On Tue, 2006-09-12 at 11:26 -0400, Jesse Keating wrote: > On Tuesday 12 September 2006 11:08, buildsys@xxxxxxxxxxxxxxxxx wrote: > > quagga > > 4: 0:0.98.6-1.fc4 (FC4-updates) > > 5: 0:0.98.6-1.FC5 (FC5-updates) > > 6: 0:0.98.6-2.1 (FC6) > > The only problem here I see is going from 4 to 5, but 4 to 6 and 5 to 6 should > work. [...] > Again, only a problem going from 4 to 5, but 4 to 6 and 5 to 6 is fine. [...] > Same story here. Yes, that's how the upgrade checker script currently works: it finds the first problematic upgrade path and reports everything onwards from it without checking later paths. Improving this is on my TODO list, but I'm not sure if it's that useful to also check upgrade paths that have "holes" in them such as going directly from FC4 to FC6 -- I think it'd just add noise to the reports making it harder to spot the actual problem. By the way, I don't have the impression that packagers are expected to make sure that upgrades like directly from FC(x) to FC(x+2) without stopping at FC(x+1) do work in the first place. Other opinions? -- fedora-extras-list mailing list fedora-extras-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-extras-list
