Details
-
Bug
-
Resolution: Fixed
-
Critical
-
3.0
-
Security Level: Public
-
Centos 6.4
6 Nodes, 2 Buckets, 1 ddoc X 2 Views
3.0.0-1163-rel
-
Untriaged
-
Unknown
-
June 30 - July 18
Description
1. Load items ~ 700M, 630M on both the buckets, 10% dgm
2. During initial indexing / loading phase, I see a number of log messages which expect rollbacks on partitions due to startSeqNo > endSeqNo.
Not sure why should the above happen - given that there is no rebalance/failover/add new node etc activity.
Logs from the cluster https://s3.amazonaws.com/bugdb/11973/bug_logs.tar
Attachments
For Gerrit Dashboard: MB-11973 | ||||||
---|---|---|---|---|---|---|
# | Subject | Branch | Project | Status | CR | V |
40762,1 | MB-11973: Only grab items from the chk manager one chk at a time | master | ep-engine | Status: ABANDONED | 0 | 0 |
40765,4 | MB-11973: Only grab items from the chk manager one chk at a time | 3.0 | ep-engine | Status: MERGED | +2 | +1 |