Details
-
Bug
-
Resolution: Fixed
-
Test Blocker
-
7.6.0
-
Triaged
-
0
-
Unknown
Description
In our weekly runs using 7.6.0-1419, we saw the issue happened intermittently. ep_dcp_replica_items_remaining can't reach 0 for hours.
2023-08-29T08:14:11 [INFO] ep_dcp_replica_items_remaining = 516936
...
2023-08-29T15:44:25 [INFO] ep_dcp_replica_items_remaining = 516936
Build timed out (after 480 minutes). Marking the build as aborted.
There is another ticket opened for the same issue (MB-58342) in 7.2.1. However, we didn't see this issue in our previous weekly pipeline using 7.6.0-1363. It's not clear if it's a new issue in 7.6 or it shares the same root cause as MB-58342. Therefore, I opened this ticket to track the issue.
Build: 7.6.0-1419
Jobs:
http://perf.jenkins.couchbase.com/job/magma-nvme-ycsb/4462/
http://perf.jenkins.couchbase.com/job/magma-nvme-ycsb/4459/
Logs:
https://s3.amazonaws.com/bugdb/jira/qe/collectinfo-2023-08-30T040233-ns_1%40172.23.100.135.zip
https://s3.amazonaws.com/bugdb/jira/qe/collectinfo-2023-08-30T040233-ns_1%40172.23.100.136.zip
https://s3.amazonaws.com/bugdb/jira/qe/collectinfo-2023-08-30T040233-ns_1%40172.23.100.137.zip
Attachments
Issue Links
- is duplicated by
-
MB-58536 DCP replication is hung with some nodes consistently TMP_OOM-ing while below HWM
- Closed
-
MB-58633 [System Test] :- Rebalance out hangs in kv service
- Closed
-
MB-58651 Tests were stuck during load/access phase in 7.6
- Closed
- is triggering
-
MB-58833 Compile with -Wshadow (and equivalent for Clang and MSVC)
- Open