can't get my first conf call - prosody stuck
cloud32 opened this issue · 2 comments
Hi,
installation passed OK
[d50c-cml-app16 JITSI]$ helm install -f jitsi.yaml myjitsi jitsi/jitsi-meet
NAME: myjitsi
LAST DEPLOYED: Tue Dec 6 08:18:40 2022
NAMESPACE: default
STATUS: deployed
REVISION: 1
NOTES:
1. Get the application URL by running these commands:
export POD_NAME=$(kubectl get pods --namespace default -l "app.kubernetes.io/name=jitsi-meet,app.kubernetes.io/component=web,app.kubernetes.io/instance=myjitsi" -o jsonpath="{.items[0].metadata.name}")
echo "Visit http://127.0.0.1:8080 to use your application"
kubectl --namespace default port-forward $POD_NAME 8080:80
[d50c-cml-app16 JITSI]$ kubectl get pod
NAME READY STATUS RESTARTS AGE
myjitsi-jitsi-meet-jicofo-b86bdb469-mkpxj 0/1 Running 0 4s
myjitsi-jitsi-meet-jvb-766c5c8984-qjjsd 0/1 Running 0 4s
myjitsi-jitsi-meet-web-7f98b75c4d-gff6j 0/1 Running 0 4s
myjitsi-prosody-0 0/1 Pending 0 4s
[d50c-cml-app16 JITSI]$ kubectl get service
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
myjitsi-jitsi-meet-jvb ClusterIP 10.104.75.209 <none> 10000/UDP 33s
myjitsi-jitsi-meet-web ClusterIP 10.101.137.224 <none> 80/TCP 33s
myjitsi-prosody ClusterIP 10.110.85.66 <none> 5280/TCP,5281/TCP,5347/TCP,5222/TCP,5269/TCP 33s
manually added NodePort services so that WEB & COLIBRI are reachable from outside the cluster
[d50c-cml-app16 JITSI]$ kubectl get service
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
jitsi-ext-jvb NodePort 10.111.232.38 <none> 8080:8444/TCP 6s
jitsi-ext-web NodePort 10.98.10.128 <none> 80:8443/TCP 118s
myjitsi-jitsi-meet-jvb ClusterIP 10.103.227.8 <none> 10000/UDP 5m23s
myjitsi-jitsi-meet-web ClusterIP 10.109.214.17 <none> 80/TCP 5m23s
myjitsi-prosody ClusterIP 10.96.250.39 <none> 5280/TCP,5281/TCP,5347/TCP,5222/TCP,5269/TCP 5m23s
opening a Chrome browser to cluster's external IP @PORT 8443 works (Jitsi portal seen)
opening a Chrome browser to cluster's external IP @PORT 8444 works (JSON colibri ststa are retrieved)
creating a new meeting works
however - the user can't connect to the meeting.
repeating reconnect messages.
same from other browsers on other hosts trying to connect.
any ideas ?
so I tried checking (with tcpdump ...) what's going on inside the web pod and it seem to try and communicate with the prosody pod but with no success.
then i noticed that one of the 4 Jitsi pods is stuck in "Pending"
checking its log showd a "FailedScheduling" warning
[d50c-cml-app16 JITSI]$ kubectl get pod
NAME READY STATUS RESTARTS AGE
myjitsi-jitsi-meet-jicofo-64dcdb55d8-6d24t 1/1 Running 0 22s
myjitsi-jitsi-meet-jvb-7f6544f744-nmpfk 1/1 Running 0 22s
myjitsi-jitsi-meet-web-7f98b75c4d-mp229 1/1 Running 0 22s
myjitsi-prosody-0 0/1 Pending 0 22s
[d50c-cml-app16 JITSI]$ kubectl describe pod myjitsi-prosody-0
Name: myjitsi-prosody-0
Namespace: default
Priority: 0
Node: <none>
Labels: app.kubernetes.io/instance=myjitsi
app.kubernetes.io/name=prosody
controller-revision-hash=myjitsi-prosody-65fd8ddb96
statefulset.kubernetes.io/pod-name=myjitsi-prosody-0
Annotations: <none>
Status: Pending
IP:
IPs: <none>
Controlled By: StatefulSet/myjitsi-prosody
Containers:
prosody:
Image: jitsi/prosody:stable-8044
Ports: 5222/TCP, 5269/TCP, 5347/TCP, 5280/TCP, 5281/TCP
Host Ports: 0/TCP, 0/TCP, 0/TCP, 0/TCP, 0/TCP
Liveness: http-get http://:bosh-insecure/http-bind delay=0s timeout=1s period=10s #success=1 #failure=3
Readiness: http-get http://:bosh-insecure/http-bind delay=0s timeout=1s period=10s #success=1 #failure=3
Environment Variables from:
myjitsi-prosody ConfigMap Optional: false
myjitsi-prosody Secret Optional: false
myjitsi-prosody-jicofo Secret Optional: false
myjitsi-prosody-jvb Secret Optional: false
myjitsi-prosody-common ConfigMap Optional: false
Environment: <none>
Mounts:
/config/data from prosody-data (rw)
/var/run/secrets/kubernetes.io/serviceaccount from kube-api-access-sdrj6 (ro)
Conditions:
Type Status
PodScheduled False
Volumes:
prosody-data:
Type: PersistentVolumeClaim (a reference to a PersistentVolumeClaim in the same namespace)
ClaimName: prosody-data-myjitsi-prosody-0
ReadOnly: false
kube-api-access-sdrj6:
Type: Projected (a volume that contains injected data from multiple sources)
TokenExpirationSeconds: 3607
ConfigMapName: kube-root-ca.crt
ConfigMapOptional: <nil>
DownwardAPI: true
QoS Class: BestEffort
Node-Selectors: <none>
Tolerations: node.kubernetes.io/not-ready:NoExecute op=Exists for 300s
node.kubernetes.io/unreachable:NoExecute op=Exists for 300s
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Warning FailedScheduling 34s default-scheduler 0/2 nodes are available: 2 pod has unbound immediate PersistentVolumeClaims.
Warning FailedScheduling 33s default-scheduler 0/2 nodes are available: 2 pod has unbound immediate PersistentVolumeClaims.
[d50c-cml-app16 JITSI]$
so apparently it's a K8s PVC issue
[d50c-cml-app16 JITSI]$ kubectl get pvc
NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE
prosody-data-myjitsi-prosody-0 Pending 9h
[d50c-cml-app16 JITSI]$ kubectl describe pvc prosody-data-myjitsi-prosody-0
Name: prosody-data-myjitsi-prosody-0
Namespace: default
StorageClass:
Status: Pending
Volume:
Labels: app.kubernetes.io/instance=myjitsi
app.kubernetes.io/name=prosody
Annotations: <none>
Finalizers: [kubernetes.io/pvc-protection]
Capacity:
Access Modes:
VolumeMode: Filesystem
Used By: myjitsi-prosody-0
Events:
Type Reason Age From Message
---- ------ ---- ---- -------
Normal FailedBinding 16s (x2202 over 9h) persistentvolume-controller no persistent volumes available for this claim and no storage class is set
closing this issue.