[rabbitmq-discuss] Mnesia db integrity

Matthias Radestock matthias at lshift.net
Fri Jan 30 21:00:04 GMT 2009


Valentino,

Valentino Volonghi wrote:
> I've tried going a bit deeper into the thing and I suspect it's
> related to how you would deal with failures in a cloud. Basically you
> start a new image that has a new hostname and I suppose mnesia would
> be waiting for the broken node to come back up.

Correct. That's exactly what would happen. Retaining the hostname across 
restarts is crucial for mnesia to be able to recover. There are ways 
around that, but they are somewhat involved.


Matthias.




More information about the rabbitmq-discuss mailing list