On Sun, Sep 28, 2014 at 10:12:11AM -0700, Cameron Norman wrote: > I would disagree that this particular bug (CanSuspend, CanShutdown, > CanHibernate, etc. being reported as false) is of a grave severity. These > are my reasons: > > 1) I can not reproduce it. This means that the package is not affecting > every (or even a majority) of users, and is not completely useless.
The problem is trivial to reproduce, so please try harder. Just do a standard Jessie xfce install, install sysvinit-core and systemd-shim to let it replace systemd-sysv. Try to suspend. Multiple people has already reported running into this and people are not able to figure out how to work around them. (And some of these users are other Debian Developers I might add, so should supposedly be on the more qualified side of the scale.) Last time I tried xfce was the default desktop. Now that we moved back to GNOME again, the problems are worse. People are reporting not being able to even login because of systemd-shim failure. (I can't help but wonder if anyone is actually testing systemd-shim at all? If so, for which usecase are you testing or are you living in some completely theoretical scenario? A default install would be a useful addition to the testcases if it's not already there.) > 2) There is absolutely no chance of data loss here. If there where I'd use the critical severity. > 3) There is no security issues here (unlike the other bug you raised the > priority of). If there where, I'd use the security tag. > > The only item above that is questionable would be (1) I think. I am not > perfect on what bug severities' qualifications are, so if the "makes the > package in question unusable" bit is satisfied by it being unusable for only > one or a few users, then I retract my disagreement. I agree with this part, but it's probably more correct to guess that you haven't actually read the policy and are just cargo-culting reasons you've seen other people use without understanding them. Please read debian policy if you really want to go on and argue about severities. (But it would be so much more helpful if people would focus on solving issues which makes users not be able to use their computers properly instead of arguing about severities. I only raised the severities because the maintainer wanted to get current issues pointed out better as quoted in my previous mail.) > > Otherwise, if you agree Andreas, please lower this back down to important. No I don't agree. I'm also at the receiving end of people being frustrated about all the random breakage systemd-shim creates. If people really want to see systemd-shim be an alternative a whole bunch of bugs in it needs to be fixed. Help with doing user support is very much welcome because currently the mere existance of systemd-shim raises whe workload alot here. If people argue that systemd-shim is fine in its current state you really need to bring and armada of people doing user support because users reporting in don't agree with you. If this armada of volunteers for user support doesn't exist, then either systemd-shim needs to be fixed or made sure it's not tricking users into using it (ie. removed from jessie). This might sound harsh, but there are a bunch of more issues which noone has bothered reassigning to systemd-shim yet as apparently noone is taking care of the issues anyway. There really is a need for a big effort to bring systemd-shim into a much better state if it's going to fill the shoes it tries to claim. These issues don't go away simply by arguing about bug severities. Feel free to argue them with the maintainer of the package and keep me out of the loop here so I can use my time to actually improve Debian. Regards, Andreas Henriksson -- To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
