[Debian-ha-maintainers] Resource cannot run anywhere

[Ferenc Wagner]

"Leonhardt,Christian" <christian.leonhardt at dako.de> writes:

> Am Donnerstag, den 16.07.2015, 15:11 +0200 schrieb Ferenc Wagner:
>
>> "Leonhardt,Christian" <christian.leonhardt at dako.de> writes:
>> 
>>> The resource p_cluster_ip_int1 can run on both nodes. I never had
>>> problems with this one. Only p_apache2, p_bind9, p_lighttpd and
>>> p_squid3 are making trouble.
>> 
>> Maybe your failcounts need a reset?
>> http://www.3open.org/d/ha/how_to_fix_resource_not_started_by_pacemaker
>
> The failcounts are fine. I did a resource cleanup before starting a new
> migration attempt. Alle counters are zero while the global threshold is
> 3.

Then try reading the logs.  Start with those of the DC (crm status will
tell you which node it is).  If you don't find any clue, upload them
somewhere so that others can take a look.  Unfortunately I can't promise
you anything as it's holiday season and both my hands are full of urgent
things.  Consider opening a thread on users at clusterlabs.org.
-- 
Regards,
Feri.