rabbitmq-operator icon indicating copy to clipboard operation
rabbitmq-operator copied to clipboard

ERROR: epmd error for host imp20rabbit-0.imp20rabbit-discovery.atom-test.svc.cluster.imp: nxdomain (non-existing domain)

Open cuijxin opened this issue 6 years ago • 1 comments

Hi, When I deploy the cr on my kubernetes cluster, I got the error "epmd error...", and I have deploy the operator done, and the status is running.

➜  deploy git:(master) ✗ kubectl get po -n atom-test
NAME                                 READY   STATUS    RESTARTS   AGE
rabbitmq-operator-76c59cf54b-krkbj   1/1     Running   0          65m

Then, I deploy the rabbitmq cr use:

➜  deploy git:(master) ✗ kubectl apply -f crds/rabbitmq_v1_rabbitmq_cr.yaml -n atom-test
rabbitmq.rabbitmq.improvado.io/imp20rabbit created

but, the result is terrible:

➜  deploy git:(master) ✗ kubectl get po -n atom-test
NAME                                 READY   STATUS             RESTARTS   AGE
imp20rabbit-0                        1/2     CrashLoopBackOff   3          2m17s
imp20rabbit-1                        0/2     Pending            0          15s
rabbitmq-operator-76c59cf54b-krkbj   1/1     Running            0          69m
➜  deploy git:(master) ✗ kubectl get pvc -n atom-test
NAME                        STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS   AGE
rabbit-data-imp20rabbit-0   Bound    pvc-8d546809-d43b-11e9-a67a-00163e088669   1Gi        RWO            standard       2m25s
rabbit-data-imp20rabbit-1   Bound    pvc-d609523a-d43b-11e9-a67a-00163e088669   1Gi        RWO            standard       23s
➜  deploy git:(master) ✗ kubectl get po -n atom-test
NAME                                 READY   STATUS             RESTARTS   AGE
imp20rabbit-0                        1/2     CrashLoopBackOff   3          2m37s
imp20rabbit-1                        0/2     Init:0/1           0          35s
rabbitmq-operator-76c59cf54b-krkbj   1/1     Running            0          70m

What does the error mean, and how can I resolve it.

cuijxin avatar Sep 11 '19 02:09 cuijxin

@cuijxin oh, sorry, I haven't seen notification about this issue. Would be great, if you will be able to provide logs from rabbitmq replicas. However, resource you've applied was created at early stages of operator development with debug goals - I doubt it is valid now. Will test tommorow and fix it / remove from repo + add viable example.

inish777 avatar Oct 07 '19 16:10 inish777