ActiveMQ Artemis offers JGroups integration for cloud use-cases such as yours, but, generally speaking, we don't really provide support for JGroups in the ActiveMQ community since we don't maintain the JGroups code-base and don't know its inner workings.
That said, I would suggest that the JGroups configuration using FILE_PING shipped in the examples may not be the best for your use-case. We chose the FILE_PING configuration because it was the simplest for the use-cases in the examples (i.e. where all the nodes are running on the same machine on different ports with access to the same disk). However, something like KUBE_PING [1] would probably be a better fit for your use-case since you are, in fact, using Kubernetes. In any event, it's hard to comment on what may be happening without more information. You've only provided a snippet of the stack-trace for the UnknownHostException so it's not clear what class is actually attempting that operation. Justin [1] https://github.com/jgroups-extras/jgroups-kubernetes On Wed, Mar 15, 2023 at 7:27 AM prateekjai...@gmail.com < prateekjai...@gmail.com> wrote: > Hi All, > > How are you? I am a bit confused about the errors reported by artemis slave > instance (running in clustered mode using jgroups). I might be missing > something here so please bear with me. > > Deployment details: I am running 2 pods of artemis in master/slave > configuration in k8s. It supports NFS v4. I am using jgroups configuration > (same file which is supplied in examples of artemis) and I can see that > correct pod names and IPs are updated in share directory > > -------- output from jgroups file ---------------- > root@jms-01-deployment-6d7dcb8ccd-2t9g2:/opt# cat > > /usr/share/artemis-jgroups/active_broadcast_channel/ab62408d-6a7a-4894-9923-2649d8bf3b28.jms-01-deployment-6d7dcb8ccd-2t9g2-29095.list > > jms-01-deployment-6d7dcb8ccd-2t9g2-29095 > ab62408d-6a7a-4894-9923-2649d8bf3b28 192.168.48.154:7800 T > jms-01-deployment-6d7dcb8ccd-j7pxl-34353 > 5b3d099b-8e5f-4dd0-ab43-604ad1ae87c3 192.168.199.2:7800 F > -------------------------------------------------------- > > POD "jms-01-deployment-6d7dcb8ccd-2t9g2" is the master and another is > slave. Correct POD ips are listed there. But in log/artemis.log file of > slave pod, I see these logs > > ------------------------------- > 2023-03-15 11:29:51,519 ERROR [org.apache.activemq.artemis.core.client] > AMQ214016: Failed to create netty connection > java.net.UnknownHostException: jms-01-deployment-6d7dcb8ccd-2t9g2 > at java.net.InetAddress$CachedAddresses.get(InetAddress.java:797) ~[?:?] > at java.net.InetAddress.getAllByName0(InetAddress.java:1524) ~[?:?] > at java.net.InetAddress.getAllByName(InetAddress.java:1382) ~[?:?] > ----------------------------------- > > jgroups is correctly creating and updating files. IPs of master/slave pods > are visible from each other. I have tested them using command > > "./artemis check node". > Right now I have created deployment objects. Is it necessary to create > stateful sets? > > Am I missing something here? Let me know if you need more information on > this. > > > Regards, > Prateek Jain > > > -------------------------------------------------------------- > EXPECTATION : Causes all troubles...... > -------------------------------------------------------------- >
