-
Notifications
You must be signed in to change notification settings - Fork 67
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
[nova] switch to projected volumes #2061
base: master
Are you sure you want to change the base?
Conversation
633b62e
to
61de3ae
Compare
This commit needs much more info. At least the JIRA issue number, but probably even better something from fe6e986 Also please rebase. |
- name: nova-bin | ||
mountPath: /var/lib/kolla/venv/bin |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Did you confirm this works? Is everything else still there in /var/lib/kolla/venv/bin
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
It seems that here we need to keep using subPath
otherwise the bin
directory will get overwritten. I think there's no problem with using subPath
, since this is the only place where nova-bin is being used. Then does it still make sense to project a volume here? Probably not.
items: | ||
- key: iptables-restore.mock | ||
path: iptables-restore | ||
name: nova-etc |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should be nova-bin
.
items: | ||
- key: nova.conf | ||
path: nova.conf | ||
- key: nova-api.conf |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should be nova-api-metadata.conf
items: | ||
- key: iptables-restore.mock | ||
path: iptables-restore | ||
name: nova-etc |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Should be nova-bin
.
- mountPath: /etc | ||
name: sudoers |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Doing this, we mount one configmap over another again, as we mount this to /etc
and then nova-etc
to /etc/nova
. We don't want that. Can we find another solution? Maybe mount the whole /etc
because of sudoers
and prepend all paths for Nova's config files with nova/
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
But then, whole /etc on the container will get overwritten with what's on our projected volume.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The way would be to use subPath for /etc/sudoers.
- configMap: | ||
items: | ||
- key: sudoers | ||
path: sudoers |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
name:
is missing, I think. That way, it doesn't know where to project from.
openstack/nova/templates/hypervisors/hypervisors-vmware-deployment-vct.yaml
Outdated
Show resolved
Hide resolved
openstack/nova/templates/hypervisors/hypervisors-vmware-deployment-vct.yaml
Outdated
Show resolved
Hide resolved
openstack/nova/templates/hypervisors/hypervisors-vmware-deployment-vct.yaml
Outdated
Show resolved
Hide resolved
61de3ae
to
3524821
Compare
3524821
to
a98e017
Compare
I've rebased the code and added some of the fixed requested. I am not 100% sure I covered all of them. |
We still have crashlooping pods when the container exits (e.g. when the process is killed manually or by exception). Fabian Ruff found a bug, that should explain the behavior: kubernetes/kubernetes#68211 (comment) This bug also contains a workaround, that's implemented here: use a projected configMap instead of mounting directly from the original ones. This way, an updated configMap will not be seen by the restarting container and thus will not lead to the crashloop we've seen. CCM-9905
a98e017
to
b5b44aa
Compare
This is part of the ticket CCM-9905