[Debian-ha-maintainers] Resource cannot run anywhere

[Adrian Vondendriesch]

Hi Leonhardt,

Am 14.07.2015 um 09:26 schrieb Leonhardt,Christian:
>>> At the moment the cluster is running pretty well so far but i stuck 
>>> at
>>> the failover tests.
>>> 'crm node standby' works flawlessly but moving a resource to an 
>>> other
>>> node with 'resource move p_lighttpd' for example failes with 
>>> 'Resource
>>> cannot run anywhere'. The resource just get stopped. The problem 
>>> occurs
>>> when moving the resources p_squid3, p_apache2, p_bind9 or 
>>> p_lighttpd.
>>> All other resouces like ip addresses or symlinks are working.
>>> Do i miss something or is it a bug? 
>>
>> I don't look too deep into your configuration, but it seems to me 
>> that
>>
>>>  <rsc_location id="cli-ban-p_lighttpd-on-s1r3drslfw1" 
>>> rsc="p_lighttpd"
>> role="Started" node="s1r3drslfw1" score="-INFINITY"/>
>>
>> is preventing the migration of p_lighttpd to node s1r3drslfw1. Did 
>> you
>> hit "crm_resource -U" ... after "crm_resource -M"? Just a thought.
>>
> Thats what i did. There were no active location constraints before
> moving the resource. cli-ban-p_lighttpd-on-s1r3drslfw1 was created
> during "crm_resource -M" because s1r3drslfw1 was the active node. The
> resource should move to s3r2drslfw1. Be aware of the first four chars
> in the server names. Its a bit tricky.

So,

> location cli-ban-p_cluster_ip_int1-on-s3r2drslfw1 p_cluster_ip_int1
> role=Started -inf: s3r2drslfw1

was not there while you did try to move p_lighttpd?

If the constraint wasn't there, could you reproduce the error? Could you
please provide the output of "crm status" and "crm configure show"
before and after the failed migration? Also I would like to have a look
at the corresponding logfiles of both servers within that timeframe and
the entered commands.

Greetings,
 - Adrian