-
Notifications
You must be signed in to change notification settings - Fork 2.4k
[grafana] Add DAC_READ_SEARCH
to initChownData
and readOnlyRootFilesystem
#3698
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
base: main
Are you sure you want to change the base?
Conversation
@andreas-heatly are you able to test this to prevent an outage on your side? |
I tested the restart usecase explained in #3699 and this fixes that at least
|
…stem Signed-off-by: Pat Riehecky <[email protected]>
Tested this - does not work with psp:
Don't think we have to give away more than the baseline policy?! |
Generally, chown is a privileged operation. If you want to be compliant with baseline, just disable the initContainer. In modern Kubernetes environment, its generally not recommend to use that approach. |
Breaking compat with There is some sort of strange interaction between the init-container and the pvc. I don't think I'll have time to look at this for a while... I'll close this PR out. Interested folks at least have a hint something is tricky with the explicit |
Sorry to interrupt here, but the chown init container will never be compatible with baseline. Changing file owner is an operation done by the root account. This is normally done by But it's still valuable to increase the hardening in that context. |
But I don't understand - shouldn't the default be the hardened? I am only here, because of the issues. Never decided to activate the init container. Never had any psp issues until the cap of this pr was added. And chown alone doesn't violate the psp. |
But how the PSP can affect your deployment, if chmod container is not enabled on your deployment? |
The default is enable chown |
Thats the root issue. this should be not enabled by default. |
This should resolve the regression from #3684