Fencing agents run sweet, tested this on the command line. We're running
2.2.3 Fencing agents are the latest as of about a month ago.
The problem occurs when a host fails because it's ipmi interface is
down, a hardware failure or the management module on the blade server
doesn't respond or someone has made a typo with the password or
something. Instead of the cobbler wrapper just going OK, bad return code
let's try the next host on the list it bails out.
EG. When this happens on a list of multiple hosts it stops at that point
instead of continuing with the next host.
Sat Mar 16 11:09:30 2013 - INFO | cobbler power configuration is:
Sat Mar 16 11:09:30 2013 - INFO | type : ipmilan
Sat Mar 16 11:09:30 2013 - INFO | address: 11.11.11.11
Sat Mar 16 11:09:30 2013 - INFO | user : admin
Sat Mar 16 11:09:30 2013 - INFO | id :
Sat Mar 16 11:09:30 2013 - INFO | running: /usr/sbin/fence_ipmilan
Sat Mar 16 11:09:49 2013 - INFO | received on stdout: Powering on
machine @ IPMI:11.11.11.11...Failed
Sat Mar 16 11:09:49 2013 - DEBUG | received on stderr:
Sat Mar 16 11:09:51 2013 - INFO | running: /usr/sbin/fence_ipmilan
Sat Mar 16 11:10:11 2013 - INFO | received on stdout: Powering on
machine @ IPMI:11.11.11.11...Failed
Sat Mar 16 11:10:11 2013 - DEBUG | received on stderr:
Sat Mar 16 11:10:13 2013 - INFO | running: /usr/sbin/fence_ipmilan
Sat Mar 16 11:10:32 2013 - INFO | received on stdout: Powering on
machine @ IPMI:11.11.11.11...Failed
Sat Mar 16 11:10:32 2013 - DEBUG | received on stderr:
Sat Mar 16 11:10:34 2013 - INFO | running: /usr/sbin/fence_ipmilan
Sat Mar 16 11:10:53 2013 - INFO | received on stdout: Powering on
machine @ IPMI:11.11.11.11...Failed
Sat Mar 16 11:10:53 2013 - DEBUG | received on stderr:
Sat Mar 16 11:10:55 2013 - INFO | running: /usr/sbin/fence_ipmilan
Sat Mar 16 11:11:14 2013 - INFO | received on stdout: Powering on
machine @ IPMI:11.11.11.11...Failed
Sat Mar 16 11:11:14 2013 - DEBUG | received on stderr:
Sat Mar 16 11:11:16 2013 - INFO | Exception occured: <class
'cobbler.cexceptions.CX'>
Sat Mar 16 11:11:16 2013 - INFO | Exception value: 'command failed
(rc=1), please validate the physical setup and cobbler config'
Sat Mar 16 11:11:16 2013 - INFO | Exception Info:
File "/usr/lib/python2.6/site-packages/cobbler/utils.py", line 126,
in die
raise CX(msg)
Sat Mar 16 11:11:16 2013 - INFO | Exception occured: <class
'cobbler.cexceptions.CX'>
Sat Mar 16 11:11:16 2013 - INFO | Exception value: 'command failed
(rc=1), please validate the physical setup and cobbler config'
Sat Mar 16 11:11:16 2013 - INFO | Exception Info:
File "/usr/lib/python2.6/site-packages/cobbler/remote.py", line 89,
in run
rc = self._run(self)
File "/usr/lib/python2.6/site-packages/cobbler/remote.py", line 259,
in runner
self.remote.power_system(object_id,self.options.get("power",""),token,logger=self.logger)
File "/usr/lib/python2.6/site-packages/cobbler/remote.py", line
1915, in power_system
rc=self.api.power_on(obj, user=None, password=None, logger=logger)
File "/usr/lib/python2.6/site-packages/cobbler/api.py", line 1019,
in power_on
return
action_power.PowerTool(self._config,system,self,user,password,logger=logger).power("on")
File "/usr/lib/python2.6/site-packages/cobbler/action_power.py",
line 123, in power
utils.die(self.logger,"command failed (rc=%s), please validate the
physical setup and cobbler config" % rc)
File "/usr/lib/python2.6/site-packages/cobbler/utils.py", line 134,
in die
raise CX(msg)
Sat Mar 16 11:11:16 2013 - ERROR | ### TASK FAILED ###
On 03/18/2013 09:47 AM, James Clendenan wrote:
Hi David
Which power plugins are you using as fence scripts? I know a few of
them can occasionally cause problems as well. Also which version are
you running at the moment for cobbler?
Do you have any logs you can share?
Thanks,
James
--
Sent from my phone.
On 2013-03-18, at 9:11 AM, David Leach <[email protected]
<mailto:[email protected]>> wrote:
Yeah, I was thinking in my head, wouldn't a simple bash loop not have
this problem?
It was me putting in the update ;) I would really appreciate it being
worked on as it's not always tech savvy people using this (excellent)
feature.
Cheers
On 03/18/2013 09:06 AM, James Cammarata wrote:
On Mon, Mar 18, 2013 at 10:59 AM, David Leach
<[email protected] <mailto:[email protected]>> wrote:
Hi,
We're trying to use cobbler as a tool to power down hosts and
power them up in mass whenever we need to do an image update or
just for a downtime.
One issue which is a little frustrating is if you run a power
command on a bunch of hosts and one fails, the command fails at
that point. I know there was a feature request made 6 months ago
to fix this, but I was wondering if there was a work around for
this.
Yes, I saw a recent update on that issue (not sure if it was you). I
can work on that if it is problematic, otherwise one work-around
would be to use a shell script from the command line in a for loop.
That way, and individual failure wouldn't stop the others from working.
_______________________________________________
cobbler mailing list
[email protected]
https://lists.fedorahosted.org/mailman/listinfo/cobbler
_______________________________________________
cobbler mailing list
[email protected] <mailto:[email protected]>
https://lists.fedorahosted.org/mailman/listinfo/cobbler
_______________________________________________
cobbler mailing list
[email protected]
https://lists.fedorahosted.org/mailman/listinfo/cobbler
_______________________________________________
cobbler mailing list
[email protected]
https://lists.fedorahosted.org/mailman/listinfo/cobbler
