I was asked a question that got me thinking: if crane is just a static file
server could we use it without pulp? For docker v1 we use pulp to generate
the crane json files. I'm not sure that's required for v2. Note: I am
really behind on docker v2 in general and the crane v2 implementation.

Why: Managing a pulp instance is overhead that is not always required. For
the v1 registry service[1] we're doing basically what is being asked: take
a tarball export from pulp, push crane json files, the layers go into an
AWS S3 bucket.

[1] https://github.com/aweiteka/raas
_______________________________________________
Pulp-list mailing list
[email protected]
https://www.redhat.com/mailman/listinfo/pulp-list

Reply via email to