Bug 1887400

Summary: quorum: Add support for nodelist callback (was dlm: gfs2 corruption when nodes rejoin too quickly)
Product: Red Hat Enterprise Linux 8 Reporter: Jan Friesse <jfriesse>
Component: corosyncAssignee: Jan Friesse <jfriesse>
Status: CLOSED ERRATA QA Contact: cluster-qe <cluster-qe>
Severity: high Docs Contact:
Priority: unspecified    
Version: 8.3CC: aahringo, ccaulfie, cluster-maint, cluster-qe, jfriesse, phagara, rpeterso, sbradley, swhiteho, teigland
Target Milestone: rc   
Target Release: 8.0   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: corosync-3.1.0-1.el8 Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: 1866905 Environment:
Last Closed: 2021-05-18 15:26:09 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1866905    

Comment 1 Jan Friesse 2020-10-12 11:33:54 UTC
Upstream patch https://github.com/corosync/corosync/commit/4eb36297288f65fe26d2adedf6c33aced7544976

For QA: Not sure how to test this feature alone. It's just a new API.

Comment 5 Patrik Hagara 2020-11-09 12:36:18 UTC
New API tested using the provided upstream unit test program:

> [root@virt-248 ~]# rpm -q corosync
> corosync-3.1.0-1.el8.x86_64
> [root@virt-248 ~]# cd bz1887400-testapp
> [root@virt-248 bz1887400-testapp]# ./testquorummodel
> quorate   1
> Waiting for quorum events, press ^C to finish
> -------------------
> nodelist notification called
>   ring id            = 1.2c
>   num members        = 5  1 2 3 4 5
>   num joined members = 0
>   num left members   = 0
> quorum notification called
>   quorate   = 1
>   ring id   = 1.2c
>   num nodes = 5  1 2 3 4 5


Stop cluster on another node:

> nodelist notification called
>   ring id            = 1.30
>   num members        = 4  1 3 4 5
>   num joined members = 0
>   num left members   = 1  2
> quorum notification called
>   quorate   = 1
>   ring id   = 1.30
>   num nodes = 4  1 3 4 5


Start the cluster on the node again:

> nodelist notification called
>   ring id            = 1.35
>   num members        = 5  1 2 3 4 5
>   num joined members = 1  2
>   num left members   = 0
> quorum notification called
>   quorate   = 1
>   ring id   = 1.35
>   num nodes = 5  1 2 3 4 5

The list of current/joined/left nodes is passed via the new nodelist notification callback API.

Moving to verified.

Comment 7 errata-xmlrpc 2021-05-18 15:26:09 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (corosync bug fix and enhancement update), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2021:1780