search

ceph / ceph-helm

Curated applications for Kubernetes
Apache License 2.0
108 stars 36 forks source link

Deploying ceph multiple times causes authentication error. #43

Closed saitejar closed 6 years ago

saitejar commented 6 years ago

Is this a request for help?: Yes

Is this a BUG REPORT or FEATURE REQUEST? (choose one): BUG REPORT

Version of Helm and Kubernetes: kubernetes 1.8.3

Which chart: ceph

What happened: I have a kubernetes deployment running. When I try to deploy ceph for the first time, it works fine. But when I purge ceph helm chart and redeploy ceph, ceph-mon, ceph-mon-check works fine but ceph-mds, ceph-mgr have an authentication failure and timeout connecting to ceph-mon.

What you expected to happen: I would expect that all the services are able to connect to monitor without any authentication issues.

How to reproduce it (as minimally and precisely as possible): Deploy ceph, purge ceph chart, deploy ceph again.

Anything else we need to know: Something to note: I am not deleting the hostpath folder in the nodes on which ceph is being re deployed. I was expecting that redeploying would persist the content of ceph.

alram commented 6 years ago

You need to delete /var/lib/ceph-helm/ceph/mon/* manually on each ceph-mon host before a re-run. We haven't been able to work on fixing this yet

saitejar commented 6 years ago

@alram thank you. Would that persist the data ?

alram commented 6 years ago

No if you redeploy there's no persistence, you'll lose your data.