Red Hat Bugzilla – Bug 200047
USB mass storage devices no longer automatically mount
Last modified: 2013-03-05 22:46:07 EST
Description of problem:
If I attach my audio player (think USB thumb drive) or external HD, both USB
highspeed devices, they do not get mounted to /media/YEPP1 or /media/USBbackup
(respectively), nor anywhere else.
Version-Release number of selected component (if applicable):
Steps to Reproduce:
1. Connect the USB device.
2. Wait 5 to 10 second
3. Do "tail -n 25 /var/log/messages" to verify that udev created the device files.
The device files are created (/dev/sdb and /dev/sdb1), but sdb1 isn't mounted
and the mount point doesn't appear on the desktop.
The VFat file system on /dev/sdb1 should be mount as /media/$FSLABEL
(/media/YEPP1 or /media/USBBackup); and a mount point icon on the desktop appears.
A lot of packages have changed, but mostly for BuildRequires reasons. This
might reach into HAL and DBUS.
I'm seeing the same thing on a Centrino Core Duo notebook since FC6T1 too
(wasn't running FC on it prior to T1). USB Keys, USB HDDs, iPods etc are all
problematic but CDs automount just fine.
rpm -qa|grep -E 'gnome-mount|udev|kernel|gnome-volume-manager|hal|dbus'|sort
Fixed on devel and FC5
Fixed is what version?
I running latest devel (rawhide) and it still doesn't work.
I speek english good. :-)
Fixed in what version?
I'm running the latest devel (rawhide) and it still doesn't work.
Ah! There's now an entry in /etc/udev/rules.d/ that refers to media. There
were no entries here before.
I'm getting a "Too long line" error from 60-libsane.rules in the directory.
Would this block the 90-hal.rules from loading?
hal-0.5.7.1-1.fc6. You need to reboot after the update.
I've already rebooted and I'm using hal-0.5.7.1-2.fc6 installed.
I'm going to try rebooting with sane-backends removed to avoid the error in #5
(Too long line) and see if it behaves better.
Removing sane-backends did remove the error message but did not help the
situation. My mass storage device still isn't being mounted.
Earlier this morning my device wouldn't automount. Then, after doing a yum
update, it now works. I'm not sure what updated package resolved the problem.
The update was:
kernel i686 2.6.17-1.2505.fc6 development 16 M
kernel-devel i686 2.6.17-1.2505.fc6 development 4.4 M
ekiga i386 2.0.2-3 development 5.1 M
gnupg i386 1.4.5-3 development 1.8 M
hal-cups-utils i386 0.6.2-3 development 26 k
initscripts i386 8.38-1 development 1.2 M
irda-utils i386 0.9.17-2.fc6 development 73 k
java-1.4.2-gcj-compat i386 220.127.116.11-40jpp_100rh development 43 k
kernel-doc noarch 2.6.17-1.2505.fc6 development 2.7 M
libselinux i386 1.30.19-5 development 89 k
libselinux-devel i386 1.30.19-5 development 126 k
libselinux-python i386 1.30.19-5 development 50 k
libtiff i386 3.8.2-5 development 312 k
libtiff-devel i386 3.8.2-5 development 492 k
lvm2 i386 2.02.06-3 development 1.0 M
module-init-tools i386 3.3-0.pre1.4.15 development 397 k
perl-IO-Socket-SSL noarch 0.994-1.fc6 development 47 k
rhpxl i386 0.27-1 development 93 k
selinux-policy noarch 2.3.3-17 development 290 k
selinux-policy-targeted noarch 2.3.3-17 development 638 k
xorg-x11-drv-ati i386 6.6.1-6.fc6 development 333 k
kernel i686 2.6.17-1.2439.fc6 installed 44 M
kernel-devel i686 2.6.17-1.2439.fc6 installed 14 M
I have not rebooted in the past 5 days.
Most likely a library that was updated finally got loaded. If you don't reboot
it is unknown when a new library will replace the older one in memory.
I rebooted and it no longer works.
Created attachment 133697 [details]
Output info from udev and hald when mass-storage device is connected
I did a "udevcontrol log_priority=info" and added the arguments "--daemon=yes
--verbose=yes --use-syslog" to hald in /etc/rc.d/init.d/haldaemon.
The attachment is the results found in /var/log/messages.
The results are hard for me to intrepret (and I'm likely wrong), but I think
the problem might be something higher on the food chain.
Reverting back to gnome-volume-manager-1.5.15-4 allowed auto mounting to work.
Reverting gnome-session-2.15.4-2 allowed gnome-volume-manager to be started at
the beginning of my gnome session.
It's really a problem to me. I have to reboot after every suspend since USB
stick is not mounted anymore.
PM: Adding info for No Bus:target2:0:1
PM: Removing info for No Bus:target2:0:1
PM: Adding info for No Bus:target2:0:2
PM: Removing info for No Bus:target2:0:2
PM: Adding info for No Bus:target2:0:3
PM: Removing info for No Bus:target2:0:3
PM: Adding info for No Bus:target2:0:4
PM: Removing info for No Bus:target2:0:4
PM: Adding info for No Bus:target2:0:5
PM: Removing info for No Bus:target2:0:5
PM: Adding info for No Bus:target2:0:6
PM: Removing info for No Bus:target2:0:6
PM: Adding info for No Bus:target2:0:7
PM: Removing info for No Bus:target2:0:7
Created attachment 135458 [details]
Another output from udev and hald
Created attachment 135459 [details]
First time device was mounted successfully, second time mount failed.
I think this info is relevant
[I] osspec.c:232: SEQNUM=1496, ACTION=add, SUBSYSTEM=block, DEVPATH=/sys/block/
[I] hotplug.c:181: /sys/block/sda/sda1 is a block device (subsystem)
[I] blockdev.c:637: block_add: sysfs_path=/sys/block/sda/sda1 dev=/dev/sda1 is_
[I] blockdev.c:547: get_luks_uuid: device_file=/dev/sda1
[I] blockdev.c:673: Ignoring hotplug event - no parent
[W] blockdev.c:1043: Not adding device object
More investigation: Parent device isn't created because it's scsi.type is
"unknown". That's all because it seems that /sys/block/sda/device/type is
created with delay.
Sep 3 21:41:53 gnome udevd-event: run_program: '/lib/udev/udev_run_hotplugd'
Sep 3 21:41:53 gnome hald: 21:41:53.308 [E] util.c:463: Cannot open
Sep 3 21:41:53 gnome udevd-event: run_program:
'/lib/udev/udev_run_hotplugd' returned with status 0
Sep 3 21:41:53 gnome hald: 21:41:53.314 [E] util.c:463: Cannot open
Sep 3 21:41:53 gnome udevd-event: run_program: '/lib/udev/udev_run_devd'
Sep 3 21:41:53 gnome hald: 21:41:53.317 [E] util.c:322: Cannot open
Sep 3 21:41:53 gnome hald: 21:41:53.320 [W] physdev.c:686: scsi.type is
Actually it's udev bug 201422 and it should be resolved in latest rawhide please
mark as a duplicate
Created attachment 136051 [details]
Sorry never used this before :)
Above patch was made by Martin Pitti from ubuntu edgy patch lists. I've been
having problems with all usb attached devices since FC5 upgrade to hal-0.5.7.1
rebuilt rpm with patch above and my drives show up again. Hope this helps
Should be fixed with hal 0.5.8.1 in latest rawhide otherwise please reopen.