0

We have a Cassandra cluster with three nodes installed on RedHat. In the past, the cluster could be stopped and started without problems. After rebooting the cluster, Cassandra no longer starts. The following error occurs

ERROR [main] 2025-02-25 11:25:58,771 CassandraDaemon.java:708 - Exception encountered during startup

java.lang.IllegalArgumentException: e9311530-b596-11e8-b635-dd45db3e561g is already bound in reverseMap to (demo,airports)


at org.apache.cassandra.utils.ConcurrentBiMap.put(ConcurrentBiMap.java:97) ~[apache-cassandra-3.11.2.jar:3.11.2]

How can this problem be solved?

There is an older discussion in https://issues.apache.org/jira/browse/CASSANDRA-15166 but no solution proposed.

eckstein151
  • 3
  • 2

1 Answers1

1

You didn't provide a lot of information but I suspect that there's something wrong with the schema for one (or more) of the tables. The full error message PLUS the full stack trace would confirm my suspicion.

In any case, what the message tells me is that at least one of the tables has a duplicate schema entry. The error is saying that the table ID is already associated with the demo.airports table but it came across another table in the schema with the same e9311530-b596-11e8-b635-dd45db3e561g ID.

For what it's worth, that table ID is a time UUID which has a timestamp embedded in it indicating that the table was created on 11 Sept 2018 07:47:19 GMT.

Something happened that caused duplicate entries in the schema. For example, if you tried to restore tables in the schema_*/ folders, something might have gone wrong to corrupt the schema. For the record, I'm just thinking out loud and not an accusation.

In any case, a possible workaround is to recreate the schema on a node then manually copying the SSTables into the new folders to recover the data as if you were restoring them from snapshots like I documented in this post. Cheers!

Erick Ramirez
  • 4,590
  • 1
  • 8
  • 30