On Thu, Oct 11, 2012 at 5:58 PM, Zeeshan Ali (Khattak) <zeeshanak@xxxxxxxxx> wrote: > On Thu, Oct 11, 2012 at 5:13 PM, David Zeuthen <davidz@xxxxxxxxxx> wrote: >> >> >>> > And can you confirm that David's updated rule fixes this issue? >> >> As I said in my other mail, there are many many reasons why things >> can fail that are unrelated to the udev rule. Just proclaiming "it >> doesn't work!" without investigating further and then scrapping a >> good udev rule is non-sense. > > I didn't scrap anything. I'm saying more like I am unable to do any > testing or comparison if things are broken for me with and without > your change. I'm telling it here so that you or anyone else could help > me figure out what is going on. > >> For example, there are some issues with the kernel loop driver and >> partitioned devices, that's probably what Zeeshan is running into >> (support for partitioned loop devices is rather new). I don't know >> what kernel it's fixed in (if fixed at all). But it's easy to check >> if the expected block devices show up. Maybe Zeeshan didn't do that, >> I don't know. > > What I mean by "doesn't work" is that the media in question gets > automatically mounted and I get the dialog "Open with Files". Somehow > the udev rule never get triggered AFAICT: > > $ osinfo-detect -f env /dev/sdb1 > OSINFO_BOOTABLE=1 > OSINFO_INSTALLER=http://fedoraproject.org/fedora/18 > OSINFO_LIVE=http://fedoraproject.org/fedora/18 > OSINFO_MEDIA=http://fedoraproject.org/fedora/18:3 > $ udevadm info --query=all /dev/sdb1|grep OSI > $ > > $ osinfo-detect -f env /dev/loop5p1 > OSINFO_BOOTABLE=1 > OSINFO_INSTALLER=http://fedoraproject.org/fedora/17 > OSINFO_MEDIA=http://fedoraproject.org/fedora/17:1 > $ udevadm info --query=all /dev/loop5p1|grep OSI > >> FWIW, the updated udev rule consistently works on my box; latest F18 >> packages, stock no customizations. That is, if the block devices >> do show up. > > Great! If you have also tested against USB stick and hw CDROM, thats > all the testing we need. I tested with hw CDROM and it worked like a charm and David tested with USB stick and loopback devices. > In any case, I have no objection on pushing this change already as I > haven't seen it to cause any regressions. I have pushed this. Now I'll try to roll out a release.. Thanks David for the patch and Christophe for sending the patch here and testing. -- Regards, Zeeshan Ali (Khattak) FSF member#5124
