On Thursday 28 of July 2016 16:34:30 Greg Sutcliffe wrote: > So I had a read back through the thread, and drew a few conclusions. The > history seems to be: > > * The thread was opened ~10 weeks ago, and went quiet ~6 weeks ago > * At that time, there were 5 upvotes (Dominic, Daniel, Ewoud, Michael, > Duncan), 1 downvote (BK), and 1 unsure (Ohad, who said he had "mixed > feelings). Others commented but did not express a preference. > * We then jump forward to a few days ago, with all the discussion that's > happened since. > > It's hard to deny that it looked OK to go ahead with, as of 3 days ago; > 5-to-1 is something we wouldn't normally challenge, and 6 weeks is plenty > of time to speak up about concerns. > > However, I do take issue with comparing it to the previous OS deprecations, > which by-and-large have been Debian/Ubuntu deprecations. These (a) don't > impact as large a part of the community, (b) have a shorter shelf-life than > EL (implying users are more comfortable with upgrading), and (c) have a > simpler packaging structure. They're also unlikely to be using any of the > RPM-centric plugins we offer, which (of course) will be more affected by > dropping EL6 (bear in mind that according to the 2016 survey, some 30% of > our community use Katello - that's not insignificant). As such, while I > agree no fallout occurred in the past, I think we do have to be more > careful this time. > > More than anything, I think it's the sudden switch from "dead thread" to > "it's done" that has raised tempers, rather than the decision itself. I > think it would have been reasonable to follow up with a "this seems > decided, I'll do it in a few days" type of email - similar to how the > branching happens. We have a similar "final call" approach in the RFCs, in > case anyone lost sight of the thread in the noise. > > In short then, we are where we are. I think these concerns *should* have > been raised earlier, but now that they *have* been raised, I think they're > valid, and we should finish the discussion. As it stands today, we have > around 5-to-4 in favour of dropping EL6 in 1.13, but that's not any kind of > consensus.
Count me in for EL6 support for at least another release. If there's some extra effort needed with which I can help somehow, I'm happy to do so. I thought we solved main problems on smart-proxy using puppet API and SCL for dynflow plugin. -- Marek > > Can we get a (re)statement of the pros/cons from each side, to inform the > choices of those on the fence? As of 10th May it seemed that the support of > EL6 was only causing minor issues - is that still true? Other than > preparation time (which is valid) are there other concerns for dropping it? > > (PS - More generally, we seem to have an issue with "closing" > discussions/RFCs with a firm decision; many discussions die out in this > way. Thoughts on how we can improve that are welcome, and I will post my > own, but lets start a new thread for that) -- You received this message because you are subscribed to the Google Groups "foreman-dev" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
