xattrs associated with the file at <BRICK_PATH>/.glusterfs/c3/e8/c3e88cc1-7e0a-4d46-9685-2d12131a5e1c

Here's what you need to execute:

# getfattr -d -m . -e hex /mnt/b1/v/.glusterfs/c3/e8/c3e88cc1-7e0a-4d46-9685-2d12131a5e1c      on the first node and

# getfattr -d -m . -e hex /mnt/b2/v/.glusterfs/c3/e8/c3e88cc1-7e0a-4d46-9685-2d12131a5e1c      on the second.

Also, it is normally advised to use a replica 3 volume as opposed to replica 2 volume to guard against split-brains.

-Krutika

On Mon, Mar 14, 2016 at 3:17 PM, Mahdi Adnan <mahdi.adnan@xxxxxxxxxxxxxxxxx> wrote:

sorry for serial posting but, i got new logs it might help..

the message appear during the migration;

/var/log/glusterfs/nfs.log


[2016-03-14 09:45:04.573765] I [MSGID: 109036] [dht-common.c:8043:dht_log_new_layout_for_dir_selfheal] 0-testv-dht: Setting layout of /New Virtual Machine_1 with [Subvol_name: testv-stripe-0, Err: -1 , Start: 0 , Stop: 4294967295 , Hash: 1 ],
[2016-03-14 09:45:04.957499] E [shard.c:369:shard_modify_size_and_block_count] (-->/usr/lib64/glusterfs/3.7.8/xlator/cluster/distribute.so(dht_file_setattr_cbk+0x14f) [0x7f27a13c067f] -->/usr/lib64/glusterfs/3.7.8/xlator/features/shard.so(shard_common_setattr_cbk+0xcc) [0x7f27a116681c] -->/usr/lib64/glusterfs/3.7.8/xlator/features/shard.so(shard_modify_size_and_block_count+0xdd) [0x7f27a116584d] ) 0-testv-shard: Failed to get trusted.glusterfs.shard.file-size for c3e88cc1-7e0a-4d46-9685-2d12131a5e1c
[2016-03-14 09:45:04.957577] W [MSGID: 112199] [nfs3-helpers.c:3418:nfs3_log_common_res] 0-nfs-nfsv3: /New Virtual Machine_1/New Virtual Machine-flat.vmdk => (XID: 3fec5a26, SETATTR: NFS: 22(Invalid argument for operation), POSIX: 22(Invalid argument)) [Invalid argument]
[2016-03-14 09:45:05.079657] E [MSGID: 112069] [nfs3.c:3649:nfs3_rmdir_resume] 0-nfs-nfsv3: No such file or directory: (192.168.221.52:826) testv : 00000000-0000-0000-0000-000000000001

Respectfully
Mahdi A. Mahd

On 03/14/2016 11:14 AM, Mahdi Adnan wrote:

So i have deployed a new server "Cisco UCS C220M4" and created a new volume;

Volume Name: testv
Type: Stripe
Volume ID: 55cdac79-fe87-4f1f-90c0-15c9100fe00b
Status: Started
Number of Bricks: 1 x 2 = 2
Transport-type: tcp
Bricks:
Brick1: 10.70.0.250:/mnt/b1/v
Brick2: 10.70.0.250:/mnt/b2/v
Options Reconfigured:
nfs.disable: off
features.shard-block-size: 64MB
features.shard: enable
cluster.server-quorum-type: server
cluster.quorum-type: auto
network.remote-dio: enable
cluster.eager-lock: enable
performance.stat-prefetch: off
performance.io-cache: off
performance.read-ahead: off
performance.quick-read: off
performance.readdir-ahead: off

same error ..

can anyone share with me the info of a working striped volume ?

On 03/14/2016 09:02 AM, Mahdi Adnan wrote:

I have a pool of two bricks in the same server;

Volume Name: k
Type: Stripe
Volume ID: 1e9281ce-2a8b-44e8-a0c6-e3ebf7416b2b
Status: Started
Number of Bricks: 1 x 2 = 2
Transport-type: tcp
Bricks:
Brick1: gfs001:/bricks/t1/k
Brick2: gfs001:/bricks/t2/k
Options Reconfigured:
features.shard-block-size: 64MB
features.shard: on
cluster.server-quorum-type: server
cluster.quorum-type: auto
network.remote-dio: enable
cluster.eager-lock: enable
performance.stat-prefetch: off
performance.io-cache: off
performance.read-ahead: off
performance.quick-read: off
performance.readdir-ahead: off

same issue ...
glusterfs 3.7.8 built on Mar 10 2016 20:20:45.

