I have an issue of getting stale repo metadata from repoquery. We use this as part of our release process, and as an ad-hoc tool to query what's in a repo.
When you specify --repofrompath on the command line, yum dutifully caches the repodata from that repo. The problem lies in the fact that for one invocation, I might say "myrepo,http://foo.bar/some/repo";. Then I run it again later, and say "myrepo,http://foo.bar/some/other/repo";. As far as yum is concerned, since the repoid is the same, it believes that it has valid cache of the second repo, where in reality it could have wildly different content. If the second repo is older than the first (causing yum to completely ignore metadata checksums in the "new" repomd.xml), you get stale content that doesn't reflect what's going on in the second repo. I'm not sure what the correct way to address this is. We could not cache anything from --repofrompath repos (but how do you tell?). We could provide a command line option to repoquery to completely ignore any cache (which is the right thing in my mind). Am I missing something here? _______________________________________________ Yum-devel mailing list [email protected] http://lists.baseurl.org/mailman/listinfo/yum-devel
