Details
-
Bug
-
Resolution: Fixed
-
Critical
-
6.0.0
-
Oceanus cluster
-
Untriaged
-
Unknown
-
CX Sprint 115
Description
Setting this ticket as Critical because DEBUG level is default logging level since build 1370.
The rate of degradation:
6.0.0-1388 and 6.0.0-1432 were tested with DEBUG logging level
And for 6.0.0-1460 I've manually switched it back to WARN
The degradation is all across the product. Query latency degraded as well
Comparison between 6.0.0-1432 (DEBUG) and 6.0.0-1460 (WARN):
Avg. ingestion rate (items/sec), 1 node, BigFUN 20M users (320M docs), 3 indexes, SSD:
http://cbmonitor.sc.couchbase.com/reports/html/?snapshot=oceanus_600-1460_sync_aa3f&label=6.0.0-1460-WARN&snapshot=oceanus_600-1432_sync_2116&label=6.0.0-1432-DEBUG
Unfortunately I didn't have IO stats working when running the DEBUG tests. Its now fixed.
I'll rerun the DEBUG test with all stats in place and update the ticket.
Logs:
172.23.96.5:kv
172.23.96.7:kv
172.23.96.8:cbas
[https://s3-us-west-2.amazonaws.com/perf-artifacts/jenkins-oceanus-766/172.23.96.5.zip
] [https://s3-us-west-2.amazonaws.com/perf-artifacts/jenkins-oceanus-766/172.23.96.7.zip
] https://s3-us-west-2.amazonaws.com/perf-artifacts/jenkins-oceanus-766/172.23.96.8.zip
Attachments
Issue Links
- relates to
-
MB-30831 Analytics: Parametrise queries are failing with service is temporarily unavailable
- Closed