[Openais] Re: [Linux-cluster] Where to go with cman ?

[Patrick Caulfield]

Steven Dake wrote:
> On Mon, 2005-08-08 at 16:30 +0100, Patrick Caulfield wrote:
> 
>>Steven Dake wrote:
>>
>>>Thats great news Patrick.  One thing you should be aware of is that I
>>>have changed some of the internal interfaces in preparation for others
>>>to use libtotem to be extremely more sanitary.  Unfortunately I may have
>>>done this a little too late in your case..  But I think you will find
>>>things are a little better.  It really only effects totempg_initialize.
>>>Also libtotem was renamed to libtotem_pg because of requests from Daniel
>>>about a name-space collision with some movie player in fc4.
>>
>>Yes I spotted that, my current "nearly-working" cman is based on the latest SVN
>>sources.
>>
>>
>>>For multihoming, I want to support the totem redundant ring protocol in
>>>the totem code.  This is an extension of totemsrp to support multiple
>>>nics per processor.  Then data is either actively or passively
>>>replicated over multiple links.  There is essentially no failover and
>>>multiple links can offer better performance and still operate properly
>>>when one entire network fails.  It looks pretty simple to implement.
>>>The paper is at:
>>>
>>>http://www.rcsc.de/pdf/icdcs02.pdf
>>
>>Excellent, thanks. I'll have a read.
>>
> 
> 
> Patrick,
> 
> Over the weekend I reorged the totem code significantly (although the
> totempg interfaces have not changed).  The reorg was painful timewise,
> but the result is that redundant ring should be pretty easy to implement
> now.  Basically I took all of the network junk out of totemsrp and put
> it in "totemnet.c".  It also allows for multiple instances of totemnet
> binds.  This is the main feature I needed to implement redundant ring in
> a clean fashion.  The ipv6 support should be a little easier to add now
> since most of the network code is limited to totemnet.

Superb! I intend to get on to the IPv6 stuff in the next week or so, other
things permitting.

> I should have a patch in a few days with a redundant ring passive and
> active implementation.



-- 

patrick