Hi Team,
I have integrated google cloud on two different Morpheus data server, one is 3 NODES HA WITH Nfs shared file system, and one is single node all in one…On both server cluster is deploying perfectly with all control and worker nodes, but it does not show any deployments, replica set container Kubernetes command is also not running…I’m attaching error which comes while running the kubectl command below…Can anyone please guide me what’s the real issue is
SINGLE NODE AIO:
ERROR:
3 NODE HA WITH NFS:
First, doublecheck that all the relevant APIs are enabled for the service accounts used when adding the clouds.
You can find those listed here.
https://docs.morpheusdata.com/en/latest/integration_guides/Clouds/google/google.html?highlight=gcp%20permission#requirements-for-integration-with-morpheus
Also would you have any services like SELinux, Apparmor or Carbon Black running. Looks like there is an issue related to managing file ownership.
Thankyou for the quick response, I have double checked all the prerequisite API’s and they enabled as mentioned in the document moreover the file permission error I m getting is on 3node HA server on which nfs share is mounted with /var/opt/morpheus/morpheus-ui…
a quick update I have resolved the permission error but still getting segmentation fault error on both 3node and single node AIO
Error on UI:
Morpheus UI LOGS:
Looking up errors around a segmentation fault and kubectl mentions that it might be a corrupt binary. Since you’ve had permissions issues, it may be related to that as well, it will be difficult to tell for sure. Note that following the below steps is at your own risk, if you feel comfortable with the process. Otherwise, creating a support ticket would be best (https://support.morpheusdata.com/).
My initial idea would be to move/backup the /var/opt/morpheus/morpheus-ui/workspace/kubernetes/runtime/41a1538e-e545-44ec-8932-d9fc700e1872/ directory using the following command:
mv /var/opt/morpheus/morpheus-ui/workspace/kubernetes/runtime/41a1538e-e545-44ec-8932-d9fc700e1872/ /var/opt/morpheus/morpheus-ui/workspace/kubernetes/runtime/41a1538e-e545-44ec-8932-d9fc700e1872-bak/
You could also remove the directory but just playing it safe. Once it is no longer available, when you run a new get nodes command, it should regenerate that folder with the config and kubectl binary and run the command.
That said, since you had this issue on two different systems (hopefully not using the same shared NFS), I think there might be a different issue going on. If you continue to receive the error, I’d recommend opening a support case, as that would be the best avenue to troubleshoot errors:
https://support.morpheusdata.com/
2 Likes
Thanks, I followed this and my issue is resolved…
