Please read the http://frugalware.org/docs/bugs page if you are new to bugreporting!
FS#3290 - Brasero does not work anymore after udev upgrade
Attached to Project:
Frugalware
Opened by Gerhard Wallraf (lackS) - Thursday, 07 August 2008, 21:05 GMT+1
Last edited by crazy (bugs) - Friday, 29 August 2008, 09:05 GMT+1
Opened by Gerhard Wallraf (lackS) - Thursday, 07 August 2008, 21:05 GMT+1
Last edited by crazy (bugs) - Friday, 29 August 2008, 09:05 GMT+1
|
DetailsAfter doing the bigger upgrade yesterday (August 6, 2008), brasero does not work anymore. cdrecord doesn't either, a group and permission change for /dev/sg* (to group "cdrom" and permission "660") devices makes it work again. I don't know if this is the best and/or cleanest solution, but at least it makes the system work again. Attached, you'll find my modified udev rules file.
|
This task depends upon
50-udev-default.rules
Also your change is wrong since you change the group for any sg* devices ( scsi etc etc ) to be owened by cdrom :)
The rule is needed for cdrom is right :
...
SUBSYSTEMS=="scsi", KERNEL=="s[grt][0-9]*", ATTRS{type}=="5", NAME="%k", GROUP="cdrom", MODE="0660"
...
Also it does work right on all my boxes ( which all uses current )
$ ls -la /dev/sg*
crw------- 1 root root 21, 0 7. Aug 16:00 /dev/sg0
crw-rw---- 1 root cdrom 21, 1 7. Aug 16:00 /dev/sg1
crw------- 1 root root 21, 2 7. Aug 16:00 /dev/sg2
crw------- 1 root root 21, 3 7. Aug 16:00 /dev/sg3
However you did upgraded from what to current ?
Thanks for your help so far,
lackS
the recent udev upgrade broke automounting of cds in kde as well, so i'm not yet closing this bug.
crazy, does system:/media ok at you? (in konqueror)
we reverted some random upstream patch, then run 'udevadm trigger', saying it's the same as a reboot. this is wrong.
if you put udevadm trigger to rc.local, then it works fine, without the reverted patch as well.
the question is what is the real fix here, then..
Note 2: The 50-udev-default.rules file is not owned by any package on -current nowadays, it used to belong to udev-124. Is this the desired behaviour? I tried to remove it, and nothing seems to have changed.
Note 3: I opened a similar bug report at #3320 because this one was closed in between and the desktop show up part of the problem was not (yet) fixed. Maybe we can close one of the two bugs to avoid people working parallely on the same problem.
Note 4: It's just a mood, but to me it seems that hal is simply not recognizing the optical drive contents until explicitly asked. However, USB sticks work fine.
( BTW that bug has nothing to do with hal , is pure udev breakage )
still not closing, but then this is not a problem for 0.9.