I would like to pick up this project. Keith can you point me at any (if 
any) code spikes that have been done in this direction? Is there any proof 
of concept code etc? I know binlog replay is a pattern that is pretty 
fleshed out in other persistence daemons. I know that is not necessarily 
what you had been thinking per your statement back in 09:

"The design I was hoping for isn't really for scalability either, just 
high availability. 

The reason I prefer peer-peer 2-way replication is because "failover" 
is ridiculously simple for clients and requires no external tools or 
fancy routing. 

I'm still very much in favor even if you really want to do it 
master-slave. Either way this will be a great improvement."

But I am not sure you even had the on disk persistence then?

In the meantime I like others am planning on running a hot spare that can 
pick up the mount with the binlog if there is a failure in the primary and 
then fencing the primary out.

-- 
You received this message because you are subscribed to the Google Groups 
"beanstalk-talk" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/beanstalk-talk/-/elaU5S2JfEcJ.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/beanstalk-talk?hl=en.

Reply via email to