Cisco Bug: CSCvv74034 - mongodb-2 in Crashloop after upgrade from 1337 to 2123 (maglev 1.5.219)
Oct 03, 2020
- Cisco DNA Center
Known Affected Releases
Symptom: *Note: this issue may cause some loose of data that may be unrecoverable. Always take backups before upgrading.* During upgrade, we have two mongodb pod that are up and one pod in crash loop. The pod in the crash loop has mongodb container that is failing during container creation stage. We have to look at the logs of this crashed mongodb container, to check if this is due to failed Rollback. Check the logs to see if these lines are present: 1) "Detected unclean shutdown - /var/data/mongodb/mongod.lock is not empty." 2) " I REPL [initandlisten] Rollback ID is" 3) "[replication-0] Fatal Assertion ...." Conditions: Upgrade or any ungraceful shut down of primary before it can replicate client data on one of the secondary, for writes that did not ask for majority write concern. This is a known issue with Mongodb. The issue is with Rollback during failover, for details please see: https://docs.mongodb.com/manual/core/replica-set-rollbacks/.
Bug details contain sensitive information and therefore require a Cisco.com account to be viewed.
Bug Details Include
- Full Description (including symptoms, conditions and workarounds)
- Known Fixed Releases
- Related Community Discussions
- Number of Related Support Cases