Details
-
Bug
-
Resolution: Fixed
-
Critical
-
5.5.0
-
Build: 5.5.0-1751
Cluster: iris
OS: CentOS 7
CPU: Data: E5-2630 v2 (24 vCPU), Query & Index: E5-2680 v3 (48 vCPU) Memory: Data & Query: 64GB, Index: 512GB Disk: Samsung Pro 850
-
Untriaged
-
Centos 64-bit
-
-
Yes
Description
Changes caused the regression:
changed projects :
|
|
goproj/src/github.com/couchbase/indexing changed from 917a8ca928985dbfef6ba89e7ebf0ae781c71dca to master
|
[+] 560eec9 Merge remote-tracking branch 'couchbase/unstable' into HEAD
|
[+] 87e8384 MB-25162: Update Go targets to use Go 1.7.6
|
[+] c7aecc7 MB-26597 : GSI - Add Log Redaction Support - Part 2
|
[+] 0586f6e MB-23000 REST endpoint for Index stats ...
|
[+] f2f961e MB-26677 Upgrade path for XATTRs ...
|
|
query-ui changed from ca84b39b6e64fdf45beca714fb8a33d69ad82385 to master
|
[+] 0647dfb Improvements to handling of large documents in new doc editor.
|
[+] d3ff86a MB-27681 - Add an expand button to the JSON editor in the *new* document editor.
|
http://172.23.120.24/builds/latestbuilds/couchbase-server/vulcan/1751/CHANGELOG
CPU profile indicates major code path related to log redaction functionality (TagUD method).
The regression most probably caused by this line:
l.Debugf("ScanPipeline::computeGroupAggr %v", l.TagUD(s.p.aggrRes))
(https://github.com/couchbase/indexing/commit/c7aecc78a40c4cf80a5f004cf98720ccbdc0cd6f#diff-e430d5eddad2ae791a8f3a4d11c6539bR193)
when entire aggregation result is being passed to TagUD()
CPU profile attached.
Affected tests/queries:
http://showfast.sc.couchbase.com/#/timeline/Linux/n1ql/count/MOI
http://showfast.sc.couchbase.com/#/timeline/Linux/n1ql/tokens/all
http://showfast.sc.couchbase.com/#/timeline/Linux/n1ql/Q5_Q7_thr/all
Attachments
Issue Links
- relates to
-
MB-27774 GSI: Observed latency and throughput regression
- Closed