{UPR}:: Rebalance-in of 2 nodes is stuck when doing Ops

1014, centos 6x

Vms:: 10.6.2.144-150

1. Create 7 node cluster
2. Create default bucket
3. Add 400 K items
4. Do mutations and rebalance-out 2 nodes
5. Do mutations and rebalance-in 2 nodes

Step 5 leads to rebalance being stuck

Test Case:: ./testrunner -i ../palm.ini get-cbcollect-info=False,get-logs=False,stop-on-failure=False,get-coredumps=True,force_kill_memached=False,verify_unacked_bytes=True,total_vbuckets=128,std_vbuckets=5 -t rebalance.rebalanceinout.RebalanceInOutTests.incremental_rebalance_out_in_with_mutation,init_num_nodes=3,items=400000,skip_cleanup=True,GROUP=IN_OUT;P0