-
Notifications
You must be signed in to change notification settings - Fork 44
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Example is not working #99
Comments
[root@node1 opt]# kubectl get jv pvc-5ea4b089-9617-402e-9a40-9a5d2d69be50 -n openebs NAME REPLICACOUNT PHASE STATUS |
[root@node1 opt]# kubectl get po,pvc NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS AGE |
Hi @cleveritcz Can you please share the output for |
[root@node1 opt]# kubectl get pods -n openebs |
Yes I don't see the replica and controller pod here. Can you share the output for |
[root@node1 opt]# kubectl -n openebs get all NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE NAME READY UP-TO-DATE AVAILABLE AGE NAME DESIRED CURRENT READY AGE NAME READY AGE |
I used these commands:
I don't know why storageClass has to be twice |
Was the volume deleted later. I see in the logs the resources got created but are missing from openebs namespace. The host-path storageclass is for the replica STS which is used for local hostpath storage for each replica |
I am sorry, I can't move to slack. I would like to get work it. |
No problem. Can you delete the old volumes and apply the hostpath storageclass mentioned in the quickstart guide. Then it should work |
I deleted pv, pvc and hostpath storage class and applied again hostpath storage class. [root@node1 opt]# kubectl -n openebs get all NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE NAME READY UP-TO-DATE AVAILABLE AGE NAME DESIRED CURRENT READY AGE NAME READY AGE |
I still don't see the ctrl and rep pods for the volume. Can you share the logs for jiva-operator pod. |
[root@node1 opt]# kubectl logs -f jiva-operator-6b69b86894-dr4p6 -n openebs |
[root@node1 opt]# kubectl -n openebs get all NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE NAME READY UP-TO-DATE AVAILABLE AGE NAME DESIRED CURRENT READY AGE NAME READY AGE |
[root@node1 opt]# kubectl get jivavolume -n openebs |
Okay now the pods got created. Can you share the output for -oyaml of the jv resource? How many nodes do are there in this cluster? Also is the networking between the nodes working properly? |
3 nodes in cluster (1 master/2 workers). I don't know if networking is ok, I deployed it by kubespray with Cilium network plugin. |
So the volume got created here. It is just that the ctrl and rep cannot connect due to network issue. |
[root@node1 opt]# kubectl -n kube-system get all NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE NAME READY UP-TO-DATE AVAILABLE AGE NAME DESIRED CURRENT READY AGE |
I haven't used kubespray or cilium but was looking for similar issues. Found this kubernetes-sigs/kubespray#4357 (comment) |
ok, I will try to redeploy it. |
I redeployed it with ingress_nginx_host_network: true and still is not working |
[root@node1 ~]# kubectl logs -f jiva-operator-6b69b86894-wbkps -n openebs |
maybe some rbac is missing |
Can you describe the pod |
root@node1 ~]# kubectl logs -f pvc-8a754aae-4a0f-48ca-a237-fd88f4089254-jiva-rep-0 -n openebs it is crashing |
I don't have domain cluster.local |
Lets try a thing. Can you exec into the replica pod and try to curl the service IP for the volume.
This will help us rule out service discovery problems |
I changed domain to cluster.local and pod is running. I am getting still this error: [root@node1 ~]# kubectl logs -f pvc-2b697d2e-e561-4acf-b264-60fd6cd392e1-jiva-rep-0 -n openebs
|
Can you share the -oyaml of the jv. The error will appear in the beginning because the repica pod take some time to come to Running state and should go away in jiva-operator after jv comes to Ready state |
|
I don't know if it is working for production use case. |
everything is up and running. I would like to use anothe domain than cluster.local. Where could I change it, please? [root@node1 ~]# kubectl -n openebs get all NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE NAME DESIRED CURRENT READY UP-TO-DATE AVAILABLE NODE SELECTOR AGE NAME READY UP-TO-DATE AVAILABLE AGE NAME DESIRED CURRENT READY AGE NAME READY AGE |
Thanks for reporting this issue. I was able to reproduce it locally. I will add this to the next release tracker and try to fix it. It is a small fix we have to replace this hard coded value https://github.com/openebs/jiva-operator/blob/811a3af7c34aa13c487f7e4a84c4704cf8f94100/pkg/controllers/jivavolume_controller.go#L75 Also if you have a single replica use case you can directly use localpv hostpath |
and what about this error? time="2021-06-16T12:09:07Z" level=info msg="Failed to get volume stats errGet "http://10.233.57.251:9501/v1/stats\": dial tcp 10.233.57.251:9501: connect: connection refused" |
This I need to look for a little more this is code is trying to watch the volume status by trying to png for the stats. This can happen because of a lot of reasons. But these errors are harmless in most cases |
After changing the cluster domain and the volume became Ready are they still appearing in the log? |
yes |
Hi @cleveritcz the error for the cluster domain got fixed in the latest release of jiva-operator. You can find the operator yamls here For the other error it is something that may happen due to some reachability issue. The volume is able to mount so the GET for the stats is successful but I still need to dig a little more. |
Hi @cleveritcz I was able to find the cause of the second error that was flooding the logs and is fixed by the PR #112. These fixes are now available In 2.11.0 release. Closing this issue for now, please feel free to reopen if any errors occur. |
What steps did you take and what happened:
[A clear and concise description of what the bug is, and what commands you ran.)
