-
Notifications
You must be signed in to change notification settings - Fork 62
Deployment failing in generic hosts with permission issues in deploy image #2375
Comments
I tried to build image locally using
notice the last line in the error that complaints about nginx.conf, from where that file is injected? |
@pranavgore09 Did you set the |
@jarifibrahim yes I added that, still not looking good. @debloper I could figure out permission error
run.sh was crying for that direcotry.
that is entrypoint for |
@jarifibrahim this issue isn't regarding the @pranavgore09 I really hate how we practice security in here. |
@debloper maybe, I do not really understand security in this case, no opinion as of now 😄 |
@debloper @pranavgore09 The issue exists because Now, the @debloper About the following error - it isn't actually fatal. I agree that it should be fixed, but the image works fine even with this error.
The PR fabric8-services/fabric8-wit#1817 uses the same [0] - https://github.com/fabric8-ui/fabric8-ui/blob/master/Dockerfile.deploy |
In docker-compose, minishift or regular docker-run, the deploy image fails with:
Reference point for this issue, from the original base-image:
https://github.com/kbsingh/openshift-nginx/blob/master/root/run.sh#L6
The particular command itself isn't fatal, but it bails out with a fail and also gives the idea that the following commands will also fail, especially the one at https://github.com/kbsingh/openshift-nginx/blob/master/root/run.sh#L9 which is important.
Should we use root user, or escalate privileges of the current user, is probably the question here.
We should also investigate, why and how this works in the CI/CD pipeline.
The text was updated successfully, but these errors were encountered: