[Pacemaker] centos 6 fence_apc parameter error

[Dejan Muhamedagic]

Hi Rasto,

On Thu, Oct 11, 2012 at 11:31:10AM +0200, Rasto Levrinc wrote:
> On Thu, Oct 11, 2012 at 10:29 AM, Dejan Muhamedagic <dejanmm at fastmail.fm> wrote:
> > On Wed, Oct 10, 2012 at 09:45:46AM -0500, Michael Brennen wrote:
> >> On Wed, 10 Oct 2012, Dejan Muhamedagic wrote:
> 
> >> >>
> >> >>=========================================
> >> >>crm configure primitive f_apc stonith:fence_apc params \
> >> >>  pcmk_host_map="fail1:1,fail2:2" \
> >> >>  pcmk_host_list="fail1 fail2" \
> >> >>  pcmk_host_check="static-list" \
> >> >>  ipaddr="1.2.3.4" \
> >> >>  action="reboot" \
> >> >>  login="user" \
> >> >>  passwd="pw"
> >> >>=========================================
> >> >>
> >> >>That results in the following message:
> >> >>
> >> >>=========================================
> >> >>ERROR: f_apc: parameter fail2 does not exist
> >> >>Do you still want to commit?
> >> >>=========================================
> >> >>
> >> >>I have not been able to find a way around this, but the
> >> >>configuration seems exactly what I have found in the lists.  Does
> >> >>anyone have any idea what might be happening?  Thanks.
> >> >
> >> >It won't work from the command line, because the crm shell parser
> >> >never sees the quotes (they're digested by /bin/sh). You can type
> >> >this in the crm shell or protect the quotes using \ (at least
> >> >those for pcmk_host_list, others not necessary).
> >> >
> >> >Thanks,
> >> >
> >> >Dejan
> >>
> >> Well... now that I think about it that makes perfect sense.
> >
> > It does, though it certainly is surprising. With two
> > lexers/parsers involved, well, it takes a computer brain to deal
> > with it :( If anybody has an idea how to deal with this in a
> > better way...
> 
> Weird thing is that python splits "pcmk_host_list=fail1 fail2" argument to
> "pcmk_host_list=fail1" and "fail2", unlike other languages. What's up with
> that?

It's just the usual shell-like lexer.

> Once that is fixed, it shouldn't be a problem.

After the chat with you, the fix is upstream now. It's (quite) a
bit of a kludge, but it seems to work. Many thanks for persisting!

Cheers,

Dejan

> Rasto
> 
> -- 
> Dipl.-Ing. Rastislav Levrinc
> rasto.levrinc at gmail.com
> Linux Cluster Management Console
> http://lcmc.sf.net/
> 
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> 
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://bugs.clusterlabs.org