Hi, On Tue, Aug 30, 2011 at 12:08:44PM +0100, Gerhardus Geldenhuis wrote: > I looked around through the documentation but could not get a description of > how to do "repo" snapshots. I might have missed it or not have a full grasp > or understanding yet but would appreciate anyone sharing thoughts on how to > do it. > > Very simply I want to be able using only cobbler to create a snapshot of a > repository and called it for example RH5.6-Feb which should contain the base > repo plus all updates that has been releases up until that point. I want to > be able to then create monthly snapshots for example and when I build a > server build it against one of these snapshots. If you can already do this > in cobbler I would appreciate some guidance or if not if anyone could share > how they achieve this with cobbler.
I think this is not implemented, you should be able to do it yourself for repos that you host (have locally as files): - 'cp -r /part/repo /part/reposnapshot' - 'cobbler repo copy ...' to make the snapshot known to cobbler > You can do this in spacewalk but I am > not sure how much is spacewalk and how much is cobbler and I am keen to not > re-invent the wheel. For various reasons at the moment unfortunately I can't > use spacewalk to manage this so the solution has to be native to cobbler. Cobbler is maintaining the repo just as a whole, spacewalk knows about the packages inside. That way a clone of a channel in spacewalk (what you call snapshot) happens mainly in the database, without creating copies of files. I think you could be fine with the cobbler approach. This one needs more diskspace, yet that could be reduced in creating hardlinks for files that exist in multiple repos. Christian _______________________________________________ cobbler mailing list [email protected] https://fedorahosted.org/mailman/listinfo/cobbler
