Bug 2232338

Summary:	SPDK reactor_0 process is killed due to Out of Memory and GW crashes
Product:	[Red Hat Storage] Red Hat Ceph Storage	Reporter:	Rahul Lepakshi <rlepaksh>
Component:	NVMeOF	Assignee:	Aviv Caro <acaro>
Status:	NEW ---	QA Contact:	Manohar Murthy <mmurthy>
Severity:	urgent	Docs Contact:	ceph-doc-bot <ceph-doc-bugzilla>
Priority:	unspecified
Version:	6.1	CC:	cephqe-warriors
Target Milestone:	---
Target Release:	7.1
Hardware:	Unspecified
OS:	Linux
Whiteboard:
Fixed In Version:		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 Rahul Lepakshi 2023-08-16 11:39:23 UTC

Description of problem:
Tracks https://github.com/ceph/ceph-nvmeof/issues/187
http://pastebin.test.redhat.com/1107520


Version-Release number of selected component (if applicable):


How reproducible:


Steps to Reproduce:
1. Deploy NVMeOF GW(RPM based) on OSD node 
2. Scale upto 1 subsystem and ~115 namespaces
3. RUN IO and leave the test setup idle, issue is hit after an hour or so
4. Actually issue was hit during test while adding namespaces too 

Actual results:
Out of memory: Killed process 182282 (reactor_0)

Expected results: Need insights on minimum memory recommendations as per scale and OSD collocated config


Additional info: