Bug 128635

Summary: fence_node: fences the node it's run on
Product: [Retired] Red Hat Cluster Suite Reporter: Derek Anderson <danderso>
Component: gfsAssignee: Christine Caulfield <ccaulfie>
Status: CLOSED CURRENTRELEASE QA Contact: Cluster QE <mspqa-list>
Severity: medium Docs Contact:
Priority: medium    
Version: 4CC: teigland
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2004-10-04 19:25:43 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:

Description Derek Anderson 2004-07-27 14:43:10 UTC
Description of problem:
Doing some rudimentary fence testing.  Set the fence agent to just
echo a couple of lines to the console so I can tell who/when got
fenced.  Quorate cluster with everyone in the same fence domain.  From
node link-12 ran:

fence_node link-10

The result is that the fence agent was run on link-12.  It then
errantly reports that the fence of link-10 was successful.  Console
output from link-12:
===================
I HAVE BEEN FENCED!
===================
Jul 27 14:37:15 link-12 ccsd[2151]: No nodes found.
Jul 27 14:37:15 link-12 ccsd[2151]: Error while processing get: No
data available
Jul 27 14:37:15 link-12 ccsd[2151]: No nodes found.
Jul 27 14:37:15 link-12 ccsd[2151]: Error while processing get: No
data available
Jul 27 14:37:15 link-12 fence_node[2230]: Fence of "link-10" was
successful

[root@link-12 root]# cat /proc/cluster/nodes
Node  Votes Exp Sts  Name
   1    1    3   M   link-12
   2    1    3   M   link-10
   3    1    3   M   link-11
[root@link-12 root]# cat /proc/cluster/services

Service          Name                              GID LID State     Code
Fence Domain:    "default"                           1   2 run       -
[1 3 2]

DLM Lock Space:  "clvmd"                             2   3 run       -
[1 2 3]

Version-Release number of selected component (if applicable):
[root@link-12 root]# fence_node -V
fence_node DEVEL.1090955557 (built Jul 27 2004 14:13:44)
Copyright (C) Red Hat, Inc.  2004  All rights reserved.

How reproducible:
Always

Steps to Reproduce:
1. Quorate cman cluster
2. Everyone in default fence domain
3. Run "fence_node nodeX" from node nodeY
  
Actual results:
nodeY is fenced.


Expected results:
nodeX fenced.

Additional info:

Comment 1 David Teigland 2004-07-28 03:21:20 UTC
I will try to figure out what's happening here but I think you
misunderstand the relation between fenced and fence_node -- the
two have nothing to do with each other.

fenced deals with the fence domain and executes a fencing agent
against a node when a node is removed from the domain, based on
input from cman.

fence_node is not used by our software but is a simple and potentially
useful utility.  fence_node looks up fencing parameters from ccs and
then calls the specified agent on the victim.  In effect, fenced
calls fence_node when it needs to fence a node that's left the domain.


Comment 2 David Teigland 2004-07-28 04:14:01 UTC
I'm using fence_manual and see no problems when using fence_node
as you've shown.

Comment 3 Derek Anderson 2004-10-04 19:25:43 UTC
Verified against:
fenced DEVEL.1096898839 (built Oct  4 2004 09:08:31)
Copyright (C) Red Hat, Inc.  2004  All rights reserved.

Comment 4 Kiersten (Kerri) Anderson 2004-11-16 19:06:04 UTC
Updating version to the right level in the defects.  Sorry for the storm.