[Pacemaker] how to realize group with colocation?

[Andrew Beekhof]

On Wed, May 19, 2010 at 1:45 PM, Dennis J. <dennisml at conversis.de> wrote:
> On 05/19/2010 08:59 AM, Andrew Beekhof wrote:
>> Which part of
>>
>> "web_start_0 failed with rc=6: Preventing web from re-starting
>> anywhere in the cluster"
>>
>> Is not clear to you?
>>
>> Have a look what rc=6 means:
>>     http://www.clusterlabs.org/doc/en-US/Pacemaker/1.0/html/Pacemaker_Explained/s-ocf-return-codes.html
>>
>
> If the error code is specifically defined then why not output the actual
> message as well like this:
>
> "web_start_0 failed with rc=6 (The resource's configuration is invalid. Eg.
> A required parameters are missing.): Preventing web from re-starting
> anywhere in the cluster"

Sorry, but I don't plan on outputting War and Peace every time an
operation fails, I feel requiring a 65" screen would have a larger
impact on usability.
The very next line does contain this though:

WARN: unpack_rsc_op: Processing failed op web_start_0 on vm2: not configured (6)

Or perhaps look at the crm_mon output:

Failed actions:
    web_start_0 (node=vm2, call=16, rc=6, status=complete): not configured


>
> That would save admins around the world some time and make it more
> straightforward to determine what the actual problem is. MySQL does the
> same and requires you to "decipher" errors using the perror binary even
> though there is always only one message associated with one error code.
> Making these lookups manual is unnecessary IMHO.
>
> Regards,
>  Dennis
>
> _______________________________________________
> 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
>