Note: This is a beta release of Red Hat Bugzilla 5.0. The data contained within is a snapshot of the live data so any changes you make will not be reflected in the production Bugzilla. Also email is disabled so feel free to test any aspect of the site that you want. File any problems you find or give feedback here.
Bug 1598969 - ocf::heartbeat:iSCSILogicalUnit -> This NodeACL already exists in configFS
Summary: ocf::heartbeat:iSCSILogicalUnit -> This NodeACL already exists in configFS
Status: ON_QA
Alias: None
Product: Red Hat Enterprise Linux 7
Classification: Red Hat
Component: resource-agents
Version: 7.5
Hardware: x86_64
OS: Linux
Target Milestone: rc
: ---
Assignee: Oyvind Albrigtsen
QA Contact:
Depends On:
Blocks: 1692413
TreeView+ depends on / blocked
Reported: 2018-07-07 12:08 UTC by Strahil Nikolov
Modified: 2019-03-26 14:46 UTC (History)
4 users (show)

Fixed In Version: resource-agents-4.1.1-20.el7
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
: 1692413 (view as bug list)
Last Closed:
Target Upstream Version:

Attachments (Terms of Use)
debug start of iSCSILogicalUnit resource and cluster backup (deleted)
2018-07-07 12:08 UTC, Strahil Nikolov
no flags Details

Description Strahil Nikolov 2018-07-07 12:08:29 UTC
Created attachment 1457154 [details]
debug start of iSCSILogicalUnit resource and cluster backup

Description of problem:
iSCSILogicalUnit fails to start the resource from the first time with the error 'This NodeACL already exists in configFS' once the resource has failed on all nodes and the resource got clear && cleanup - the resource starts successfully.

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

How reproducible:

Steps to Reproduce:
1. Disable firewall or allow port 7000/tcp
2. Set selinux to permissive
3. Configure 3 nodes with drbd and test them prior to setting it up in the cluster configuration:
3.1. Create drbd0 config file:
# cat /etc/drbd.d/drbd0.res 
resource drbd0 {
protocol C;
net {
        cram-hmac-alg sha1;
        shared-secret "SomePass";

volume 0 {
	device /dev/drbd0;
	disk /dev/drbd/drbd0;
	meta-disk internal;
on drbd1 {
	node-id 1;
on drbd2 {
	node-id 2;
on drbd3 {
	node-id 3;
connection-mesh {
  hosts drbd1 drbd2 drbd3;

3.2 Prepare local storage for drbd on all nodes: 
drbdadm create-md drbd0
drbdadm up drbd0
3.3 Make one node primary for initial sync:
drbdadm primary --force drbd0

4.Configure the cluster by using the "pcs config restore cluster.bkp.tar.bz2" -> provided in the attached zip file

Actual results:
iSCSILogicalUnit fails to start properly.Manual intervention is needed via:
'pcs resource clear iscsi-lun0 && pcs resource cleanup iscsi-lun0'.
Once the nodes get rebooted -> they need to fail starting the resource and cleaned up again.

Expected results:
iSCSILogictalUnit to work as expected

Additional info:
I'm attaching the cluster config backup and output of the debug-start generated via:
pcs resource disable iscsi-lun0
#export OCF_ROOT=/usr/lib/ocf
#export OCF_RESKEY_allowed_initiators=''
#export OCF_RESKEY_lun=0
#export OCF_RESKEY_path=/dev/drbd0
#export OCF_RESKEY_target_iqn=''
#bash -x /usr/lib/ocf/resource.d/heartbeat/iSCSILogicalUnit start

Comment 2 Strahil Nikolov 2018-09-04 19:37:03 UTC
Have you managed to reproduce the issue ?

Comment 3 Oyvind Albrigtsen 2019-02-20 09:40:41 UTC
It seems like this patch should solve this issue:

Comment 5 Patrik Hagara 2019-03-26 09:30:24 UTC
qa_ack+, reproducer in comment#4

Note You need to log in before you can comment on or make changes to this bug.