Re: Failures during rebalance on gluster distributed disperse volume

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 


Hi Sunil,

many thanks to you too.
I will follow your suggestions and the guide for upgrading to 3.12

Crossing fingers :-)
Regards,
Mauro

Il giorno 15 set 2018, alle ore 11:57, Sunil Kumar Heggodu Gopala Acharya <sheggodu@xxxxxxxxxx> ha scritto:

Hi Mauro,

As Nithya highlighted FALLOCATE support for EC volumes went in 3.11 as part of https://bugzilla.redhat.com/show_bug.cgi?id=1454686. Hence, upgrading to 3.12 as suggested before would be a right move.

Here is the documentation for upgrading to 3.12: https://docs.gluster.org/en/latest/Upgrade-Guide/upgrade_to_3.12/

Regards,
Sunil kumar Acharya

Senior Software Engineer

T: +91-8067935170    



On Sat, Sep 15, 2018 at 3:42 AM, Mauro Tridici <mauro.tridici@xxxxxxx> wrote:

Hi Nithya,

thank you very much for your answer.
I will wait for @Sunil opinion too before starting the upgrade procedure.

Since it will be the first upgrade of our Gluster cluster, I would like to know if it could be a “virtually dangerous" procedure and if it will be the risk of losing data :-) 
Unfortunately, I can’t do a preventive copy of the volume data in another location.
If it is possible, could you please illustrate the right steps needed to complete the upgrade procedure from the 3.10.5 to the 3.12 version?

Thank you again, Nithya.
Thank you to all of you for the help!

Regards,
Mauro

Il giorno 14 set 2018, alle ore 16:59, Nithya Balachandran <nbalacha@xxxxxxxxxx> ha scritto:

Hi Mauro,


The rebalance code started using fallocate in 3.10.5 (https://bugzilla.redhat.com/show_bug.cgi?id=1473132) which works fine on replicated volumes. However, we neglected to test this with EC volumes on 3.10. Once we discovered the issue, the EC fallocate implementation was made available in 3.11.

At this point, I'm afraid the only option I see is to upgrade to at least 3.12.

@Sunil, do you have anything to add?

Regards,
Nithya

On 13 September 2018 at 18:34, Mauro Tridici <mauro.tridici@xxxxxxx> wrote:

Hi Nithya,

thank you for involving EC group.
I will wait for your suggestions.

Regards,
Mauro

Il giorno 13 set 2018, alle ore 13:38, Nithya Balachandran <nbalacha@xxxxxxxxxx> ha scritto:

This looks like an issue because rebalance switched to using fallocate which EC did not have implemented at that point.

@Pranith, @Ashish, which version of gluster had support for fallocate in EC?


Regards,
Nithya

On 12 September 2018 at 19:24, Mauro Tridici <mauro.tridici@xxxxxxx> wrote:
Dear All,

I recently added 3 servers (each one with 12 bricks) to an existing Gluster Distributed Disperse Volume.
Volume extension has been completed without error and I already executed the rebalance procedure with fix-layout option with no problem.
I just launched the rebalance procedure without fix-layout option, but, as you can see in the output below, I noticed that some failures have been detected.

[root@s01 glusterfs]# gluster v rebalance tier2 status
                                    Node Rebalanced-files          size       scanned      failures       skipped               status  run time in h:m:s
                               ---------      -----------   -----------   -----------   -----------   -----------         ------------     --------------
                               localhost            71176         3.2MB       2137557       1530391          8128          in progress       13:59:05
                                 s02-stg                0        0Bytes             0             0             0            completed       11:53:28
                                 s03-stg                0        0Bytes             0             0             0            completed       11:53:32
                                 s04-stg                0        0Bytes             0             0             0            completed        0:00:06
                                 s05-stg               15        0Bytes         17055             0            18            completed       10:48:01
                                 s06-stg                0        0Bytes             0             0             0            completed        0:00:06
Estimated time left for rebalance to complete :        0:46:53
volume rebalance: tier2: success

In the volume rebalance log file, I detected a lot of error messages similar to the following ones:

