Re: Bodhi policy for pushing updates to stable

2015-01-26 Thread Adam Williamson
On Sat, 2015-01-24 at 11:21 -0800, Adam Williamson wrote: > On Sat, 2015-01-24 at 11:05 -0800, Adam Williamson wrote: > > > > With a change along those lines, I think we could plausibly look > > at hard enforcement of the upgrade path, and it would be a good > > improvement. It may be necessary

Re: Bodhi policy for pushing updates to stable

2015-01-25 Thread Kevin Fenzi
On Mon, 26 Jan 2015 03:02:32 +0100 Kevin Kofler wrote: > The solution for that is to finally allow multi-release update > groups, where the karma would be added together for all releases. If > we push the exact same security fix to 3 releases, it should not need > separate testing for each. Whil

Re: Bodhi policy for pushing updates to stable

2015-01-25 Thread Kevin Kofler
Adam Williamson wrote: > So thinking this through again...the bit I forgot to mention - the > reason why 'updates' matters - is that on fedup to Branched, updates- > testing is (usually) not used, because fedup takes its repo set from > the release being upgraded from. I don't know if it'd be pract

Re: Bodhi policy for pushing updates to stable

2015-01-25 Thread Kevin Kofler
Adam Williamson wrote: > With a change along those lines, I think we could plausibly look at > hard enforcement of the upgrade path, and it would be a good > improvement. It may be necessary to have *some* kind of override > mechanism for the case where we have a major security issue we really > ne

Re: Bodhi policy for pushing updates to stable

2015-01-24 Thread Bruno Wolff III
On Sat, Jan 24, 2015 at 11:05:52 -0800, Adam Williamson wrote: Someone proposed that using the 'updates' repository during the Branched period could help solve quite a few things here, and I think that's an interesting idea. The other thing it fixes is packages dgowing up as orphaned when th

Re: Bodhi policy for pushing updates to stable

2015-01-24 Thread Adam Williamson
On Sat, 2015-01-24 at 16:11 -0600, Dennis Gilmore wrote: > > With a change along those lines, I think we could plausibly look > > at hard enforcement of the upgrade path, and it would be a good > > improvement. It may be necessary to have *some* kind of override > > mechanism for the case where

Re: Bodhi policy for pushing updates to stable

2015-01-24 Thread Dennis Gilmore
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Sat, 24 Jan 2015 11:05:52 -0800 Adam Williamson wrote: > On Sat, 2015-01-24 at 17:42 +0100, Kevin Kofler wrote: > > Kamil Paral wrote: > > > So, enforcing upgrade path for stable releases sounds good. But > > > when we add development releases in

Re: Bodhi policy for pushing updates to stable

2015-01-24 Thread Adam Williamson
On Sat, 2015-01-24 at 11:05 -0800, Adam Williamson wrote: > > With a change along those lines, I think we could plausibly look at > hard enforcement of the upgrade path, and it would be a good > improvement. It may be necessary to have *some* kind of override > mechanism for the case where we h

Re: Bodhi policy for pushing updates to stable

2015-01-24 Thread Adam Williamson
On Sat, 2015-01-24 at 17:42 +0100, Kevin Kofler wrote: > Kamil Paral wrote: > > So, enforcing upgrade path for stable releases sounds good. But > > when we add development releases into the mix, we need to break > > upgrade path in certain cases. And we probably need to come up > > with a differ

Re: Bodhi policy for pushing updates to stable

2015-01-24 Thread Kevin Kofler
Kamil Paral wrote: > So, enforcing upgrade path for stable releases sounds good. But when we > add development releases into the mix, we need to break upgrade path in > certain cases. And we probably need to come up with a different solution > to ensure you can correctly upgrade to it on the releas

Re: Bodhi policy for pushing updates to stable

2015-01-21 Thread Kamil Paral
> Hi all. > > When upgrading F20 to F21 using FedUp, some users had a problem > with some packages not being upgraded (e.g. [1]). The problem was > caused by broken update path F20 -> F21. Just a note: That problem is even more complex. Fedora Branched some has freeze periods, when you can't pus

Re: Bodhi policy for pushing updates to stable

2015-01-16 Thread Tomas Hozza
On 01/15/2015 05:15 PM, Luke Macken wrote: > On Thu, Jan 15, 2015 at 10:19:19AM +0100, Tomas Hozza wrote: > > Hi all. > > > > When upgrading F20 to F21 using FedUp, some users had a problem > > with some packages not being upgraded (e.g. [1]). The problem was > > caused by broken update path F20 ->

Re: Bodhi policy for pushing updates to stable

2015-01-15 Thread Luke Macken
On Thu, Jan 15, 2015 at 10:19:19AM +0100, Tomas Hozza wrote: > Hi all. > > When upgrading F20 to F21 using FedUp, some users had a problem > with some packages not being upgraded (e.g. [1]). The problem was > caused by broken update path F20 -> F21. > > For example in wget's case I pushed updates

Re: Bodhi policy for pushing updates to stable

2015-01-15 Thread Matěj Cepl
On 2015-01-15, 09:19 GMT, Tomas Hozza wrote: > I think bodhi should enforce the update path is not broken and > hold the update for F20 until the update for F21 is in stable. Gosh, I thought bodhi already enforces update policy ... :( Matěj -- devel mailing list devel@lists.fedoraproject.org ht

Bodhi policy for pushing updates to stable

2015-01-15 Thread Tomas Hozza
Hi all. When upgrading F20 to F21 using FedUp, some users had a problem with some packages not being upgraded (e.g. [1]). The problem was caused by broken update path F20 -> F21. For example in wget's case I pushed updates for the same NVR in F20 and F21 with auto-karma. However the wget update f