On Sat, Mar 17, 2018 at 10:52 AM, Martin Horák <[email protected]> wrote:
> Hi Michael. > Thank you for the answer. You just copied my own thoughts. CephFS is > certainly an option, I asked our Ceph admins for it, so If they provide > some for me, I'll use it. As a backup scenario I thought about exporting > RBD (Ceph block device) via NFS, but i'd rather avoid that - it's > unnecessarily complicated, probably slow, error prone. > Anyway I think I could still go with unshared storage. If I had all the > containers using vlp volume in one pod. Of course I couldn't scale up > workers then. It wouldn't be optimal solution, but maybe possible. What do > you think? > The only problem I can think of is that when you upload content, an httpd process receives the bits and needs to write them into /var/lib/pulp. If you could avoid using the upload feature entirely, you'd be ok. Otherwise you'll either need to run httpd also in the same pod, or find another storage option. -- Michael Hrivnak Principal Software Engineer, RHCE Red Hat
_______________________________________________ Pulp-list mailing list [email protected] https://www.redhat.com/mailman/listinfo/pulp-list
