Details
-
Bug
-
Resolution: Fixed
-
Critical
-
7.1.6
-
7.1.6-3962
-
Untriaged
-
0
-
Yes
Description
fresh install
2 nodes w/kv,n1ql,index
-the following steps happen in the test setup
the test also clusters these two nodes together
then the test creates an empty bucket
then the test deletes an empty bucket
then the test removes one node from the cluster
then clusters them together again
then the test creates an empty bucket and runs the query
run query
cbq> SELECT * FROM system:indexes WHERE indexes.bucket_id is missing;
{
"requestID": "2295f81f-6bff-41f5-b488-cc390d8ea92f",
"errors": [
],
"status": "fatal",
"metrics":
}
not when i do it through the UI it is fine, but through cbq and curl it throws hte service shut down error
attaching logs
this test passes in 7.2.4 and trinity
also the test that runs into this issue is here
http://qa.sc.couchbase.com/job/test_suite_executor/635411/consoleFull
./testrunner -i /data/workspace/debian-p0-query-vset00-00-collections_system_7.0_P0/testexec.36507.ini -p bucket_size=256,skip_primary_index=true,skip_load=True,nodes_init=2,GROUP=P0 -t tuqquery.tuq_system_keyspaces.QuerySystemKeyspacesTests.test_keyspaces_basic,force_clean=True,GROUP=P0,bucket_size=256,skip_primary_index=true,skip_load=True,nodes_init=2
it reliably hits the problem but I have yet to come up with a manual reproduction
Attachments
Issue Links
- is triggered by
-
MB-58732 Topology change: PrepareTopologyChange exceeding call time limit
- Closed