Bug 1390022

Summary:	[ceph-iscsi-config] rbd-target-gw crash if the running host is not defined to the active configuration
Product:	Red Hat Storage Console	Reporter:	Paul Cuzner <pcuzner>
Component:	ceph-ansible	Assignee:	Paul Cuzner <pcuzner>
Status:	CLOSED ERRATA	QA Contact:	Tejas <tchandra>
Severity:	high	Docs Contact:
Priority:	high
Version:	2	CC:	adeza, aschoen, ceph-eng-bugs, ceph-qe-bugs, gmeno, jdillama, kdreyer, kurs, nthomas, pcuzner, sankarshan, tchandra
Target Milestone:	---
Target Release:	2
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:	ceph-iscsi-config-1.5-1.el7cp	Doc Type:	If docs needed, set a value
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2016-11-22 23:42:25 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:

Description Paul Cuzner 2016-10-30 22:41:02 UTC

Description of problem:
If the rbd-target-gw service is started BEFORE the host is registered to the configuration object, the service fails with a python backtrace.

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


How reproducible:
Every time

Steps to Reproduce:
1. start the rbd-target-gw service before configuratin OR
   create a one node configuration then try and start the service on node 2
2.
3.

Actual results:
backtrace and rbd-target-gw service fails

Expected results:
Service should check whether there is any work to do, and message accordingly

Additional info:

Comment 3 Paul Cuzner 2016-11-02 04:37:23 UTC

Please test against 1.5test release

Comment 4 Tejas 2016-11-02 14:25:10 UTC

Paul, 

  Had a question here. Should'nt  the rbd-target-gw service be masked when the gateway is configured on the node?

the steps I tried:
1. started the rbd-target-gw service on a node which does not have the gateway configured.
2. configred the iscsi gateway on that node.
3. I am still able to manipulate the rbd-target-gw service.

Let me know if I have misunderstood this.

Thanks,
Tejas

Comment 6 Jason Dillaman 2016-11-02 16:03:20 UTC

What is the impact of this issue? Is this the admin starting the service before they add the node the playbook and re-run ansible? If so, that seems like this could be documented and fixed in the next release.

Comment 7 Paul Cuzner 2016-11-03 23:23:53 UTC

Issue is resolved with 1.5 released to QE today (fri 04 nov)

Crash is exactly that - admin starts the service, prior to any playbook run. Rather than let the daemon fail and cause confusion, the code now checks if there is anything to do, no work = no-op.

Comment 8 Paul Cuzner 2016-11-03 23:26:36 UTC

(In reply to Tejas from comment #4)
> Paul, 
> 
>   Had a question here. Should'nt  the rbd-target-gw service be masked when
> the gateway is configured on the node?
> 
> the steps I tried:
> 1. started the rbd-target-gw service on a node which does not have the
> gateway configured.
> 2. configred the iscsi gateway on that node.
> 3. I am still able to manipulate the rbd-target-gw service.
> 
> Let me know if I have misunderstood this.
> 
> Thanks,
> Tejas

rbd-target-gw replaces the target service - so it needs to be present and active on all nodes. It is responsible for rebuilding the iscsi configuration following a reboot, by reading and applying the current configuration object.

if you start the rbd-target-gw on a gateway that is not yet configured, the daemon should not crash.

please retest this scenario on the 1.5 downstream release.

Comment 12 Tejas 2016-11-04 17:13:06 UTC

Verified in build:
ceph-iscsi-config-1.5-1.el7cp
No crash seen.

Comment 14 errata-xmlrpc 2016-11-22 23:42:25 UTC

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2016:2817