Bug 2310380

Summary: When scaling down the number of GWs, namespaces of the removed GWs ANA grp, remain orphan
Product: [Red Hat Storage] Red Hat Ceph Storage Reporter: Aviv Caro <acaro>
Component: NVMeOFAssignee: Leonid <lchernin>
Status: VERIFIED --- QA Contact: Rahul Lepakshi <rlepaksh>
Severity: urgent Docs Contact: ceph-doc-bot <ceph-doc-bugzilla>
Priority: unspecified    
Version: 8.0CC: aindenba, bdavidov, bkunal, cephqe-warriors, gbregman, kjosy, lchernin, linuxkidd, mmurthy, pdhange, rlepaksh, sunnagar, tserlin, vumrao
Target Milestone: ---Keywords: DeliveryBlocker, External
Target Release: 8.0   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: ceph-19.1.1-16.el9cp Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 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:

Description Aviv Caro 2024-09-06 09:34:08 UTC 
The fix is as follows: 

1. The removed GW will temporarily remain in monitor DB, but in "Deleting" State.

2. Monitor triggers immediate Failover of Deleting GW(s). 

3. Another GW will get the group of the  GW in state “Deleting” (i.e. Failover).

4. In case that another GW is added while Deleting still exits on a GW, the new GW will be assigned to the same grp id (Failback).