I followed manual for Quickstart deploy and I couldn't get running example app. The pvc didn't bound it.
What did you expect to happen:
I expect example app working
The output of the following commands will help us better understand what's going on:
(Pasting long output into a GitHub gist or other pastebin is fine.)
[root@node1 opt]# kubectl logs -f jiva-operator-6b69b86894-5lmht -n openebs
I0615 19:16:01.931555 1 request.go:655] Throttling request took 1.046617453s, request: GET:https://10.233.0.1:443/apis/cert-manager.io/v1alpha2?timeout=32s
time="2021-06-15T19:16:02Z" level=info msg="Go Version: go1.14.7"
time="2021-06-15T19:16:02Z" level=info msg="Go OS/Arch: linux/amd64"
time="2021-06-15T19:16:02Z" level=info msg="Version of jiva-operator: 2.9.0"
time="2021-06-15T19:16:05Z" level=info msg="start bootstraping jiva componentsJivaVolume: pvc-9331eb72-6657-4145-8a97-0e7a2768d199"
time="2021-06-15T19:16:05Z" level=info msg="Creating a new serviceService.NamespaceopenebsService.Namepvc-9331eb72-6657-4145-8a97-0e7a2768d199-jiva-ctrl-svc"
time="2021-06-15T19:16:06Z" level=info msg="Updating JivaVolume with iscsi specISCSISpec{10.233.49.69 3260 iqn.2016-09.com.openebs.jiva:pvc-9331eb72-6657-4145-8a97-0e7a2768d199}"
time="2021-06-15T19:16:06Z" level=info msg="Creating a new deploymentDeploy.NamespaceopenebsDeploy.Namepvc-9331eb72-6657-4145-8a97-0e7a2768d199-jiva-ctrl"
time="2021-06-15T19:16:06Z" level=info msg="Creating a new StatefulsetStatefulset.NamespaceopenebsSts.Namepvc-9331eb72-6657-4145-8a97-0e7a2768d199-jiva-rep"
time="2021-06-15T19:16:07Z" level=info msg="Creating a new pod disruption budgetPdb.NamespaceopenebsPdb.Namepvc-9331eb72-6657-4145-8a97-0e7a2768d199-pdb"
time="2021-06-15T19:16:07Z" level=info msg="start bootstraping jiva componentsJivaVolume: pvc-5ea4b089-9617-402e-9a40-9a5d2d69be50"
time="2021-06-15T19:16:07Z" level=info msg="Creating a new serviceService.NamespaceopenebsService.Namepvc-5ea4b089-9617-402e-9a40-9a5d2d69be50-jiva-ctrl-svc"
time="2021-06-15T19:16:08Z" level=info msg="Updating JivaVolume with iscsi specISCSISpec{10.233.5.29 3260 iqn.2016-09.com.openebs.jiva:pvc-5ea4b089-9617-402e-9a40-9a5d2d69be50}"
time="2021-06-15T19:16:08Z" level=info msg="Creating a new deploymentDeploy.NamespaceopenebsDeploy.Namepvc-5ea4b089-9617-402e-9a40-9a5d2d69be50-jiva-ctrl"
time="2021-06-15T19:16:08Z" level=info msg="Creating a new StatefulsetStatefulset.NamespaceopenebsSts.Namepvc-5ea4b089-9617-402e-9a40-9a5d2d69be50-jiva-rep"
time="2021-06-15T19:16:08Z" level=info msg="Creating a new pod disruption budgetPdb.NamespaceopenebsPdb.Namepvc-5ea4b089-9617-402e-9a40-9a5d2d69be50-pdb"
time="2021-06-15T19:16:08Z" level=info msg="Failed to get volume stats errGet "http://10.233.49.69:9501/v1/stats\": dial tcp 10.233.49.69:9501: connect: connection refused"
time="2021-06-15T19:16:08Z" level=info msg="Failed to get volume stats errGet "http://10.233.5.29:9501/v1/stats\": dial tcp 10.233.5.29:9501: connect: connection refused"
time="2021-06-15T19:16:09Z" level=info msg="Failed to get volume stats errGet "http://10.233.49.69:9501/v1/stats\": dial tcp 10.233.49.69:9501: connect: connection refused"
time="2021-06-15T19:16:09Z" level=info msg="Failed to get volume stats errGet "http://10.233.5.29:9501/v1/stats\": dial tcp 10.233.5.29:9501: connect: connection refused"
time="2021-06-15T19:16:11Z" level=info msg="Failed to get volume stats errGet "http://10.233.5.29:9501/v1/stats\": dial tcp 10.233.5.29:9501: connect: connection refused"
Anything else you would like to add:
[Miscellaneous information that will assist in solving the issue.]
Environment:
/etc/os-release
): centos 7.9The text was updated successfully, but these errors were encountered: