Bug 2274757

Summary: [rook][GSS][ODF 4.14.6] Notify/alert end user if legacy OSDs are LVM backed
Product: [Red Hat Storage] Red Hat OpenShift Data Foundation Reporter: Prashant Dhange <pdhange>
Component: rookAssignee: Travis Nielsen <tnielsen>
Status: CLOSED ERRATA QA Contact: Vishakha Kathole <vkathole>
Severity: urgent Docs Contact:
Priority: urgent    
Version: 4.14CC: ableisch, asriram, bkunal, bniver, ebenahar, gabrioux, gsternag, kbg, kelwhite, mmanjuna, muagarwa, nojha, odf-bz-bot, pdhange, rafrojas, roemerso, sostapov, tdesala, tnielsen
Target Milestone: ---   
Target Release: ODF 4.16.0   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
.Legacy LVM-based OSDs are in crashloop state Previously, starting from OpenShift Data Foundation 4.14, the legacy OSDs were crashing in the init container that resized the OSD. This was because, the legacy OSDs that were created in OpenShift Container Storage 4.3 and since upgraded to a future version might have failed. With this fix, the crashing resize init container was removed from the OSD pod spec. As a result, the legacy OSD starts, however it is recommended that the legacy OSDs re replaced soon.
Story Points: ---
Clone Of: 2273398
: 2279928 (view as bug list) Environment:
Last Closed: 2024-07-17 13:19:22 UTC Type: ---
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: 2273398, 2276532, 2276533    
Bug Blocks: 2260844, 2274657, 2279928    

Comment 4 Santosh Pillai 2024-04-15 08:03:20 UTC
Should It prevent entire ODF upgrade or just the upgrade of the lvm based OSDs?

Comment 5 Prashant Dhange 2024-04-15 22:13:43 UTC
(In reply to Santosh Pillai from comment #4)
> Should It prevent entire ODF upgrade or just the upgrade of the lvm based
> OSDs?

Prevention of upgrade for lvm based OSDs is preferred but if we alert the end user before the start of the upgrade then we could avoid the unexpected data-loss situation in advance.

Comment 15 Travis Nielsen 2024-05-09 22:45:33 UTC
Rook will add status under status.storage.legacyOSDs to the CephCluster CR such as the following:

  status:
    storage:
      deviceClasses:
      - name: hdd
      legacyOSDs:
      - id: 0
        reason: LVM-based OSD on a PVC (id=0) is deprecated and should be replaced
      - id: 1
        reason: LVM-based OSD on a PVC (id=1) is deprecated and should be replaced
      - id: 2
        reason: LVM-based OSD on a PVC (id=2) is deprecated and should be replaced
      osd:
        storeType:
          bluestore: 3

I will clone this BZ to get the needed alert raised based on this status.

Comment 20 Vishakha Kathole 2024-06-04 11:44:20 UTC
Moving it to the verified state based on the 4.16 CI regression runs.

Comment 22 errata-xmlrpc 2024-07-17 13:19:22 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 (Important: Red Hat OpenShift Data Foundation 4.16.0 security, enhancement & bug fix 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/RHSA-2024:4591