Details
-
Bug
-
Resolution: Fixed
-
Critical
-
3.0
-
Security Level: Public
-
Platform = Physical
OS = CentOS 6.5
CPU = Intel Xeon E5-2630
Memory = 64 GB
Disk = HDD
-
Untriaged
-
Centos 64-bit
-
-
No
-
June 30 - July 18
Description
1 of 4 nodes is being re-added after failover.
100M x 2KB items, 10K mixed ops/sec.
Steps:
1. Failover one of nodes.
2. Add it back.
3. Enabled delta recovery.
4. Sleep 20 minutes.
5. Rebalance cluster.
Rebalance in "full recovery" mode takes 19 min.
Rebalance in "delta recovery" mode takes 25 min.
In the beginning of rebalance ton of Temp OOM errors are thrown.
Working set is screwed (cache miss rate goes up to 50%) during rebalance.
Also there is a spike in ops latency due to rapid vbmap changes.