Details
Assignee
UnassignedUnassignedReporter
zelmar.michelinizelmar.micheliniNeeds QA
YesAffects versions
Priority
High
Details
Details
Assignee
Unassigned
UnassignedReporter
zelmar.michelini
zelmar.micheliniNeeds QA
Yes
Affects versions
Priority
Smart Checklist
Smart Checklist
Smart Checklist
Created 5 days ago
Updated 3 days ago
If you restore a backup into a new k8s cluster you will get
Time monotonicity violation
errors on cfg servers and mongos and the pods will start to restart:Time monotonicity violation error:
{"t":{"$date":"2025-04-02T18:12:27.649+00:00"},"s":"E", "c":"ASSERT", "id":4457000, "ctx":"CatalogCache-3","msg":"Tripwire assertion","attr":{"error":{"code":6493100, "codeName":"Location6493100","errmsg":"Time monotonicity violation: lookup time { chunkVersion: { e: ObjectId('67ed72e8739da6f4f6c648d6'), t: Timestamp(1743614696, 13), v: Timestamp(1, 0) }, forcedRefreshSequenceNum: 9, epochDisambiguatingSequenceNum: 340 } which is less than the earliest expected timeInStore { chunkVersion: { e: Object Id('67ed778e6982c61201c9628f'), t: Timestamp(1743615886, 10), v: Timestamp(1, 0) }, forcedRefreshSequenceNum: 9, epochDisambiguatingSequenceNum: 8 }."},"location":"{file Name:\"src/mongo/util/read_through_cache.h\", line:549, functionName:\"operator()\"}"}}
zelmar@LAPTOP-MD0FVN06:~/CS0053759/CS0053759_oldcluster-dump/cluster-dump/dbaas-mongodb-rs-mongodb$ kubectl get pods NAME READY STATUS RESTARTS AGE percona-server-mongodb-operator-7f7764cd57-rzmjq 1/1 Running 0 84m my-cluster-name-rs0-1 2/2 Running 0 81m my-cluster-name-rs0-0 2/2 Running 1 (79m ago) 84m my-cluster-name-rs0-2 2/2 Running 0 79m my-cluster-name-cfg-0 2/2 Running 7 (9m36s ago) 84m my-cluster-name-cfg-1 2/2 Running 7 (7m23s ago) 81m my-cluster-name-mongos-0 1/1 Running 9 (2m44s ago) 59m my-cluster-name-cfg-2 2/2 Running 5 (2m17s ago) 79m my-cluster-name-mongos-1 1/1 Running 9 (2m19s ago) 58m my-cluster-name-mongos-2 1/1 Running 9 (2m9s ago) 58m
To reproduce:
Deploy a cluster (Operator 1.19.1 and percona-server-mongodb:7.0.16-10)
Insert some data
Take a backup
Deploy a new cluster
Restore the backup.