Bug 1343677

Summary:	Unassigned host cannot be used in new cluster
Product:	[Red Hat Storage] Red Hat Storage Console	Reporter:	Martin Kudlej <mkudlej>
Component:	Ceph	Assignee:	Shubhendu Tripathi <shtripat>
Status:	CLOSED WONTFIX	QA Contact:	sds-qe-bugs
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	2	CC:	mbukatov, sankarshan
Target Milestone:	---
Target Release:	3
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2018-11-19 05:41:33 UTC	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 Martin Kudlej 2016-06-07 16:15:54 UTC

Description of problem:
I've tried to create cluster but task failed. No host has been marked as assigned to any cluster. So I've tried to create cluster with different name again and it failed with:
ERROR    cluster.go:86 CreateCluster]#033[0m admin:3a53bed1-afa2-489c-9d87-09e524038efc-Nodes [_host1_] already participating in a cluster. New Cluster cannot be created using the node
But _host1_ is not assigned according UI.

Version-Release number of selected component (if applicable):
server:
ceph-ansible-1.0.5-19.el7scon.noarch
ceph-installer-1.0.11-1.el7scon.noarch
rhscon-ceph-0.0.20-1.el7scon.x86_64
rhscon-core-0.0.21-1.el7scon.x86_64
rhscon-ui-0.0.34-1.el7scon.noarch

monitor with calamari:
calamari-server-1.4.0-0.12.rc15.el7cp.x86_64
ceph-base-10.2.1-13.el7cp.x86_64
ceph-common-10.2.1-13.el7cp.x86_64
ceph-mon-10.2.1-13.el7cp.x86_64
ceph-selinux-10.2.1-13.el7cp.x86_64
libcephfs1-10.2.1-13.el7cp.x86_64
python-cephfs-10.2.1-13.el7cp.x86_64
rhscon-agent-0.0.9-1.el7scon.noarch


Steps to Reproduce:
1. try to create cluster but task failed
2. try to create another cluster with same nodes because they ae marked as unassigned

Actual results:
It is possible to start task for creating cluster with nodes which seems unassigned. The task failed because nodes are already assigned.

Expected results:
It will be possible to create cluster from nodes which are marked as unassigned.

Comment 3 Shubhendu Tripathi 2016-06-22 10:36:03 UTC

This actually depends on at what stage the cluster creation has failed. If cluster creation failed while storage bits installation or just before OSD configuration on node, we would need to revert the clusterid field populated in node and then it could be used in another cluster.

If the cluster creation failed at a stage when OSD creation was in progress, some of the disks might be already partitioned and this node ideally we should not use in another cluster before cleaning the disks properly.

Comment 4 Shubhendu Tripathi 2016-06-23 13:41:59 UTC

Moving this to 3.0 and created a doc BZ#1349458 for getting a troubleshooting section added in documentation.

Comment 5 Shubhendu Tripathi 2018-11-19 05:41:33 UTC

This product is EOL now