[regression] hald ignores changes to files in /usr/share/hal/fdi and /etc/hal/fdi

Bug #304793 reported by z3non
2
Affects Status Importance Assigned to Milestone
hal (Ubuntu)
Fix Released
High
Martin Pitt

Bug Description

Binary package hint: hal

since one of the last upgrades, the hald isn't configurable anymore.

It ignores changes to or new files in /usr/share/hal/fdi and /etc/hal/fdi

I tried to comment out the
 <!-- merge key="info.ignore" type="bool">true</merge --> in /usr/share/hal/fdi/preprobe/10osvendor/20-broken-usb-sticks.fdi
for my Huawei USB-modem as I want to access the microSD card (which did work fine with earlier versions ..). lshal reports it as ignored even after restarting hal or the whole system. Also creating a file to /etc/hal/fdi/preprobe or adding

<?xml version="1.0" encoding="UTF-8"?>
<deviceinfo version="0.2">
  <device>
     <!-- Ignore the Huawei E220/E180 CD emulation -->
     <match key="@storage.originating_device:usb.vendor_id" int="0x12d1">
       <match key="@storage.originating_device:usb.product_id" int="0x1003">
         <merge key="storage.automount_enabled_hint" type="bool">false</merge>
         <match key="storage.drive_type" string="cdrom">
           <merge key="info.ignore" type="bool">true</merge>
         </match>
       </match>
     </match>
   </match>
  </device>
</deviceinfo>

to /etc/hal/fdi/policy/preferences.fdi doesn't show any effect. So I tried to probe hal in a more general way with the following test file (test.fdi):

<?xml version="1.0" encoding="UTF-8"?> <!-- -*- SGML -*- -->

<deviceinfo version="0.2">
  <device>
    <match key="block.is_volume" bool="true">
      <merge key="volume.ignore" type="bool">true</merge>
    </match>
  </device>
</deviceinfo>

which is ignored in /etc/hal/policy and /usr/share/hal/fdi/policy/20thirdparty/ , usb volumes still show up ...

ProblemType: Bug
Architecture: i386
DistroRelease: Ubuntu 8.10
NonfreeKernelModules: nvidia
Package: hal 0.5.11-4ubuntu4 [modified: usr/bin/lshal usr/bin/hal-get-property usr/bin/hal-set-property usr/bin/hal-find-by-capability usr/bin/hal-find-by-property usr/bin/hal-device usr/bin/hal-is-caller-locked-out usr/bin/hal-lock usr/bin/hal-disable-polling usr/bin/hal-is-caller-privileged usr/bin/hal-setup-keymap usr/sbin/hald usr/sbin/umount.hal usr/lib/hal/hald-probe-input usr/lib/hal/hald-probe-hiddev usr/lib/hal/hald-probe-storage usr/lib/hal/hald-probe-volume usr/lib/hal/hald-probe-printer usr/lib/hal/hald-probe-pc-floppy usr/lib/hal/hald-probe-smbios usr/lib/hal/hald-probe-serial usr/lib/hal/hald-probe-ieee1394-unit usr/lib/hal/hald-probe-net-bluetooth usr/lib/hal/hald-probe-video4linux usr/lib/hal/hald-addon-hid-ups usr/lib/hal/hald-addon-input usr/lib/hal/hald-addon-storage usr/lib/hal/hald-addon-pmu usr/lib/hal/hald-addon-acpi usr/lib/hal/hald-addon-imac-backlight usr/lib/hal/hald-addon-macbookpro-backlight usr/lib/hal/hald-addon-macbook-backlight usr/lib/hal/hald-addon-cpufreq usr/lib/hal/hald-addon-usb-csr usr/lib/hal/hald-addon-dell-backlight usr/lib/hal/hald-generate-fdi-cache usr/lib/hal/hald-runner usr/lib/hal/hal-ipw-killswitch-linux usr/lib/hal/hal-storage-mount usr/lib/hal/hal-storage-unmount usr/lib/hal/hal-storage-eject usr/lib/hal/hal-storage-closetray usr/lib/hal/hal-storage-cleanup-mountpoint usr/lib/hal/hal-storage-cleanup-all-mountpoints usr/lib/hal/hal-system-setserial usr/lib/hal/hal-system-power-pm-is-supported usr/lib/hal/hal-system-power-pmu usr/lib/hal/hal-system-sonypic usr/lib/hal/hal-acl-tool]
ProcEnviron:
 LANGUAGE=de
 PATH=/home/username/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games
 LANG=de_AT.UTF-8
 SHELL=/bin/bash
SourcePackage: hal
Uname: Linux 2.6.27-10-generic i686

Tags: apport-bug
Revision history for this message
z3non (tom-uttenthaler) wrote :
Revision history for this message
z3non (tom-uttenthaler) wrote :

The root of the problem was a future timestamp of the fdi-cache:

-rw-r--r-- 1 root root 1039604 2024-07-20 02:45 fdi-cache

after deleting it, everything works fine again. The future time probably came from a PM_TRACE, which I have activated, since the S2Mem is f*ing instable since intrepid ... nevertheless, I regard this behavior as a bug in hal. It is or has been discussed there for Debian and Ubuntu, so there should:

http://<email address hidden>/msg572726.html
http://groups.google.com/group/linux.debian.bugs.dist/browse_thread/thread/83d30ea7b629e4a4

and here you find a patch (which is secure, as it deletes the fdi-cache on every hal-start, but increases boot-time):

http://groups.google.com/group/linux.debian.bugs.dist/attach/b524a9234fe6547d/hal.fdi-cache.debdiff?part=3

regards,
tom

z3non (tom-uttenthaler)
Changed in hal:
assignee: nobody → pitti
Martin Pitt (pitti)
Changed in hal:
importance: Undecided → High
status: New → In Progress
Revision history for this message
Martin Pitt (pitti) wrote :

This was fixed in intrepid in version 0.5.11-4ubuntu1.

Changed in hal:
status: In Progress → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.