"Schnabl, Sebastian" <[EMAIL PROTECTED]> writes:

>
> I know that I can give java (meaning the controller-process) more memory
> per configuration, but I'm asking myself if this is a permanent lurking
> problem: 
>
> If one controller will fail and the second controller will handle a very
> lot of requests in the time until the failing controller will be
> restored - will this make restore log impossible, until I give the
> server appropriate memory...?

Could you try to set-up a non-hsqldb recoverylog? Try for instance
MySQL. hsqldb may have some limitations with big databases.



_______________________________________________
Sequoia mailing list
[email protected]
https://forge.continuent.org/mailman/listinfo/sequoia

Reply via email to