Problem with Kernel 667?
Brian Millett
bpm at ec-group.com
Wed Nov 3 20:53:27 UTC 2004
> I have been running FC3 for a few days - no problems with kernels 639,
> 640, 643 etc. I just installed 667 and rebooted, and when it tried to
> start KsCD, I was told that /dev/cdrom had the wrong permissions or no
> audio CD was in the drive. I was listening to a CD before I rebooted, so
> I know it was working. But after the reboot, /dev/cdrom does not exist
> at all. Any idea why? What is the 'correct' way to restore that?
>
> Is this a known issue? Do I need to create a bugzilla entry?
Looks like udev-039-6.FC3.1
Rolled back to udev-039-6 and got back the dvd, cdrom & cdrecord
Not sure quite what the diff is between the two, but it is:
[bpm]$ diff -wruN /etc/udev /tmp/udev-039-6.FC3.1
diff -wruN /etc/udev/scripts/pam_console.dev
/tmp/udev-039-6.FC3.1/scripts/pam_console.dev
--- /etc/udev/scripts/pam_console.dev 2004-10-21 10:21:03.000000000 -0500
+++ /tmp/udev-039-6.FC3.1/scripts/pam_console.dev 2004-11-02
10:29:51.000000000 -0600
@@ -2,6 +2,9 @@
[ "$ACTION" != "add" ] && exit 0
+# we do not have console users in rc.sysinit
+[ -n "$IN_INITLOG" ] && exit 0
+
if [ -x /sbin/pam_console_setowner -a -f /var/run/console/console.lock \
-a -e "$DEVNAME" ]; then
@@ -35,7 +38,8 @@
fi
SYMLINKS=""
- for i in $(/usr/bin/udevinfo -q symlink -n "$DEVNAME"); do
+ for i in $(/usr/bin/udevinfo -q symlink -p "$DEVPATH"); do
+ [ $? -gt 0 ] && break
SYMLINKS="$SYMLINKS ${udev_root%%/}/$i"
done
debug_mesg "Restoring console permissions for $DEVNAME $SYMLINKS"
diff -wruN /etc/udev/udev.conf /tmp/udev-039-6.FC3.1/udev.conf
--- /etc/udev/udev.conf 2004-10-21 10:21:03.000000000 -0500
+++ /tmp/udev-039-6.FC3.1/udev.conf 2004-11-02 10:29:51.000000000 -0600
@@ -32,4 +32,4 @@
default_group="root"
# udev_log - set to "yes" if you want logging, else "no"
-udev_log="yes"
+udev_log="no"
--
Brian Millett
Enterprise Consulting Group "Shifts in paradigms
(314) 205-9030 often cause nose bleeds."
bpmATec-groupDOTcom Greg Glenn
More information about the fedora-test-list
mailing list