Ruby driver fails to connect with JRuby 9.3

Sudhir_Rao · 2023-06-30T18:57:42.163Z

Connecting to Amazon DocumentDB (mongodb 5.0) from EKS pods. Some pods connect and work, others fail to connect. The failing pods stay at “server description unknown.” Log below which keeps repeating. Downgrading JRuby to 9.2 works.

Ruby driver version: 2.19.0 (also tried 2.18.2)
Jruby version: 9.3.9.0
Mongoid version: 7.5.3 (also tried 8)

Debug log:

MONGODB | Server description for cdris-cluster-dev.cluster-clrxvva0yuqz.us-west-2.docdb.amazonaws.com:27017 changed from 'unknown' to 'unknown'.
MONGODB | There was a change in the members of the 'Unknown' topology.
MONGODB | Error checking cdris-cluster-dev.cluster-clrxvva0yuqz.us-west-2.docdb.amazonaws.com:27017: Java::JavaUtilConcurrent::RejectedExecutionException: Task java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask@3c76cafa[Not completed, task = java.util.concurrent.Executors$RunnableAdapter@4913ed11[Wrapped task = org.jruby.ext.timeout.Timeout$TimeoutTask@62f85778]] rejected from java.util.concurrent.ScheduledThreadPoolExecutor@2b87cd3a[Terminated, pool size = 0, active threads = 0, queued tasks = 0, completed tasks = 0]
MONGODB | Server description for cdris-cluster-dev.cluster-clrxvva0yuqz.us-west-2.docdb.amazonaws.com:27017 changed from 'unknown' to 'unknown'.
MONGODB | There was a change in the members of the 'Unknown' topology.
MONGODB | Error checking cdris-cluster-dev.cluster-clrxvva0yuqz.us-west-2.docdb.amazonaws.com:27017: Java::JavaUtilConcurrent::RejectedExecutionException: Task java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask@6621f2b0[Not completed, task = java.util.concurrent.Executors$RunnableAdapter@731d9b2e[Wrapped task = org.jruby.ext.timeout.Timeout$TimeoutTask@79b6aead]] rejected from java.util.concurrent.ScheduledThreadPoolExecutor@2b87cd3a[Terminated, pool size = 0, active threads = 0, queued tasks = 0, completed tasks = 0]
MONGODB | Server description for cdris-cluster-dev.cluster-clrxvva0yuqz.us-west-2.docdb.amazonaws.com:27017 changed from 'unknown' to 'unknown'.

alexbevi · 2023-07-03T09:45:01.853Z

Hey @Sudhir_Rao,

Thanks for letting us know about this issue. Though our Ruby Driver’s Compatibility matrix does list JRuby 9.3 as working with both 2.19 and 2.18, we can only guarantee this behavior when connecting to a MongoDB cluster.

The Ruby Driver (as well as Mongoid) are tested extensively against JRuby, however DocumentDB is an external product that is managed by AWS. They have some documentation for troubleshooting connection issues, but aside from that if you have difficulties with DocumentDB you’d need to work with their support team to determine why these issues appear to persist with JRuby 9.3 when JRuby 9.2 works as expected.

Imdad_Ali_Khan · 2024-06-27T11:34:40.792Z

I have a similar issue but using Ruby 3.3.3 and Docker

MONGODB | Topology type ‘unknown’ initializing.
web-1 | 11:07:04 web.1 | MONGODB | There was a change in the members of the ‘Unknown’ topology.

Any idea on how to fix it?

P.S: This worked fine and after restarting the dev machine this started poping in and reinstalling docker fixes this but data is lost

alexbevi · 2024-06-27T14:06:51.952Z

This thread was for a JRuby error. It might make more sense for you to create a new thread for your issue, including any errors and driver/mongoid version.