Xavier,
I configured the load balancer with the completion policy
WaitForCompletionPolicy.FIRST
So, the raidb1 load balancer executes each write request on each
backends and wait for the success of one of them.
Yes. In fact it waits for the success of both of them because if one
fails but the other succeeds, the failed backend must be disabled.
However, the result is returned to the client as soon as one node is
successful.
But what about the read requests ?
It one of the read request is sent to one invalid backend, is the read
request redirected to the second backend ?
A read request is only sent to an active backend (not disabled). If that
backend happened to fail during the query execution, it is retried
automatically on another backend.
Hope this clarifies things,
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
