Bug 2071078

Summary:	[RFE] Support the ocf:heartbeat:iscsi resource agent (RHEL 8)
Product:	Red Hat Enterprise Linux 8	Reporter:	Reid Wahl <nwahl>
Component:	resource-agents	Assignee:	Oyvind Albrigtsen <oalbrigt>
Status:	POST ---	QA Contact:	cluster-qe <cluster-qe>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	8.7	CC:	agk, cfeist, cluster-maint, fdinitto, heinzm, oalbrigt, slevine
Target Milestone:	rc	Keywords:	FutureFeature, Triaged
Target Release:	8.7
Hardware:	All
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Enhancement
Doc Text:		Story Points:	---
Clone Of:	2071077	Environment:
Last Closed:		Type:	Feature Request
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:	2071077
Bug Blocks:

Description Reid Wahl 2022-04-01 18:33:00 UTC

+++ This bug was initially created as a clone of Bug #2071077 +++

Description of problem:

This RFE is for a resource agent that can connect to, disconnect from, and monitor iSCSI targets from the initiator side.

Currently, we have two agents that act at the target side, and we have crude approaches (e.g., resource-agents-deps.target.d) that ensure the iSCSI initiator service is started before pacemaker. It would be beneficial to have more granular control over connections to individual targets, as well as failure detection and recovery.

One proposed use case is single node OpenShift (SNO).

There is an existing upstream resource agent that appears to accomplish this task:
  - https://github.com/ClusterLabs/resource-agents/blob/main/heartbeat/iscsi

Some of the comments there are confusing; they make it sound as if the RA manages things at the target side. However, everything seems to happen at the initiator side, via iscsiadm. It allows specification of portal, target, etc.

The last substantive update to the RA was in 2016, so I hope that it works on RHEL 8/9.