Replication failed error,badarg

[csemio]

Hi team,
I'have a problem when i start a new replication between two existing databases, i got this message :
"_replication_state_reason": "{error,badarg}"

We have a 3 nodes clusters on couchdb 3.0.0
We have a database with 30Millions docs
Replication doc :
{ "_id": "090e714507f62a025bab8d166000753d", "_rev": "2-14d1008fc3552d842bcffcb101e7ebfa", "user_ctx": { "name": "admin", "roles": [ "_admin", "_reader", "_writer" ] }, "source": { "url": "http://localhost:5984/activity-logs", "headers": { "Authorization": "Basic XXXXXXXXXXXXXXXX" } }, "target": { "url": "http://localhost:5984/activity-logs-new", "headers": { "Authorization": "Basic XXXXXXXXXXXXXXXX" } }, "create_target": false, "continuous": false, "owner": "admin", "_replication_state": "failed", "_replication_state_time": "2021-10-15T15:46:02Z", "_replication_state_reason": "{error,badarg}" }

many thanks.

[nickva]

See if there are any log lines with then badarg term in them. Then capture a few lines around that that may have stacktraces as well. Primarily look for logs on the CouchDB instance where the replication jobs runs, but there may be a chance it's on one of the endpoint instances as well.

Also 3.0.0 seems a bit old and we already released 3.2.0. Another option is to try to upgrade and see if the error still persists.