Respectfully
Mahdi A. Mahdi

Systems Administrator
IT. Department
Earthlink Telecommunications

Cell: 07903316180
Work: 3352
Skype: mahdi.adnan@xxxxxxxxxxx

On 03/14/2016 08:11 AM, Niels de Vos wrote:

On Mon, Mar 14, 2016 at 08:12:27AM +0530, Krutika Dhananjay wrote:

It would be better to use sharding over stripe for your vm use case. It
offers better distribution and utilisation of bricks and better heal
performance.
And it is well tested.

Basically the "striping" feature is deprecated, "sharding" is its
improved replacement. I expect to see "striping" completely dropped in
the next major release.

Niels

Couple of things to note before you do that:
1. Most of the bug fixes in sharding have gone into 3.7.8. So it is advised
that you use 3.7.8 or above.
2. When you enable sharding on a volume, already existing files in the
volume do not get sharded. Only the files that are newly created from the
time sharding is enabled will.
    If you do want to shard the existing files, then you would need to cp
them to a temp name within the volume, and then rename them back to the
original file name.

HTH,
Krutika

On Sun, Mar 13, 2016 at 11:49 PM, Mahdi Adnan <mahdi.adnan@xxxxxxxxxxxxxxxxx

wrote:

I couldn't find anything related to cache in the HBAs.
what logs are useful in my case ? i see only bricks logs which contains
nothing during the failure.

###
[2016-03-13 18:05:19.728614] E [MSGID: 113022] [posix.c:1232:posix_mknod]
0-vmware-posix: mknod on
/bricks/b003/vmware/.shard/17d75e20-16f1-405e-9fa5-99ee7b1bd7f1.511 failed
[File exists]
[2016-03-13 18:07:23.337086] E [MSGID: 113022] [posix.c:1232:posix_mknod]
0-vmware-posix: mknod on
/bricks/b003/vmware/.shard/eef2d538-8eee-4e58-bc88-fbf7dc03b263.4095 failed
[File exists]
[2016-03-13 18:07:55.027600] W [trash.c:1922:trash_rmdir] 0-vmware-trash:
rmdir issued on /.trashcan/, which is not permitted
[2016-03-13 18:07:55.027635] I [MSGID: 115056]
[server-rpc-fops.c:459:server_rmdir_cbk] 0-vmware-server: 41987: RMDIR
/.trashcan/internal_op (00000000-0000-0000-0000-000000000005/internal_op)
==> (Operation not permitted) [Operation not permitted]
[2016-03-13 18:11:34.353441] I [login.c:81:gf_auth] 0-auth/login: allowed
user names: c0c72c37-477a-49a5-a305-3372c1c2f2b4
[2016-03-13 18:11:34.353463] I [MSGID: 115029]
[server-handshake.c:612:server_setvolume] 0-vmware-server: accepted client
from gfs002-2727-2016/03/13-20:17:43:613597-vmware-client-4-0-0 (version:
3.7.8)
[2016-03-13 18:11:34.591139] I [login.c:81:gf_auth] 0-auth/login: allowed
user names: c0c72c37-477a-49a5-a305-3372c1c2f2b4
[2016-03-13 18:11:34.591173] I [MSGID: 115029]
[server-handshake.c:612:server_setvolume] 0-vmware-server: accepted client
from gfs002-2719-2016/03/13-20:17:42:609388-vmware-client-4-0-0 (version:
3.7.8)
###

ESXi just keeps telling me "Cannot clone T: The virtual disk is either
corrupted or not a supported format.
error
3/13/2016 9:06:20 PM
Clone virtual machine
T
VCENTER.LOCAL\Administrator
"

My setup is 2 servers with a floating ip controlled by CTDB and my ESXi
server mount the NFS via the floating ip.





On 03/13/2016 08:40 PM, pkoelle wrote:

Am 13.03.2016 um 18:22 schrieb David Gossage:

On Sun, Mar 13, 2016 at 11:07 AM, Mahdi Adnan <
mahdi.adnan@xxxxxxxxxxxxxxxxx

wrote:

My HBAs are LSISAS1068E, and the filesystem is XFS.

I tried EXT4 and it did not help.
I have created a stripted volume in one server with two bricks, same
issue.
and i tried a replicated volume with just "sharding enabled" same issue,
as soon as i disable the sharding it works just fine, niether sharding
nor
striping works for me.
i did follow up with some of threads in the mailing list and tried some
of
the fixes that worked with the others, none worked for me. :(

Is it possible the LSI has write-cache enabled?

Why is that relevant? Even the backing filesystem has no idea if there is
a RAID or write cache or whatever. There are blocks and sync(), end of
story.
If you lose power and screw up your recovery OR do funky stuff with SAS
multipathing that might be an issue with a controller cache. AFAIK thats
not what we are talking about.

