On Fri, Mar 25, 2011 at 11:36:09AM +0000, António Meireles wrote: > in the bump (from a fl:2 user point of view when updating) python > changes and conary changes and toolchain changes, at same time,... > i just want to avoid that for the common case the probabilities of ppl > getting stuck in a state where they can't neither rollback nor mv > forward are absolutely minimal. > (i do know that some bugs related to this were identified and fixed in > both sides (conary and us). want us to be absolutely positive that it > is enough)
from the discussion on irc i have glimpsed mention of making a new label fl:2.5, and treating the transition from fl:2 to fl:2.5 just like the one from fl:1 to fl:2. i like this idea because it allows us to keep new users away from fl:2-qa, and what's more, i think it also allows testing the the transition from fl:2 to fl:2.5 in exactly the same manner as an fl:2 user would experience it, whereas an upgrade from fl:2 now to a future fl:2 with a push from fl:2-qa can not be tested without actually pushing first because i believe that a migration may behave differently than an updateall. (but maybe i am wrong here) it is really unfortunate that i am just to busy these days to hang out on irc. it is much easier to follow a discussion on lists when i am busy, because i can pick subjects which are interesting to me, i can read them any time, and reply any time later, and catch up on other stuff later. greetings, martin. -- cooperative communication with sTeam - caudium, pike, roxen and unix services: debugging, programming, training, linux sysadmin, web development -- pike programmer working in china community.gotpike.org foresight developer (open-steam|caudium).org foresightlinux.org unix sysadmin iaeste.at realss.com Martin Bähr http://www.iaeste.at/~mbaehr/ is.schon.org _______________________________________________ Foresight-devel mailing list Foresight-devel@lists.rpath.org http://lists.rpath.org/mailman/listinfo/foresight-devel
