Bug 1222440

Summary: [doc] Issue in "CREATING AN LVM-ACTIVATE RESOURCE"
Product: [openSUSE] PUBLIC SUSE Linux Enterprise High Availability Extension 15 SP5 Reporter: Jiri Belka <jiri.belka>
Component: DocumentationAssignee: Tahlia Richardson <tahlia.richardson>
Status: RESOLVED FIXED QA Contact: Tanja Roth <taroth>
Severity: Normal    
Priority: P5 - None CC: ha-bugs, jiri.belka, ygao, zzhou
Version: unspecified   
Target Milestone: ---   
Hardware: Other   
OS: Other   
Whiteboard: https://jira.suse.com/browse/DOCTEAM-1357
Found By: --- Services Priority:
Business Priority: Blocker: ---
Marketing QA Status: --- IT Deployment: ---

Description Jiri Belka 2024-04-08 08:34:01 UTC
CREATING AN LVM-ACTIVATE RESOURCE:

https://documentation.suse.com/sle-ha/15-SP5/html/SLE-HA-all/cha-ha-clvm.html#pro-ha-clvm-rsc-lvm-activate

There is not "glue" between lvmlockd and the actual VG, thus VG might be (and often is) activated before lvmlockd is ready and thus it would fail...

An example (from a customer case):

group base-group dlm clvmd
group group_GSXP Cluster_IP_gasxp VG_gasxp01vg_new FS_gasxp01lv01_new FS_gasxp01lv02_new FS_gasxp01lv03_new FS_gasxp01lv04_new FS_gasxp01lv05_new FS_gasxp01lv06_new FS_gasxp01lv07_new FS_gasxp01lv09_new FS_gasxp01lv10_new FS_gasxp01lv11_new FS_gasxp01lv12_new FS_gasxp01lv13_new FS_gasxp01lv14_new FS_gasxp01lv15_new rwe_GASXP \
        meta target-role=Started

clone base-clone base-group \
        meta interleave=true target-role=Started

# to be added: glue between clvmd and group using FS on shared VG
colocation col-group_GSXP-with-base-clone inf: group_GSXP base-clone
order ord-base-clone-before-group_GSXP Mandatory: base-clone group_GSXP

Could you improve the docs please? 12-SP5 has same issue for clvmd.
Comment 1 Tahlia Richardson 2024-04-09 00:37:59 UTC
Thanks Jiri, I've added this to my queue.
Comment 2 Tahlia Richardson 2024-05-02 05:19:29 UTC
Here's the PR: https://github.com/SUSE/doc-sleha/pull/373

Jiri, let me know if this solves the problem :) 

Yan, could you please look too? Does this make sense?
Comment 3 Yan Gao 2024-05-03 13:36:37 UTC
(In reply to Tahlia Richardson from comment #2)
> Here's the PR: https://github.com/SUSE/doc-sleha/pull/373
> 
> Jiri, let me know if this solves the problem :) 
> 
> Yan, could you please look too? Does this make sense?

LGTM. only a nitpicking there.

I'm looping in Roger to take a look as well.
Comment 4 Jiri Belka 2024-05-07 08:53:49 UTC
(In reply to Yan Gao from comment #3)
> (In reply to Tahlia Richardson from comment #2)
> > Here's the PR: https://github.com/SUSE/doc-sleha/pull/373
> > 
> > Jiri, let me know if this solves the problem :) 
> > 
> > Yan, could you please look too? Does this make sense?
> 
> LGTM. only a nitpicking there.
> 
> I'm looping in Roger to take a look as well.

I liked Yan's comment. Thx.
Comment 5 Roger Zhou 2024-05-10 07:42:14 UTC
(In reply to Jiri Belka from comment #4)
> I liked Yan's comment. Thx.

Thanks Jiri to bring up this DOC improvement!

(In reply to Yan Gao from comment #3)
> (In reply to Tahlia Richardson from comment #2)
> > Here's the PR: https://github.com/SUSE/doc-sleha/pull/373
[...]
> 
> I'm looping in Roger to take a look as well.

I agree most of your conclusion in the PR. Thx!
Comment 6 Tahlia Richardson 2024-05-16 01:01:07 UTC
Thanks all. This is now complete.