Bug 1091986

Summary:	We need a command to bump-up the glusterd op-version.
Product:	[Red Hat Storage] Red Hat Gluster Storage	Reporter:	Ravishankar N <ravishankar>
Component:	core	Assignee:	Kaushal <kaushal>
Status:	CLOSED ERRATA	QA Contact:	SATHEESARAN <sasundar>
Severity:	unspecified	Docs Contact:
Priority:	high
Version:	rhgs-3.0	CC:	bmohanra, divya, nsathyan, psriniva, rhs-bugs, sasundar, storage-qa-internal
Target Milestone:	---	Keywords:	TestBlocker
Target Release:	RHGS 3.0.0
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:	glusterfs-3.6.0-1.0.el6rhs.x86_64	Doc Type:	Enhancement
Doc Text:	Feature: A new cluster option, 'cluster.op-version' has been introduced which can be used to bump the cluster operating version. The cluster operating version can be bumped using the command # gluster volume set all cluster.op-version <OP-VERSION> The op-version will be bumped only if - all the peers in the cluster support it, and - the new op-version is greater than the current cluster op-version This set operation will not do any other change other than changing and saving the cluster op-version in the glusterd.info file. It will NOT, - change any existing volume - add the option to the global options list - fix the cluster op-version to the given version, it can be bumped up by other volume set commands. This feature is only useful for gluster storage pools which have been upgraded from RHS-2.1 to RHS-3.0. In such a cluster the only valid value to the key is 3, the op-version of RHS-3.0. Reason: Several features in RHS-3.0, including snapshots and remove and add brick commands, only work when all the peers are RHS-3.0. But, an existing cluster is upgraded to RHS-3.0, the cluster continues operating in the previous operating version. This would prevent the above mentioned commands from running. Result (if any): Setting the option 'cluster.op-version' on 'all' volume will bump up the cluster operating version and allow newer features to be used.	Story Points:	---
Clone Of:
Clones:	1092592 1102055 (view as bug list)		Environment:
Last Closed:	2014-09-22 19:36:12 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:
Bug Depends On:	1092592
Bug Blocks:	1084443, 1102055

Description Ravishankar N 2014-04-28 12:48:13 UTC

Description of problem:
Both  the snapshot feature and persistent AFR  xattrs implementation will work
only if op-version >=4.  To that effect we would need a gluster CLI
command to bump up the glusterd op-version when an existing cluster is
upgraded to the release containing these features.

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

How reproducible:

Steps to Reproduce:
1. Form a cluster and create a volume with a glusterfs release having cluster op-version less than 4.
2. Upgrade all nodes to the release having snapshot/ persistent AFR xattrs features.
3. Try snapshot/ add-brick/ remove-brick operations

Actual results:
Add/remove brick operations fail with the message:
One or more nodes do not support the required op-version. Cluster op-version must atleast be 4.

Similarly snapshots would also fail with an appropriate message (not checked the exact message yet).

Expected results:
These features should work.

Comment 2 Kaushal 2014-04-29 14:17:03 UTC

http://review.gluster.org/7603 (glusterd: Allow bumping up the cluster op-version) posted (#1) for review

Comment 3 Pavithra 2014-05-08 07:01:29 UTC

I discussed with Ravi and Kaushal and understood the context of the bug and where it needs to be documented. 

Here is the mail conversation:

Hi Ravi,

I spoke to Kaushal and got all the information about the cluster.op-version command.
Can you please tell me in which step of the upgrade this command needs to be executed?
Here is the document we have:
http://documentation-devel.engineering.redhat.com/site/documentation/en-US/Red_Hat_Storage/2.1/html/Installation_Guide/chap-Installation_Guide-Software_Updates-ISSU.html

Here is the associated bug number:
https://bugzilla.redhat.com/show_bug.cgi?id=1091986

Thanks,
Pavithra
IRC name: psriniva

Hi Pavithra,
The command needs to be executed after upgrading all servers and starting all the volumes. It can be added in the following sections

7.1. Updating Red Hat Storage in the Offline Mode: 
After step 5 (i.e. starting the volumes after upgrade of the servers) and before step 6 (i.e.before upgrading the clients).

7.2.3. In Service Software Upgrade
After step 12 (i.e all server nodes have been upgraded).

However, I think it would be better to put  this in a separate section, say "7.2.6. Post-upgrade Tasks" where we say why we need this command and then give the syntax of the command to run.

Regards,
Ravi

Bhavana will be be taking up this task, as she has been the primary author of Rolling upgrade.

Comment 4 Nagaprasad Sathyanarayana 2014-05-19 10:56:35 UTC

Setting flags required to add BZs to RHS 3.0 Errata

Comment 6 SATHEESARAN 2014-05-28 16:51:40 UTC

Tested with glusterfs-3.6.0.8-1.el6rhs

A new option "cluster.op-version" available to set the op-version

Comment 8 errata-xmlrpc 2014-09-22 19:36:12 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.

http://rhn.redhat.com/errata/RHEA-2014-1278.html