[Pacemaker] Question about behavior when a resource was put into unmanaged mode

[David Vossel]

----- Original Message -----
> From: "Kazunori INOUE" <inouekazu at intellilink.co.jp>
> To: "pacemaker at oss" <pacemaker at oss.clusterlabs.org>
> Sent: Wednesday, July 31, 2013 1:20:58 AM
> Subject: [Pacemaker] Question about behavior when a resource was put into	unmanaged mode
> 
> Hi,
> 
> I'm using pacemaker-1.1.10. (4ae1f17)
> When I put a primitive (p1) of group resource into unmanaged mode, low-order
> primitive (p2) restarted.
> In the case of pacemaker-1.0.13, p2 doesn't stop (it continues "Started").
> Is this the assumed behavior?

I played around with this and I'm seeing the same thing you are... I would not expect setting a resource as un-managed to restart dependencies like this.  Please file a bug at bugs.clusterlabs.org so we can make sure this gets fixed.

I also tried setting order and colocation constraints between the two resources in a way that would make them work the same as a group.  Setting the first resource in the chain to unmanaged did not result in a restart of the other resources. This issue looks limited to group usage.

-- Vossel