Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

Bug 632415

Summary:	RFE: A "--restart-all" flag to force a full condor_restart of all subsystems
Product:	Red Hat Enterprise MRG	Reporter:	Erik Erlandson <eerlands>
Component:	wallaby-utils	Assignee:	Robert Rati <rrati>
Status:	CLOSED WONTFIX	QA Contact:	Lubos Trilety <ltrilety>
Severity:	medium	Docs Contact:
Priority:	low
Version:	1.0	CC:	eerlands, iboverma, ltrilety, matt, mhusnain, mkudlej, rrati, willb
Target Milestone:	2.0	Keywords:	FutureFeature
Target Release:	---
Hardware:	All
OS:	Linux
Whiteboard:
Fixed In Version:	condor-wallaby-client-4.0-1	Doc Type:	Enhancement
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2011-05-25 17:35:01 UTC	Type:	---
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:	679143
Bug Blocks:

Description Erik Erlandson 2010-09-09 21:07:17 UTC

Not sure if this falls under qmf-config or wallaby utilities, but I think it's a condor-specific issue.

Either way, it would be nice to provide admin with a way to say "I want to just restart every subsystem on the node when I activate"   This gets around any possible ambiguities or bugs in subsystem memberships for parameters, and also happens to provide an easy way around the "90-startd restart" problem, or the generic "nonstandard subsystem name" problem.   Just restart it all, please.

might look like:   "--activate --restart-all"

Comment 2 Robert Rati 2011-02-23 15:15:24 UTC

*** Bug 679143 has been marked as a duplicate of this bug. ***

Comment 3 Robert Rati 2011-02-23 15:18:28 UTC

The wallaby shell has a new command "force-restart" which will force specified
nodes (or groups) to restart all subsystems.

This command is not intended to be packaged and it only available in the
wallaby repo in lib/mrg/grid/config/shell

Comment 4 Martin Kudlej 2011-03-08 12:11:37 UTC

Is this fixed by command parameters of CCP or just by wallaby console command?
Are these command parameters accessible in any available package?

Comment 5 Robert Rati 2011-03-08 14:24:12 UTC

This is only available through the wallaby shell and is not packaged in the wallaby shell package.  To use, you need to grab the command/rb file from the wallaby repo.

Comment 6 Lubos Trilety 2011-05-24 15:14:53 UTC

If I run first force-restart only for current node and then for all nodes, it doesn't restart master daemon on current node.

Scenario:
1. run command for current node
# wallaby force-restart -n `hostname`
Console Connection Established...

2. see ConfigLog
05/24 16:47:17 INFO: Retrieving configuration version "1306248420855274" from the store
05/24 16:47:19 INFO: Retrieved configuration from the store
05/24 16:47:20 DEBUG: Daemons to restart: [u'master']
05/24 16:47:20 DEBUG: Daemons to reconfig: []
05/24 16:47:20 DEBUG: Sending command "condor_restart" to subsystem "master"

3. run force-restart for all nodes
# wallaby force-restart -a
Console Connection Established...

4. see ConfigLog
05/24 16:52:18 INFO: Retrieving configuration version "1306248721187698" from the store
05/24 16:52:20 INFO: Retrieved configuration from the store
05/24 16:52:21 DEBUG: Daemons to restart: []
05/24 16:52:21 DEBUG: Daemons to reconfig: []
05/24 16:52:21 DEBUG: Sending command "condor_reconfig" to subsystem "master"

The restart is done using special parameter WALLABY_FORCE_RESTART, which is changed with any running of the wallaby force-restart. If option '-n' is wallaby set this value for current node in the store, if '-a' is used wallaby do the same for default-group. Problem is that setting for current node has bigger priority, so change in default-group for the same parameter is ignored.

(One other notice, if wallaby force-restart is run without any argument it ends without error, but nothing happened. Maybe it will be better if it prints some error or run on local node.)

>>> ASSIGNED

Comment 8 Misha H. Ali 2011-05-30 04:28:21 UTC

    Technical note added. If any revisions are required, please edit the "Technical Notes" field
    accordingly. All revisions will be proofread by the Engineering Content Services team.
    
    New Contents:
Condor now includes a shell command for wallaby which forces a restart operation for the specified nodes or group. The user must get the "command/rb" file from the wallaby repository to use the force-reset command. The syntax for this command is "force-restart <parameter> <hostname>", in which the parameter can be either "-n" to for the current node and "-a" for all nodes in the default group.

Comment 9 Misha H. Ali 2011-06-06 03:24:53 UTC

Technical note can be viewed in the release notes for 2.0 at the documentation stage here:

http://documentation-stage.bne.redhat.com/docs/en-US/Red_Hat_Enterprise_MRG/2.0/html-single/MRG_Release_Notes/index.html#tabl-MRG_Release_Notes-GRID_Update_Notes-RHM_Known_Issues

Comment 10 Matthew Farrellee 2011-06-06 14:47:55 UTC

Deleted Technical Notes Contents.

Old Contents:
Condor now includes a shell command for wallaby which forces a restart operation for the specified nodes or group. The user must get the "command/rb" file from the wallaby repository to use the force-reset command. The syntax for this command is "force-restart <parameter> <hostname>", in which the parameter can be either "-n" to for the current node and "-a" for all nodes in the default group.