Matt,
Hey guys, all the help you have provided me thus far has been awesome,
I appreciate it immensely. I was wondering though if its possible to
setup a Distributed Recovery Log for a vdb. I could have sworn that I
read something in the documentation before but I can't seem to find it
any more. If anyone has tried this I'd like to know your success rate
and any problems you may have encountered. I would be running the
recovery logs on mysql backends and each vdb would have their own
recovery log, keeping with standards. I just know that when you create
a distributed vdb which is what the recovery log will be, it also
needs a recovery log.
Yes, it is possible to use a distributed recovery log. For the
RecoveryLog JDBC URL, you point to another virtual database (possibly
hosted by the same controllers). However, the virtual database that you
use as the recovery log must not have a recovery log itself else it ends
up in an infinite recursion.
If that looks cool on paper I don't think there is much practical
interest in having a distributed recovery log. Having a local recovery
log on each controller performs much better and offers as much
reliability. Management procedures become much more complex with a
distributed recovery log, you have to make sure that everything is
started in the proper order.
Hope this helps,
Emmanuel
--
Emmanuel Cecchet
Chief Scientific Officer, Continuent
Blog: http://emanux.blogspot.com/
Open source: http://www.continuent.org
Corporate: http://www.continuent.com
Skype: emmanuel_cecchet
Cell: +33 687 342 685
_______________________________________________
Sequoia mailing list
[email protected]
https://forge.continuent.org/mailman/listinfo/sequoia
