[ClusterLabs] Antw: Re: Antw: Ordering constraint restart second resource group

[Ulrich Windl]

>>> Andrei Borzenkov <arvidjaar at gmail.com> schrieb am 13.08.2015 um 11:33 in
Nachricht
<CAA91j0WaYcPPNCtMZnwDz4_QDFWgxPrO6DbB=gA3bv+_OOOxEA at mail.gmail.com>:
> On Thu, Aug 13, 2015 at 11:25 AM, Ulrich Windl
> <Ulrich.Windl at rz.uni-regensburg.de> wrote:
>> And what exactly is your problem?
> 
> Real life example. Database resource depends on storage resource(s).
> There are multiple filesystems/volumes with database files. Database
> admin needs to increase available space. You add new storage,
> configure it in cluster ... pooh, your database is restarted. There is
> zero need to restart database because it does not even use new
> resource yet.
> 
> I do the above routinely with other cluster implementation without any
> visible impact.

Hi!

So maybe that's why planning usually is done before implementation. We had similar problems with our NFS exports, and we redesigned to speed up the resource start and stop times, as well as allowing changes without restarting the NFS server (and most importantly) local (dependent) client resources...

What you could do: Add your new IP resource with a location constraint only (it will be started on the right node then). Then put the IP resource into the group, and the cluster will see that every resource is in the desired state, and nothing will be restarted.

Regards,
Ulrich

> 
>>                                                    If you change a resource, 
> it will be
>> restrted, and if a resource is restarted, constraints will be followed...
>>
>> Despite of that: If I understand your configuration correctly, it's very 
> much
>> the same as
>>
>> resource_group
>>   ip1
>>   ip2
>>   apache1
>>
>> Regards,
>> Ulrich
>>
>>>>> John Gogu <ionut.gogu at gmail.com> schrieb am 12.08.2015 um 18:35 in
>> Nachricht
>> <CAMESV9DUj3owj16oT5DSYjxZWeZX1f5wV63=mUYtA3VV0KKVGQ at mail.gmail.com>:
>>> Hello,
>>> in my cluster configuration I have following situation:
>>>
>>> resource_group_A
>>>    ip1
>>>    ip2
>>> resource_group_B
>>>    apache1
>>>
>>> ordering constraint resource_group_A then resource_group_B symetrical=true
>>>
>>> When I add a new resource from group_A, resources from group_B are
>>> restarted. If I remove constraint all ok but I need to keep this ordering
>>> constraint.
>>>
>>>
>>> John
>>>
>>>
>>>
>>
>>
>>
>>
>> _______________________________________________
>> Users mailing list: Users at clusterlabs.org 
>> http://clusterlabs.org/mailman/listinfo/users 
>>
>> Project Home: http://www.clusterlabs.org 
>> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf 
>> Bugs: http://bugs.clusterlabs.org 
> 
> _______________________________________________
> Users mailing list: Users at clusterlabs.org 
> http://clusterlabs.org/mailman/listinfo/users 
> 
> Project Home: http://www.clusterlabs.org 
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf 
> Bugs: http://bugs.clusterlabs.org