Details
-
Bug
-
Resolution: Fixed
-
Critical
-
7.1.0
-
Untriaged
-
-
1
-
Yes
Description
Build : 7.1.0-1822
Test : -test tests/integration/neo/test_neo_magma_milestone3.yml -scope tests/integration/neo/scope_neo_magma.yml
Scale : 3
Iteration : 1st
Seeing cbq-engine crash on multiple nodes multiple times with the following stack trace :
This is from the node 172.23.96.48, but the crashes are there on other query nodes as well - 172.23.96.254, 172.23.97.112.
2021-12-02T01:00:40.190-08:00 [Info] GSIC[default/bucket5-scope_0-coll_3-1638435400165951464] logstats "bucket5" {"gsi_scan_count":7,"gsi_scan_duration":26670421956,"gsi_throttle_duration":1394237936,"gsi_prime_duration":170473496,"gsi_blocked_duration":21840893003,"gsi_total_temp_files":16}
|
panic: runtime error: index out of range [-1]
|
|
goroutine 620415 [running]:
|
github.com/couchbase/query/inferencer.(*UnifiedDocumentRetriever).GetNextDoc(0xc016f42400, 0x28702c0, 0xc000160780, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0)
|
/home/couchbase/jenkins/workspace/couchbase-server-unix/goproj/src/github.com/couchbase/query/inferencer/document_retrievers.go:657 +0x1b2a
|
github.com/couchbase/query-ee/updstat.doUpdateStatistics(0x28782d0, 0xc0001efe10, 0x2847f80, 0xc016f42400, 0xc01b6374e0, 0x5, 0x8, 0x0, 0x6b0a3, 0xc002fc3a70, ...)
|
/home/couchbase/jenkins/workspace/couchbase-server-unix/goproj/src/github.com/couchbase/query-ee/updstat/upd_statistics.go:370 +0x1ad
|
github.com/couchbase/query-ee/updstat.(*DefaultStatUpdater).UpdateStatistics(0xc00037e8d0, 0x28782d0, 0xc0001efe10, 0xc000fa1e80, 0x3, 0x4, 0xc01b637440, 0xf, 0x12, 0x287def0, ...)
|
/home/couchbase/jenkins/workspace/couchbase-server-unix/goproj/src/github.com/couchbase/query-ee/updstat/upd_statistics.go:290 +0x9ac
|
created by github.com/couchbase/query/execution.(*UpdateStatistics).RunOnce.func1
|
/home/couchbase/jenkins/workspace/couchbase-server-unix/goproj/src/github.com/couchbase/query/execution/update_stat.go:87 +0x365
|
2021-12-02T01:00:42.002-08:00 [INFO] Current nofiles rlimit: 200000 (max: 200000)
|
Last known good build : 7.1.0-1809
Possible change that might have caused this regression:
Commit: eedbcdd25937df6fa0eb730bf9b30d6837b591dd in build: couchbase-server-7.1.0-1822
MB-49860 Need to reset document retriever if multiple passes needed for UPDATE STATISTICS
Change-Id: I56c09a5f96f7ac923020590f1394417486d01c21
Reviewed-on: https://review.couchbase.org/c/query-ee/+/166792
Reviewed-by: Donald Haggart <donald.haggart@couchbase.com>
Tested-by: Bingjie Miao <bingjie.miao@couchbase.com>
Author: miaobingjie <bingjie.miao@couchbase.com>
Committer: Bingjie Miao <bingjie.miao@couchbase.com>
Attachments
For Gerrit Dashboard: MB-49875 | ||||||
---|---|---|---|---|---|---|
# | Subject | Branch | Project | Status | CR | V |
166894,3 | MB-49875 Close index connection on retriever reset and directly indicate when cache is active and should be used. | master | query | Status: MERGED | +2 | +1 |
166937,2 | MB-49875 Revise fix. | master | query | Status: MERGED | +2 | +1 |