I'm afraid but unless the OP has some logs from the server, a
reproducible testcase or a backtrace from client or server this isn't
getting us anywhere.

cheers
Paul

On 03/13/2016 06:54 PM, David Gossage wrote:

On Sun, Mar 13, 2016 at 8:16 AM, Mahdi Adnan <
mahdi.adnan@xxxxxxxxxxxxxxxxx> wrote:

Okay so i have enabled shard in my test volume and it did not help,

stupidly enough, i have enabled it in a production volume
"Distributed-Replicate" and it currpted  half of my VMs.
I have updated Gluster to the latest and nothing seems to be changed in
my situation.
below the info of my volume;

I was pointing at the settings in that email as an example for
corruption
fixing. I wouldn't recommend enabling sharding if you haven't gotten the
base working yet on that cluster. What HBA's are you using and what is
layout of filesystem for bricks?


Number of Bricks: 3 x 2 = 6

Transport-type: tcp
Bricks:
Brick1: gfs001:/bricks/b001/vmware
Brick2: gfs002:/bricks/b004/vmware
Brick3: gfs001:/bricks/b002/vmware
Brick4: gfs002:/bricks/b005/vmware
Brick5: gfs001:/bricks/b003/vmware
Brick6: gfs002:/bricks/b006/vmware
Options Reconfigured:
performance.strict-write-ordering: on
cluster.server-quorum-type: server
cluster.quorum-type: auto
network.remote-dio: enable
performance.stat-prefetch: disable
performance.io-cache: off
performance.read-ahead: off
performance.quick-read: off
cluster.eager-lock: enable
features.shard-block-size: 16MB
features.shard: on
performance.readdir-ahead: off


On 03/12/2016 08:11 PM, David Gossage wrote:


On Sat, Mar 12, 2016 at 10:21 AM, Mahdi Adnan <
<mahdi.adnan@xxxxxxxxxxxxxxxxx>mahdi.adnan@xxxxxxxxxxxxxxxxx> wrote:

Both servers have HBA no RAIDs and i can setup a replicated or

dispensers without any issues.
Logs are clean and when i tried to migrate a vm and got the error,
nothing showed up in the logs.
i tried mounting the volume into my laptop and it mounted fine but,
if i
use dd to create a data file it just hang and i cant cancel it, and i
cant
unmount it or anything, i just have to reboot.
The same servers have another volume on other bricks in a distributed
replicas, works fine.
I have even tried the same setup in a virtual environment (created two
vms and install gluster and created a replicated striped) and again
same
thing, data corruption.

I'd look through mail archives for a topic "Shard in Production" I
think
it's called.  The shard portion may not be relevant but it does discuss
certain settings that had to be applied with regards to avoiding
corruption
with VM's.  You may want to try and disable the
performance.readdir-ahead
also.

On 03/12/2016 07:02 PM, David Gossage wrote:



On Sat, Mar 12, 2016 at 9:51 AM, Mahdi Adnan <
<mahdi.adnan@xxxxxxxxxxxxxxxxx>mahdi.adnan@xxxxxxxxxxxxxxxxx> wrote:

Thanks David,

My settings are all defaults, i have just created the pool and
started
it.
I have set the settings as your recommendation and it seems to be the
same issue;

Type: Striped-Replicate
Volume ID: 44adfd8c-2ed1-4aa5-b256-d12b64f7fc14
Status: Started
Number of Bricks: 1 x 2 x 2 = 4
Transport-type: tcp
Bricks:
Brick1: gfs001:/bricks/t1/s
Brick2: gfs002:/bricks/t1/s
Brick3: gfs001:/bricks/t2/s
Brick4: gfs002:/bricks/t2/s
Options Reconfigured:
performance.stat-prefetch: off
network.remote-dio: on
cluster.eager-lock: enable
performance.io-cache: off
performance.read-ahead: off
performance.quick-read: off
performance.readdir-ahead: on

Is their a raid controller perhaps doing any caching?

In the gluster logs any errors being reported during migration
process?
Since they aren't in use yet have you tested making just mirrored
bricks
using different pairings of servers two at a time to see if problem
follows
certain machine or network ports?

On 03/12/2016 03:25 PM, David Gossage wrote:



On Sat, Mar 12, 2016 at 1:55 AM, Mahdi Adnan <
<mahdi.adnan@xxxxxxxxxxxxxxxxx>mahdi.adnan@xxxxxxxxxxxxxxxxx> wrote:

Dears,

I have created a replicated striped volume with two bricks and two
servers but I can't use it because when I mount it in ESXi and try
to
migrate a VM to it, the data get corrupted.
Is any one have any idea why is this happening ?

Dell 2950 x2
Seagate 15k 600GB
CentOS 7.2
Gluster 3.7.8

Appreciate your help.

Most reports of this I have seen end up being settings related.  Post
gluster volume info. Below is what I have seen as most common
recommended
settings.
I'd hazard a guess you may have some the read ahead cache or prefetch
on.

quick-read=off
read-ahead=off
io-cache=off
stat-prefetch=off
eager-lock=enable
remote-dio=on

Mahdi Adnan
System Admin


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

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

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

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

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

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

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

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

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

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