[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-ha-dev
Subject:    Re: [Linux-ha-dev] application integration
From:       Alan Robertson <alanr () unix ! sh>
Date:       2002-07-30 1:33:54
[Download RAW message or body]

Brien wrote:
>>>If I wanted to use the api would i need to create a custom message 
>>>that i send over to the node that owns of the ip address and then 
>>>write a plugin that handles the message and simply issues a 
>>>T_ASKRESOURCES message?  Do I have to worry about race conditions?
>>>
>>If you just run the hb_standby command (or send the message 
>>it sends), you 
>>don't have to worry about race conditions.
>>
> 
> For my application I cannot rely on being able to run commands remotely.
> Regardless, I don't see how this avoids race conditions.  Why would
> running a remote command be superior to sending a message via heartbeat
> to the other node?

Sorry.  I just meant that hb_standby handles the race conditions - 
regardless of how it's invoked.

> 
> 
>>>If I do in fact have to create a custom message/handler, is 
>>>
>>there any 
>>
>>>value in rolling this into the general API?
>>>
>>This is the kind of thing the API is meant to do.
>>
> 
> Sorry if I wasn't clear, but what I am asking is if "tell another node
> to go into standby" can be considered general purpose functionality and
> therefore should be handled natively by the heartbeat daemon.  It _was_
> very simple to implement, but requiring another process to handle this
> functionality seems like overkill.


This sounds reasonable.  There are ways to do this using the "shell API", 
where you don't have to have a continually running process.

	-- Alan Robertson
	   alanr@unix.sh

_______________________________________________________
Linux-HA-Dev: Linux-HA-Dev@lists.community.tummy.com
http://lists.community.tummy.com/mailman/listinfo/linux-ha-dev
Home Page: http://linux-ha.org/
[prev in list] [next in list] [prev in thread] [next in thread]