Hi Andreas, On Fri, May 15, 2015 at 11:41:08AM +0200, Andreas Henriksson wrote: > Hello Ron! > > On Thu, May 14, 2015 at 06:50:52AM +0930, Ron wrote: > > On Wed, May 13, 2015 at 12:37:06PM +0200, Karel Zak wrote: > > > On Wed, May 13, 2015 at 08:36:04AM +0200, Andreas Henriksson wrote: > > > > Possibly changing the output format now could be risky. > > > > > > Yes. Ron is right, the current behaviour is strange, but I'm not sure > > > if we want to fix it after 3 years. > > > > Hmm. Being that old complicates things a bit more, yeah. I guess Wheezy > > missed out on getting that update, because in the case of Debian stable > > releases at least, right now it's a regression that's still only about > > a month old :) > [...] > > It was this kind of thinking ...
I think you might be overthinking this :) Or at least reacting to something in your own head and not what I wrote or was thinking ... That was merely a data point for Karel: that I didn't realise it was already this old upstream, and that maybe there could be a brand new wave of fallout still to come for something upstream thought had now long ago been 'resolved'. Nowhere did I suggest, or even think, that an appropriate answer here would be for Debian to fork some alternative behaviour to the upstream code as an ongoing thing. My preferred option is for everywhere to converge on whatever we consider the correct behaviour to be. Anything else would be perpetuating insanity of one brand or another. > I also don't agree that even in Debian viewpoint this is "still only > about a month old" because the updated package has been available in > Debian for almost a year already. People who are concerned with issues > that might arise should ofcourse help out test our distribution *before* > it's deemed "stable". Changing the behaviour after "stable" has already > been released is not something I see as a possibility. Sorry, but there you are Just Wrong. As a DD, I've been running Jessie systems since the day after Wheezy first released, and at no time before I reported it did this problem become evident. In fact I only noticed it this time because I had one machine with the systemd logging verbosity dialled up, and wondered why it was filling my logs with complaints about fstab every time I installed some new package. A stable release doesn't magically stop regressions from being regressions. It doesn't even stop new ones from being introduced (security updates do that all the time and they are promptly fixed when they are noticed too). All a stable release does is vastly increase the number of people who might be exposed to them. For all of those people, this is less than a month old. Insisting they should have been pre- or omniscient in what they tested "if they care" doesn't change that in any meaningful way. No more than it would for me to insist that if the maintainer didn't notice, then they must not have been concerned with issues either ... :) Yes, it would have been great if *somebody* would have noticed it before I did. But they didn't. If we could be perfect about that, stable releases would never have any bugs. But they do. All we can do is decide on the best way to handle them, and in this case it most definitely is upstream's call. I just put out all of the facts and thoughts that I had about it, to help identify the best decision based on the evidence at hand. > Many thanks to Karel Zak for helping out with suggestions on how people > should best handle the situation. I'm hopefull the situation you > described can be handled without too much pain at your end thanks to his > input. Here we are back on the same page again though :) I wasn't aware of the existence of lsblk, or that DEVNAME was magic and not really a tag. This isn't something I need or use very often, that preseed script one-liner being about the only thing I have ever used it for directly myself. I am however always very interested in seeing misbehaviour fixed in the long term, since it's almost impossible to guess how it might burn me next time, and if I wait until then to address that it will be *far* too late, and any fix will still be years away from being able to be used without a workaround. That was my primary interest in the further discussion on this, since we do all agree the current behaviour is odd, and that changing it (again) could be tricky. Being odd forever seems suboptimal, so the interesting question is how or if we can manage the tricky part. I tossed out a few options that have worked for situations like this in the past (before I learned that lsblk was already yet another alternative here) - but Karel clearly understands the problem, and the options and alternatives surrounding it, so My Work Here Was Done. I reported a regression in Jessie that I ran into, and it's now Under Control upstream. Everything worked like it should. (Aside from maybe the little side-track above :) Cheers, Ron -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
