Kubernetes cluster recipes not seeing nodes

fred974

@olivierlambert said in Kubernetes cluster recipes not seeing nodes:

Your first Debian VM isn't booting at all. Are you sure your host is working and supporting HVM guest?

Yes, the hosts has other VM running on them. Also on my original post, the master was created and also the nodes. They simply couldn't see each other

@olivierlambert said in Kubernetes cluster recipes not seeing nodes:

Can you boot the Debian 11 template from the XO Hub?

Yes, I can deploy Debian 11 from the XO Hub but it it won't stay up. The Vm does start and get stuck at the boot process

vm.stats
{
  "id": "9082c067-e945-0ee0-2aca-4b46ab5c02be"
}
{
  "message": "Cannot read properties of undefined (reading 'statusCode')",
  "name": "TypeError",
  "stack": "TypeError: Cannot read properties of undefined (reading 'statusCode')
    at when (/usr/local/lib/node_modules/xo-server/node_modules/xen-api/src/index.js:415:41)
    at matchError (/usr/local/lib/node_modules/xo-server/node_modules/promise-toolbox/_matchError.js:17:103)
    at onError (/usr/local/lib/node_modules/xo-server/node_modules/promise-toolbox/retry.js:64:9)
    at AsyncResource.runInAsyncScope (node:async_hooks:203:9)
    at cb (/usr/local/lib/node_modules/xo-server/node_modules/bluebird/js/release/util.js:355:42)
    at tryCatcher (/usr/local/lib/node_modules/xo-server/node_modules/bluebird/js/release/util.js:16:23)
    at Promise._settlePromiseFromHandler (/usr/local/lib/node_modules/xo-server/node_modules/bluebird/js/release/promise.js:547:31)
    at Promise._settlePromise (/usr/local/lib/node_modules/xo-server/node_modules/bluebird/js/release/promise.js:604:18)
    at Promise._settlePromise0 (/usr/local/lib/node_modules/xo-server/node_modules/bluebird/js/release/promise.js:649:10)
    at Promise._settlePromises (/usr/local/lib/node_modules/xo-server/node_modules/bluebird/js/release/promise.js:725:18)
    at _drainQueueStep (/usr/local/lib/node_modules/xo-server/node_modules/bluebird/js/release/async.js:93:12)
    at _drainQueue (/usr/local/lib/node_modules/xo-server/node_modules/bluebird/js/release/async.js:86:9)
    at Async._drainQueues (/usr/local/lib/node_modules/xo-server/node_modules/bluebird/js/release/async.js:102:5)
    at Immediate.Async.drainQueues [as _onImmediate] (/usr/local/lib/node_modules/xo-server/node_modules/bluebird/js/release/async.js:15:14)
    at processImmediate (node:internal/timers:471:21)
    at process.callbackTrampoline (node:internal/async_hooks:130:17)"
}

I removed the template and re-installed it again. This time i successfully managed to deploy Debian 11 from XO Hub.
I then tried the Kubernetes recipe again and this time the master is created but is not powering on. Should I manually start the VM?

fred974

@olivierlambert

xoa.recipe.createKubernetesCluster
{
  "masterName": "K8S Master",
  "nbNodes": 3,
  "network": "37559e16-f83d-b1f4-a1ae-b9c3336dca22",
  "sr": "a20ee08c-40d0-9818-084f-282bbca1f217",
  "sshKey": "ssh-ed25519 AAAAC3NzaC1lZDI1Nhbilj;jojj'hufsdklk,kp;joFWlpk8BlpKpTO3T2R44YHP6b+g Kubernetes cluster"
}
{
  "code": "SR_BACKEND_FAILURE_46",
  "params": [
    "",
    "The VDI is not available [opterr=Failed to attach VDI during \"prepare thin\": The VDI is not available [opterr=Plugin linstor-manager failed]]",
    ""
  ],
  "call": {
    "method": "VM.start",
    "params": [
      "OpaqueRef:6ae3475f-7455-4328-904d-168b35b8ce5f",
      false,
      false
    ]
  },
  "message": "SR_BACKEND_FAILURE_46(, The VDI is not available [opterr=Failed to attach VDI during \"prepare thin\": The VDI is not available [opterr=Plugin linstor-manager failed]], )",
  "name": "XapiError",
  "stack": "XapiError: SR_BACKEND_FAILURE_46(, The VDI is not available [opterr=Failed to attach VDI during \"prepare thin\": The VDI is not available [opterr=Plugin linstor-manager failed]], )
    at Function.wrap (/usr/local/lib/node_modules/xo-server/node_modules/xen-api/src/_XapiError.js:16:12)
    at /usr/local/lib/node_modules/xo-server/node_modules/xen-api/src/transports/json-rpc.js:35:21"
}

olivierlambert

Okay so your problem isn't k8s recipe related, but elsewhere (probably the storage).

fred974

@olivierlambert said in Kubernetes cluster recipes not seeing nodes:

Okay so your problem isn't k8s recipe related, but elsewhere (probably the storage).

Yup.. I just ran the K8S recipe on local storage and I have cluster..

debian@master:~$ kubectl get node
NAME     STATUS   ROLES           AGE     VERSION
master   Ready    control-plane   20m     v1.26.3
node-1   Ready    <none>          5m24s   v1.26.3
node-2   Ready    <none>          7m32s   v1.26.3
node-3   Ready    <none>          4m18s   v1.26.3

Issue seem to be with the XOSTOR SR. Any chance @ronan-a could test if the K8S recipe work on his XOSTOR SR? I just want to know if it is specific to my setup or XOSTOR in general (bug)

olivierlambert

I think it's totally unrelated. If you have issues to create the XO Hub Debian 11 VM, nothing else could work after that.

fred974

@olivierlambert said in Kubernetes cluster recipes not seeing nodes:

If you have issues to create the XO Hub Debian 11 VM, nothing else could work after that.

What do you mean? I can create XO Hub Debian 11 on XOSTOR with no issue. I am not sure I understand you ?

olivierlambert

What I mean is the fact that the issue with your k8s recipe is not related to the recipe itself, but a problem on your own XOSTOR cluster.

fred974

@olivierlambert said in Kubernetes cluster recipes not seeing nodes:

What I mean is the fact that the issue with your k8s recipe is not related to the recipe itself, but a problem on your own XOSTOR cluster.

How can we be sure its not a bug and only and issue with my system? This is way i was asking if anyone else running XOSTOR could try to confirm the issue is with my system

olivierlambert

Because we identified some issues on XOSTOR recently, and that's exactly what causing your issue. If the VM doesn't boot, there's 0 chance the k8s recipe can work.

fred974

@olivierlambert ok so is it a matter to create K8S on local storage and moved them manually to XOSTOR untill XOSTOR issue get a fix?

olivierlambert

I would advise to not combine problems, XOSTOR is not completely bug free yet, so if you have an issue to report, get rid of XOSTOR in the equation first.

fred974

@olivierlambert Ok will do. Kubernetes is purely to learn .. so we good. Thank you

fred974

@Raymondello you right, the issue is from the storage and I am looking into that. Thank you for sharing the UAT Testing, I didn't know about it