[system test] couchbase server of 2.0 node does not start after added to 1.8.1 cluster

Cluster information:

• 8 centos 6.2 64bit server with 4 cores CPU
• Each server has 32 GB RAM and 400 GB SSD disk.
• 24.8 GB RAM for couchbase server at each node
• SSD disk format ext4 on /data
• Each server has its own SSD drive, no disk sharing with other server.
• Create cluster with 6 nodes installed couchbase server 1.8.1-937
• Cluster has 2 buckets, default (12GB) and saslbucket (12GB).
• Load 16 million items to each bucket with key size from 512 to 1024 bytes

10.6.2.37
10.6.2.38
10.6.2.44
10.6.2.45
10.6.2.42
10.6.2.43

• Mutate 16 million items with size from 1024 to 1500 bytes
• Install couchbase server 2.0.0-1885 on node 44 and 45

• Link to manifest file of build 2.0.0-1885 http://builds.hq.northscale.net/latestbuilds/couchbase-server-enterprise_x86_64_2.0.0-1885-rel.deb.manifest.xml

• Do online upgrade on cluster from 1.8.1 to 2.0.0 by using swap rebalance.
• Add node 44 and remove node 39. Rebalance passed.
• Add node 45 and remove node 40. Rebalance passed

• Uninstall couchbase server 1.8.1 on node 39. Then delete dir /opt/couchbase and all files/dir in /data/
• Install couchbase server 2.0.0-1885 on node 39. Go through initial setup and set data/view path to /data/
• Add node 39 back to cluster to do swap rebalance. Add node failed. Filed bub MB-7000

• Open web console on node 39. Failed.
• ssh to node 39 and try to start couchbase server, couchbase server failed to start. Got erlang crash dump files.

Link to erlang crash dump files https://s3.amazonaws.com/packages.couchbase/erlang/orange/201210/node39-erl-crash-dump-cb-failed-start-20121024-152214.tgz

Link to collect info of node 39 https://s3.amazonaws.com/packages.couchbase/collect_info/orange/2_0_0/201210/10.6.2.39-10242012-diag.zip