• EN
Question

k10 dashboard access

  • 29 February 2024
  • 4 comments
  • 56 views
E
Userlevel 1

Hello,

I try to have acces to the K10 dashboard outside my k8s cluster labs,

I try the loadbalancer option but I can’t have an external ip (always in pending)

kubectl get svc gateway-ext --namespace kasten-io -o wide
NAME          TYPE           CLUSTER-IP      EXTERNAL-IP   PORT(S)        AGE   SELECTOR
gateway-ext   LoadBalancer   10.97.236.105   <pending>     80:32244/TCP   52s   service=gateway

 

All my pods in kasten-io are running:

kubectl get pods -n kasten-io
NAME                                     READY   STATUS    RESTARTS   AGE
aggregatedapis-svc-58ff985f69-lbksl      1/1     Running   0          21s
auth-svc-8899c7bfc-g2b7h                 1/1     Running   0          22s
catalog-svc-5d4fd45787-wb5wx             2/2     Running   0          22s
controllermanager-svc-57dfb767b4-lw5hf   1/1     Running   0          22s
crypto-svc-576b9b56d6-b57bl              4/4     Running   0          22s
dashboardbff-svc-7fb956f4dc-pjvtb        2/2     Running   0          22s
executor-svc-76c6857865-42qwv            1/1     Running   0          22s
executor-svc-76c6857865-59pxw            1/1     Running   0          22s
executor-svc-76c6857865-vbr2d            1/1     Running   0          22s
frontend-svc-cfbcc679d-92hxs             1/1     Running   0          21s
gateway-c5b5ffb45-5xkjb                  0/1     Running   0          20s
jobs-svc-6dd769fb7c-9n9wm                1/1     Running   0          22s
k10-grafana-58cc78b885-94qnw             1/1     Running   0          23s
kanister-svc-64ffbfbf5c-kf7xj            1/1     Running   0          20s
logging-svc-8646cdbfdf-dbs44             1/1     Running   0          22s
metering-svc-cd568f7d9-svmds             1/1     Running   0          21s
prometheus-server-7cfc5fcfbf-jwdbx       1/2     Running   0          23s
state-svc-7bd7f77d-6svt9                 3/3     Running   0          21s

 

 



I install kasten with these options:

helm install k10 kasten/k10 --namespace=kasten-io --set global.persistence.storageClass=something-sc \
> --set externalGateway.create=true \
> --set auth.tokenAuth.enabled=true

 

I’ve try this:

http://ipworker:32244/k10

 

and I have a blank page

 

and to test the creation of a token I have an error too:

kubectl --namespace kasten-io create token my-kasten-sa --duration=24h
error: failed to create token: serviceaccounts "my-kasten-sa" not found

 

So I don’t know what to do next or I misunderstanding something.

Thanks in advance for your help

4 comments

C

Make sure you have a loadbalancer like Metallb running on your cluster. This will assign IP address to your Loadbalancers service or add fixed ip address to the gateway-ext service

 

regarding service account :

#first create service account:

kubectl create serviceaccount my-kast-sa -n kasten-io

# add k10-admin cluster role to serviceaccount

kubectl create clusterrolebinding k10-admin-sa --clusterrole=k10-admin --serviceaccount=kasten-io:my-kasten-sa

# add cluster-admin role (bit rough) to service account

kubectl create clusterrolebinding k10-admin-cluster --clusterrole=cluster-admin --serviceaccount=kasten-io:kasten-sa
 

# rerun the token procedure

 

E
Userlevel 1

sorry for the late answer

I’ve installed metallb, try a loadbalancer port with a nginx deployment with no problem but with kasten i see the gateways pods crashing and I have theses errors on one gateway pod:

Events:
  Type     Reason                  Age                   From               Message
  ----     ------                  ----                  ----               -------
  Normal   Scheduled               22m                   default-scheduler  Successfully assigned kasten-io/gateway-68fbc949dc-zjxb9 to k8s-worker02
  Normal   Pulling                 22m                   kubelet            Pulling image "gcr.io/kasten-images/emissary:6.5.8"
  Normal   Pulled                  22m                   kubelet            Successfully pulled image "gcr.io/kasten-images/emissary:6.5.8" in 34.133s (34.688s including waiting)
  Normal   Created                 22m                   kubelet            Created container ambassador
  Normal   Started                 22m                   kubelet            Started container ambassador
  Warning  FailedCreatePodSandBox  19m                   kubelet            Failed to create pod sandbox: rpc error: code = Unknown desc = failed to setup network for sandbox "4b954eddb0b2d442577fe75ca74d6b035f6f441a9b22d683e5adbad585e13003": plugin type="flannel" failed (add): loadFlannelSubnetEnv failed: open /run/flannel/subnet.env: no such file or directory
  Warning  FailedCreatePodSandBox  19m                   kubelet            Failed to create pod sandbox: rpc error: code = Unknown desc = failed to setup network for sandbox "a3602588ba6dec7d3ff3c6391daaf1837c4de3a501e4294acb9374fe1cabce1c": plugin type="flannel" failed (add): loadFlannelSubnetEnv failed: open /run/flannel/subnet.env: no such file or directory
  Warning  FailedCreatePodSandBox  19m                   kubelet            Failed to create pod sandbox: rpc error: code = Unknown desc = failed to setup network for sandbox "0207cd0a7e68ddbc359c8d786a7c8a358fd7f044615f81ee14f62bd9faa9027f": plugin type="flannel" failed (add): loadFlannelSubnetEnv failed: open /run/flannel/subnet.env: no such file or directory
  Normal   SandboxChanged          18m (x4 over 19m)     kubelet            Pod sandbox changed, it will be killed and re-created.
  Normal   Pulled                  18m                   kubelet            Container image "gcr.io/kasten-images/emissary:6.5.8" already present on machine
  Normal   Created                 18m                   kubelet            Created container ambassador
  Normal   Started                 18m                   kubelet            Started container ambassador
  Warning  Unhealthy               17m (x3 over 17m)     kubelet            Liveness probe failed: HTTP probe failed with statuscode: 503
  Normal   Killing                 17m                   kubelet            Container ambassador failed liveness probe, will be restarted
  Warning  BackOff                 9m12s (x29 over 15m)  kubelet            Back-off restarting failed container ambassador in pod gateway-68fbc949dc-zjxb9_kasten-io(e37a91e0-2268-40d4-9377-a8552a02d72a)
  Warning  Unhealthy               3m56s (x40 over 18m)  kubelet            Readiness probe failed: HTTP probe failed with statuscode: 503

 

but now my loadbalancer port in kasten-io namespace is OK :

k get -n kasten-io svc gateway-ext -o wide
NAME          TYPE           CLUSTER-IP       EXTERNAL-IP     PORT(S)        AGE     SELECTOR
gateway-ext   LoadBalancer   10.101.224.229   192.168.1.241   80:31856/TCP   8m29s   service=gateway

 

and all other kasten pods are ok:

 kubectl get pods --namespace kasten-io
NAME                                    READY   STATUS             RESTARTS         AGE
aggregatedapis-svc-89d76759c-65gkw      1/1     Running            0                13m
auth-svc-6dcf6685cc-84w9f               1/1     Running            0                13m
catalog-svc-85478b567b-jdnkd            2/2     Running            0                12m
controllermanager-svc-99f87f7f6-9n7fv   1/1     Running            0                13m
crypto-svc-6bfbfc4c54-b9b8t             4/4     Running            0                13m
dashboardbff-svc-7b54b8f649-k4f9l       2/2     Running            0                13m
executor-svc-5978df4789-4vmtv           1/1     Running            0                13m
executor-svc-5978df4789-cnbgm           1/1     Running            0                13m
executor-svc-5978df4789-ctj6f           1/1     Running            0                13m
frontend-svc-64c49d55bb-cnrdj           1/1     Running            0                13m
gateway-5775954c65-zmkgb                0/1     CrashLoopBackOff   7 (4m32s ago)    13m
gateway-68fbc949dc-zjxb9                0/1     CrashLoopBackOff   10 (4m31s ago)   29m
jobs-svc-f75878f54-x6h4w                1/1     Running            0                13m
k10-grafana-7687dfc495-rq7ch            1/1     Running            0                13m
kanister-svc-598d7688b5-glgd7           1/1     Running            0                13m
logging-svc-7668586c9d-ddrst            1/1     Running            0                13m
metering-svc-7cb6fc9bc7-pjkm7           1/1     Running            0                13m
prometheus-server-85d896ddb7-pj94d      2/2     Running            0                29m
state-svc-6475f95b88-qrvsv              3/3     Running            0                13m

 

 

 

What I don’t understand is why my flannel network plugin seems in error when all other deployments with a loadbalancer port working correctly

 

E
Userlevel 1

OK, forget about my last reply
I have removed my kasten deployment
reinstall it with helm and now have only one gateway pod
here’s the log from the gateway pod:
https://pastebin.com/M3bSLq2w

And here’s what I see in the events:

Events:
  Type     Reason                  Age                   From               Message
  ----     ------                  ----                  ----               -------
  Normal   Scheduled               23m                   default-scheduler  Successfully assigned kasten-io/gateway-5775954c65-6sz8f to k8s-worker01
  Normal   Created                 22m (x2 over 23m)     kubelet            Created container ambassador
  Normal   Started                 22m (x2 over 23m)     kubelet            Started container ambassador
  Normal   Pulled                  21m (x3 over 23m)     kubelet            Container image "gcr.io/kasten-images/emissary:6.5.8" already present on machine
  Warning  Unhealthy               21m (x6 over 22m)     kubelet            Liveness probe failed: HTTP probe failed with statuscode: 503
  Normal   Killing                 21m (x2 over 22m)     kubelet            Container ambassador failed liveness probe, will be restarted
  Warning  Unhealthy               18m (x31 over 22m)    kubelet            Readiness probe failed: HTTP probe failed with statuscode: 503
  Warning  FailedCreatePodSandBox  13m                   kubelet            Failed to create pod sandbox: rpc error: code = Unknown desc = failed to setup network for sandbox "73515b7a039ead65b02ecb4c62ef61ed4aae456fcb553ddfe82413d5a0e7931a": plugin type="flannel" failed (add): loadFlannelSubnetEnv failed: open /run/flannel/subnet.env: no such file or directory
  Warning  FailedCreatePodSandBox  13m                   kubelet            Failed to create pod sandbox: rpc error: code = Unknown desc = failed to setup network for sandbox "3af05071bf11cc72e65de617c8ce8606a2c9b73537e748a205929e7748fef397": plugin type="flannel" failed (add): loadFlannelSubnetEnv failed: open /run/flannel/subnet.env: no such file or directory
  Warning  FailedCreatePodSandBox  13m                   kubelet            Failed to create pod sandbox: rpc error: code = Unknown desc = failed to setup network for sandbox "33714bc5a5c2ce3c005402fc0896e7477d6f66c76680b01da5ff31e1b42c32d8": plugin type="flannel" failed (add): loadFlannelSubnetEnv failed: open /run/flannel/subnet.env: no such file or directory
  Normal   SandboxChanged          13m (x4 over 13m)     kubelet            Pod sandbox changed, it will be killed and re-created.
  Normal   Pulled                  12m (x2 over 13m)     kubelet            Container image "gcr.io/kasten-images/emissary:6.5.8" already present on machine
  Normal   Created                 12m (x2 over 13m)     kubelet            Created container ambassador
  Normal   Started                 12m (x2 over 13m)     kubelet            Started container ambassador
  Normal   Killing                 12m                   kubelet            Container ambassador failed liveness probe, will be restarted
  Warning  Unhealthy               12m (x4 over 12m)     kubelet            Liveness probe failed: HTTP probe failed with statuscode: 503
  Warning  Unhealthy               12m (x7 over 13m)     kubelet            Readiness probe failed: HTTP probe failed with statuscode: 503
  Warning  BackOff                 3m41s (x31 over 11m)  kubelet            Back-off restarting failed container ambassador in pod gateway-5775954c65-6sz8f_kasten-io(523dc826-915a-4551-88ba-00561a827676)

 

 

E
Userlevel 3
Badge +1

Hello @eztech 

 

So, I am looking over this issue and I am noticing that it looks to be you are having problems in your environment with your CNI and the Gateway-svc pods.

 

failed to setup network for sandbox "33714bc5a5c2ce3c005402fc0896e7477d6f66c76680b01da5ff31e1b42c32d8": plugin type="flannel" failed (add): loadFlannelSubnetEnv failed: open /run/flannel/subnet.env: no such file or directory

 

I would recommend taking a look over flannel logs during the time the Gateway Pod attempts to start.

 

Thanks

Emmanuel

Comment


Terms & Conditions