[2018-09-12 13:15:50.756703] E [MSGID: 0] [dht-rebalance.c:1696:dht_migrate_file] 0-tier2-dht: Create dst failed on - tier2-disperse-6 for file - /CSP/sp1/CESM/archive/sps_200508_003/atm/hist/postproc/sps_200508_003.cam.h0.2005-12_grid.nc
[2018-09-12 13:15:50.757025] E [MSGID: 109023] [dht-rebalance.c:2733:gf_defrag_migrate_single_file] 0-tier2-dht: migrate-data failed for /CSP/sp1/CESM/archive/sps_200508_003/atm/hist/postproc/sps_200508_003.cam.h0.2005-12_grid.nc
[2018-09-12 13:15:50.759183] E [MSGID: 109023] [dht-rebalance.c:844:__dht_rebalance_create_dst_file] 0-tier2-dht: fallocate failed for /CSP/sp1/CESM/archive/sps_200508_003/atm/hist/postproc/sps_200508_003.cam.h0.2005-09_grid.nc on tier2-disperse-9 (Operation not supported)
[2018-09-12 13:15:50.759206] E [MSGID: 0] [dht-rebalance.c:1696:dht_migrate_file] 0-tier2-dht: Create dst failed on - tier2-disperse-9 for file - /CSP/sp1/CESM/archive/sps_200508_003/atm/hist/postproc/sps_200508_003.cam.h0.2005-09_grid.nc
[2018-09-12 13:15:50.759536] E [MSGID: 109023] [dht-rebalance.c:2733:gf_defrag_migrate_single_file] 0-tier2-dht: migrate-data failed for /CSP/sp1/CESM/archive/sps_200508_003/atm/hist/postproc/sps_200508_003.cam.h0.2005-09_grid.nc
[2018-09-12 13:15:50.777219] E [MSGID: 109023] [dht-rebalance.c:844:__dht_rebalance_create_dst_file] 0-tier2-dht: fallocate failed for /CSP/sp1/CESM/archive/sps_200508_003/atm/hist/postproc/sps_200508_003.cam.h0.2006-01_grid.nc on tier2-disperse-10 (Operation not supported)
[2018-09-12 13:15:50.777241] E [MSGID: 0] [dht-rebalance.c:1696:dht_migrate_file] 0-tier2-dht: Create dst failed on - tier2-disperse-10 for file - /CSP/sp1/CESM/archive/sps_200508_003/atm/hist/postproc/sps_200508_003.cam.h0.2006-01_grid.nc
[2018-09-12 13:15:50.777676] E [MSGID: 109023] [dht-rebalance.c:2733:gf_defrag_migrate_single_file] 0-tier2-dht: migrate-data failed for /CSP/sp1/CESM/archive/sps_200508_003/atm/hist/postproc/sps_200508_003.cam.h0.2006-01_grid.nc

Could you please help me to understand what is happening and how to solve it?

Our Gluster implementation is based on Gluster v.3.10.5

Thank you in advance,
Mauro


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



-------------------------
Mauro Tridici

Fondazione CMCC
CMCC Supercomputing Center
presso Complesso Ecotekne - Università del Salento -
Strada Prov.le Lecce - Monteroni sn
73100 Lecce  IT

mobile: (+39) 327 5630841




-------------------------
Mauro Tridici

Fondazione CMCC
CMCC Supercomputing Center
presso Complesso Ecotekne - Università del Salento -
Strada Prov.le Lecce - Monteroni sn
73100 Lecce  IT

mobile: (+39) 327 5630841




-------------------------
Mauro Tridici

Fondazione CMCC
CMCC Supercomputing Center
presso Complesso Ecotekne - Università del Salento -
Strada Prov.le Lecce - Monteroni sn
73100 Lecce  IT

mobile: (+39) 327 5630841
 

_______________________________________________
Gluster-users mailing list
Gluster-users@xxxxxxxxxxx
https://lists.gluster.org/mailman/listinfo/gluster-users
[Index of Archives]     [Gluster Development]     [Linux Filesytems Development]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux