[Bugs] [Bug 1387502] New: Incorrect volume type in the "glusterd_state" file generated using CLI "gluster get-state"

bugzilla at redhat.com bugzilla at redhat.com
Fri Oct 21 06:21:11 UTC 2016 

https://bugzilla.redhat.com/show_bug.cgi?id=1387502

            Bug ID: 1387502
           Summary: Incorrect volume type in the  "glusterd_state" file
                    generated using CLI "gluster get-state"
           Product: GlusterFS
           Version: 3.9
         Component: glusterd
          Severity: high
          Assignee: bugs at gluster.org
          Reporter: sbairagy at redhat.com
                CC: amukherj at redhat.com, bsrirama at redhat.com,
                    bugs at gluster.org, rhinduja at redhat.com,
                    rhs-bugs at redhat.com, storage-qa-internal at redhat.com,
                    vbellur at redhat.com
        Depends On: 1382277, 1385839
            Blocks: 1353427



+++ This bug was initially created as a clone of Bug #1385839 +++

+++ This bug was initially created as a clone of Bug #1382277 +++

Description of problem:
=======================
Showing Incorrect volume type in the glusterd state dump.

[Volumes]
Volume1.name: Dis-Rep         
Volume1.id: fd13329e-35c9-476f-a5d0-ca2be1c488c0
Volume1.type: Replicate      <==============
Volume1.transport_type: tcp
Volume1.status: Started
Volume1.brickcount: 4

Volume info details:
~]# gluster volume info

Volume Name: Dis-Rep
Type: Distributed-Replicate   <============= 
Volume ID: fd13329e-35c9-476f-a5d0-ca2be1c488c0
Status: Started
Snapshot Count: 0
Number of Bricks: 2 x 2 = 4
Transport-type: tcp

......




Version-Release number of selected component (if applicable):
=============================================================
glusterfs-3.8.4-2


How reproducible:
=================
Always


Steps to Reproduce:
===================
1. Have 2 or more cluster nodes
2. Create a 2 * 2 volume type and start it
3. take glusterd state dump using the cmd "gluster get-state"
4. Check for the volume type in the dump file.

Actual results:
===============
Incorrect volume type in the glusterd state dump


Expected results:
=================
Volume type should show correctly


Additional info:

--- Additional comment from Atin Mukherjee on 2016-10-06 04:50:16 EDT ---

Byreddy - please do not use the term statedump here as that indicates something
different. Please change the subject/summary to mention cluster state

--- Additional comment from Atin Mukherjee on 2016-10-06 04:51:50 EDT ---

Correction, it should be local state.

--- Additional comment from Byreddy on 2016-10-14 04:54:00 EDT ---

Some other issues related to this one:

1) replica count entry is not showing, it should be there

2) Correct the peer related info which is discussed // similar to CLI peer
status output

3) Remove the below duplicate entries
Volume1.rebalance.data: 0
Volume1.rebalance.data: 0

4) Entry "Volume1.Brick1.signedin:" is showing false for running brick process 

5)Remove the not needed entries eg:
Volume1.Brick1.filesystem_type:
Volume1.Brick1.mount_options:

I didn't seen these entries filled in my testing.

--- Additional comment from Byreddy on 2016-10-14 07:08:52 EDT ---

+
6) Sever quorum ratio entry is missing, which is needed.

--- Additional comment from Byreddy on 2016-10-14 07:09:32 EDT ---

(In reply to Byreddy from comment #7)
> +
> 6) Server quorum ratio entry is missing, which is needed.

--- Additional comment from Samikshan Bairagya on 2016-10-14 07:46:51 EDT ---

(In reply to Byreddy from comment #7)
> +
> 6) Sever quorum ratio entry is missing, which is needed.

cluster.server-quorum-ratio will be there in the local state data under the
global options section, in case it is explicitly set to some other value other
than the default.

--- Additional comment from Byreddy on 2016-10-17 00:58:09 EDT ---

(In reply to Samikshan Bairagya from comment #9)
> (In reply to Byreddy from comment #7)
> > +
> > 6) Sever quorum ratio entry is missing, which is needed.
> 
> cluster.server-quorum-ratio will be there in the local state data under the
> global options section, in case it is explicitly set to some other value
> other than the default.

Yes Samikshan, i am seeing when i set it explicitly under the global option
section.

--- Additional comment from Byreddy on 2016-10-17 01:03:50 EDT ---

For Tiered volume type,

we have to separate the cold and hot bricks based on the below info in the
file.

Volume4.tier_info.cold_tier_type: Replicate
Volume4.tier_info.cold_brick_count: 3
Volume4.tier_info.cold_replica_count: 3
Volume4.tier_info.cold_disperse_count: 0
Volume4.tier_info.cold_dist_leaf_count: 3
Volume4.tier_info.cold_redundancy_count: 0
Volume4.tier_info.hot_tier_type: Distribute
Volume4.tier_info.hot_brick_count: 2
Volume4.tier_info.hot_replica_count: 1
Volume4.tier_info.promoted: 0
Volume4.tier_info.demoted: 0

--- Additional comment from Byreddy on 2016-10-17 01:52:10 EDT ---

rebalanced data entry is showing with out data unit.

Volume1.rebalance.id: e7227829-1357-4153-8b8f-78e05aec5fe1
Volume1.rebalance.status: started
Volume1.rebalance.failures: 0
Volume1.rebalance.skipped: 0
Volume1.rebalance.lookedup: 988
Volume1.rebalance.files: 684
Volume1.rebalance.data: 93433255    <======= NO DATA UNIT

I think data unit is needed here

--- Additional comment from Worker Ant on 2016-10-17 15:15:13 EDT ---

REVIEW: http://review.gluster.org/15662 (cli, glusterd: Address issues in
get-state cli output) posted (#1) for review on master by Samikshan Bairagya
(samikshan at gmail.com)

--- Additional comment from Worker Ant on 2016-10-19 04:15:35 EDT ---

REVIEW: http://review.gluster.org/15662 (cli, glusterd: Address issues in
get-state cli output) posted (#2) for review on master by Samikshan Bairagya
(samikshan at gmail.com)

--- Additional comment from Worker Ant on 2016-10-20 06:12:33 EDT ---

COMMIT: http://review.gluster.org/15662 committed in master by Atin Mukherjee
(amukherj at redhat.com) 
------
commit daea58a51b70f80ab04f115e49f8bf8790b6046a
Author: Samikshan Bairagya <samikshan at gmail.com>
Date:   Thu Oct 13 17:13:54 2016 +0530

    cli, glusterd: Address issues in get-state cli output

    This fixes the following data points:
    1. Volume type
    2. Peer state
    3. List of other hostnames for a peer
    4. Data unit information for rebalance

    The following data points are removed:
    1. Mount options and filesystem types for bricks
    2. global-option-version from list of global options

    The following data points are added:
    1. Replica Count
    2. Tier type for bricks belonging to hot/cold tier

    Change-Id: I5011250e863fdc4929b203cdb345d79b2f16c6a5
    BUG: 1385839
    Signed-off-by: Samikshan Bairagya <samikshan at gmail.com>
    Reviewed-on: http://review.gluster.org/15662
    Reviewed-by: mohammed rafi  kc <rkavunga at redhat.com>
    CentOS-regression: Gluster Build System <jenkins at build.gluster.org>
    NetBSD-regression: NetBSD Build System <jenkins at build.gluster.org>
    Smoke: Gluster Build System <jenkins at build.gluster.org>
    Reviewed-by: Atin Mukherjee <amukherj at redhat.com>


Referenced Bugs:

https://bugzilla.redhat.com/show_bug.cgi?id=1353427
[Bug 1353427] [RFE] CLI to get local state representation for a cluster
https://bugzilla.redhat.com/show_bug.cgi?id=1382277
[Bug 1382277] Incorrect volume type in the  "glusterd_state" file generated
using CLI "gluster get-state"
https://bugzilla.redhat.com/show_bug.cgi?id=1385839
[Bug 1385839] Incorrect volume type in the  "glusterd_state" file generated
using CLI "gluster get-state"
-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.

More information about the Bugs mailing list