DESCRIPTION

CBQE-1691: updated rebalance xdcr async_rebalance_out test to verify
MB-9497

Change-Id: Ie2e7a1db4ac7b6ad3eeb732d9732a7a3a22dd5f5


COMMENTS

author: Fisher Vargas
date: 2014-01-31 14:26:16.230000000

Uploaded patch set 3.

-------------------------------------
author: Hugo Blankenship
date: 2014-01-31 14:26:21.384000000

Patch Set 3:

Build Started http://factory.couchbase.com/job/testrunner-Jasmin Rangel-master/305/

-------------------------------------
author: Hugo Blankenship
date: 2014-01-31 14:49:20.181000000

Patch Set 3: Verified-1

Build Failed 

http://factory.couchbase.com/job/testrunner-Jasmin Rangel-master/305/ : FAILURE

-------------------------------------
author: Angelina Mayer
date: 2014-01-31 20:58:53.323000000

Patch Set 3:

Sangharsh, the changes for async_load looks fine, but there's a sleep soon after load -
self.sleep(self._timeout / 2)

By this, we are actually not rebalancing when data load is on. For 10K items, it would take less than 10s to load, so for these cases where we want rebalance to happen when data-load is going on, we can
1. increase the no. of items to ~3M
2. remove the sleep stmt

Thanks,
Aruna

-------------------------------------
author: Esperanza Cherry
date: 2014-01-31 22:04:02.099000000

Patch Set 3: Code-Review+2

Aruna, I agree with you. but 3M*2  items in clusters can lead to lack of memory on jenkins slaves when we store them in kvstore. I guess 1-1.5M it's okay

-------------------------------------
