[Pacemaker] Feature request: cleanup resource on primitive definition change

[Rasto Levrinc]

On Wed, Dec 21, 2011 at 7:38 AM, Vladislav Bogdanov
<bubble at hoster-ok.com> wrote:
> 21.12.2011 09:11, Rasto Levrinc wrote:
>> On Wed, Dec 21, 2011 at 5:24 AM, Vladislav Bogdanov
>> <bubble at hoster-ok.com> wrote:
>>> 21.12.2011 06:21, Andrew Beekhof wrote:
>>>> On Tue, Dec 13, 2011 at 11:32 PM, Vladislav Bogdanov
>>>> <bubble at hoster-ok.com> wrote:
>>>>> Hi Andrew, all,
>>>>>
>>>>> I'm now testing latest changes in git, everything goes much cleaner then
>>>>> a week ago, I'll (hopefully) make report later.
>>>>>
>>>>> One feature came into mind during testing:
>>>>> Imagine resource which was tried to be started on all nodes and failed
>>>>> everywhere.
>>>>> Then administrator changes resource definition to fix start problem.
>>>>> Now, resource is not tried to start until it is cleaned up (or both its
>>>>> failure-timeout passes and cluster state re-checked).
>>>>> Idea is to make automatic resource cleanup after definition change.
>>>>
>>>> Not good if the device wasn't failed.
>>>> Imagine someone changing the 'ip' for an IPaddr resource, the old
>>>> value would never be removed.
>>>
>>> I meant only failed resources.
>>>
>>
>> LCMC GUI does this for you.
>
> If you need GUI, have where to run it continuously and is willing to do
> it ;)

It's java it runs everywhere. :) But it doesn't work the way you think it
does and it doesn't have to run continuously.

What I meant is, that if you make a change in resource with LCMC and
apply/commit it and it was failed it cleans it up automatically. It doesn't
run in the background and does something while you make changes
with crm shell.

Rasto

-- 
Dipl.-Ing. Rastislav Levrinc
rasto.levrinc at gmail.com
Linux Cluster Management Console
http://lcmc.sf.net/