The --abort function acts differently with -g option. about gluster-kubernetes HOT 6 CLOSED

...yes? This is intentional so as to allow gk-deploy to delete the GlusterFS pods if they were improperly created.

from gluster-kubernetes.

@jarrpa so the situation we faced is when we are deploying everything and it fails. abort was called without ${GLUSTER}=1. I am not sure why. so abort will always have ${GLUSTER}=1 when it is internally called is it.

from gluster-kubernetes.

@jarrpa I mean when we do gk-deploy -g

from gluster-kubernetes.

I'm still not sure I understand the problem. When you call gk-deploy --abort -g you will always have GLUSTER=1, and thus it will try to take down any existing GlusterFS pods and related resources. if you don't specify -g with --abort then it won't try to do that, as no state is persisted between calls of gk-deploy. What am I missing?

from gluster-kubernetes.

@jarrpa You are not missing anything. We thought there was a issue when we did it. I now understand it. We can close this issue. As it was just a confusion.

from gluster-kubernetes.

Ah, whew. No worries! :)

from gluster-kubernetes.