Details
Description
Rebalance (in 3->4) + views (1 ddoc x 3 views) + front-end workload (2K ops/sec per node, 200 stake=ok queries/s).
Master events for both 2.0 and 2.0.1 are attached.
2.0 diags and debug stats:
http://172.23.96.10:8080/job/apollo-views/17/artifact/
2.0.1 diags and debug stats:
http://172.23.96.10:8080/job/apollo-views/18/artifact/
Charts are rather interesting imho, even for 2.0 there are periods of slowness. And it definitively correlates with data (bucket) compaction.
Attachments
For Gerrit Dashboard: MB-7676 | ||||||
---|---|---|---|---|---|---|
# | Subject | Branch | Project | Status | CR | V |
24397,2 | MB-7676: prioritize view compactions heavily when we wait for it | 2.0.1 | ns_server | Status: MERGED | +2 | +1 |
24522,1 | Merge remote-tracking branch 'origin/2.0.1' into b202 | 2.0.2 | ns_server | Status: MERGED | +2 | +1 |
24555,1 | Merge remote-tracking branch 'couchbase/2.0.2' | 2.0.1 | ns_server | Status: ABANDONED | 0 | 0 |
24556,1 | Merge remote-tracking branch 'couchbase/2.0.2' | master | ns_server | Status: MERGED | +2 | +1 |