Summary: | Hwbrowser dies. | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Product: | [Retired] Red Hat Linux | Reporter: | Ali-Reza Anghaie <ali> | ||||||||
Component: | hwbrowser | Assignee: | Brent Fox <bfox> | ||||||||
Status: | CLOSED RAWHIDE | QA Contact: | |||||||||
Severity: | medium | Docs Contact: | |||||||||
Priority: | medium | ||||||||||
Version: | 8.0 | CC: | baulv, bearl, brian, dimiduk.1, dwbrewer, epl, hops, jaco_botha, john.mizell, laroche, mitr, mpeters, nicholsg, peter_s_sportino, ronan_murphy, spacewar, uchima | ||||||||
Target Milestone: | --- | ||||||||||
Target Release: | --- | ||||||||||
Hardware: | i386 | ||||||||||
OS: | Linux | ||||||||||
Whiteboard: | |||||||||||
Fixed In Version: | Doc Type: | Bug Fix | |||||||||
Doc Text: | Story Points: | --- | |||||||||
Clone Of: | Environment: | ||||||||||
Last Closed: | 2003-05-25 14:31:01 UTC | Type: | --- | ||||||||
Regression: | --- | Mount Type: | --- | ||||||||
Documentation: | --- | CRM: | |||||||||
Verified Versions: | Category: | --- | |||||||||
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||||
Cloudforms Team: | --- | Target Upstream Version: | |||||||||
Attachments: |
|
Description
Ali-Reza Anghaie
2002-10-03 03:28:04 UTC
I am experiencing exactly the same bug on a Dell i8100 laptop with Red Hat 8.0. It does NOT occur on other PC's on which I have also installed Red Hat 8.0. There is no difference in the installed packages or setups that might account for this crashing on the i8100 but not on the other machines. Getting something similar. Starting from the menu it pops up and dies (each time the same). Starting it from a terminal it gives me (each time the same): (A lot of these lines first, 15 in total) =================start of output (DeviceList.py:13911): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: line 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed (DeviceList.py:13911): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: line 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed Traceback (most recent call last): File "/usr/share/hwbrowser/DeviceList.py", line 203, in ? reread_device_type_list () File "/usr/share/hwbrowser/DeviceList.py", line 150, in reread_device_type_list list.set (iter, 0, hardware.get_category_string (key), 1, key) File "/usr/share/hwbrowser/DeviceList.py", line 91, in get_category_string return self.type_mapping[key] KeyError: 20 =================end of output Installed version hwbrowser-0.6-1 libgnomecanvas-2.0.2-1 Reproduces the same results every time. Same as above: 1. Run 'Hardware Browser' from menu or 'hwbrowser' from prompt. 2. Type in passwd if needed. 3. Watch it go *bang* .. Results are that it dies on me. Additional information used LVM (for partitions / /usr /var /home /tmp) and separate /boot SB Audigy is detected but does not give sound Me too on a HP Pavilion 9995. Created attachment 78944 [details]
List of RPMs installed on HP Pav 9996
Ditto on this RH 8.0 P4 1.7 box Have the same problem on RH 8.0, hwbrowser dies with the following errors: $ hwbrowser (DeviceList.py:6672): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: lin e 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed (DeviceList.py:6672): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: lin e 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed (DeviceList.py:6672): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: lin e 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed (DeviceList.py:6672): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: lin e 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed (DeviceList.py:6672): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: lin e 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed (DeviceList.py:6672): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: lin e 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed (DeviceList.py:6672): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: lin e 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed Traceback (most recent call last): File "/usr/share/hwbrowser/DeviceList.py", line 204, in ? reread_device_type_list () File "/usr/share/hwbrowser/DeviceList.py", line 151, in reread_device_type_lis t list.set (iter, 0, hardware.get_category_string (key), 1, key) File "/usr/share/hwbrowser/DeviceList.py", line 92, in get_category_string return self.type_mapping[key] KeyError: 10 Tried a fix which worked, as suggested by mitr. Added the following to line 67 of /usr/share/hwbrowser/DeviceList.py : kudzu.CLASS_FIREWIRE : _("Firewire"), hwbrowser gives the following output when run but does not die: $ hwbrowser (DeviceList.py:11130): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: line 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed (DeviceList.py:11130): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: line 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed (DeviceList.py:11130): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: line 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed (DeviceList.py:11130): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: line 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed (DeviceList.py:11130): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: line 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed (DeviceList.py:11130): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: line 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed (DeviceList.py:11130): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: line 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed I have a Creative Audigy soundcard, the firewire port shows up in hwbrowser. I too have this same problem on Dell Inspiron 8000 I also get the exact same error (with same error text) Dell Latitude C800 Did not experience this error on my Dell Dimension XPS-R400 Tried the fix described above and that works for me as well. Sames output as described, but the hwbrowser does not die on me. Tried a fix which worked, as suggested by mitr. Added the following to line 67 of /usr/share/hwbrowser/DeviceList.py : kudzu.CLASS_FIREWIRE : _("Firewire"), The fix worked for me. Now if redhat can roll it out with rhn. Adding the Firewire changed nothing for me: (DeviceList.py:1376): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: line 1142 (gnome_canvas_path_def_any_closed): assertion `path != NULL' failed Traceback (most recent call last): File "/usr/share/hwbrowser/DeviceList.py", line 204, in ? reread_device_type_list () File "/usr/share/hwbrowser/DeviceList.py", line 151, in reread_device_type_list list.set (iter, 0, hardware.get_category_string (key), 1, key) File "/usr/share/hwbrowser/DeviceList.py", line 92, in get_category_string return self.type_mapping[key] KeyError: 20 Firewire fix worked for me as well. The browser doesn't show my firewire devices,though - three hard drives, two CD-Rs, and one DVD-R, on two host adapters, a Creative Labs Audigy and a generic OHCI card. I don't think that's the browser's fault, kudzu isn't finding and configuring the drives. I'm having the same problem on an HP Pavilion 7960 desktop. From a console or through GUI, the window outline for hwbrowser appears and then exits. The following two lines appear in the messages.log file each time hwbrowser is started: Oct 15 11:17:30 localhost kernel: parport0: PC-style at 0x378 (0x778) [PCSPP,TRISTATE,EPP] Oct 15 11:14:53 localhost kernel: parport0: irq 7 detected The "irq 7" makes me wonder if it's related to RH 8.0's failure to use the Ensoniq ES1371 sound card in this system. Still, hwbrowser should stay alive. Same here on an Atlon XP 2000+ i686 Redhat 8.0. Removing and re-installing does not make a difference. *** Bug 76632 has been marked as a duplicate of this bug. *** Firewire fix worked for me also. All devices including firewire show. Still gets canvas errors, but does run. *** Bug 76666 has been marked as a duplicate of this bug. *** *** Bug 75767 has been marked as a duplicate of this bug. *** Same problem here as well (Sony vaio FX702 laptop): Traceback (most recent call last): File "/usr/share/hwbrowser/DeviceList.py", line 203, in ? reread_device_type_list () File "/usr/share/hwbrowser/DeviceList.py", line 150, in reread_device_type_list list.set (iter, 0, hardware.get_category_string (key), 1, key) File "/usr/share/hwbrowser/DeviceList.py", line 91, in get_category_string return self.type_mapping[key] KeyError: 20 I get the same results. See enclosed strace-hwbrowser .txt file Created attachment 85304 [details]
console strace of "hwbrowser"
Tried the workaround described above (editing the DeviceList.py file), and it helps. The Hardware Browser will now start up without crashing. HOWEVER... it only works properly if I do not have a hard drive partition mounted via the Firewire interface. If a partition on the Firewire hard drive is mounted, Hardware Browser hangs (but does not crash) until the partition is unmounted. *** Bug 79527 has been marked as a duplicate of this bug. *** I had the same problem on a sony vaio laptop with a fire wire port. The one line entry suggested worked and the hwbroser didnt die, but compalined of stuff. Thanks a lot guys. I am running RH 8.0 on a Dell Inspiron 8100. I get exactly the same problem as described in other messages above. The proposed fix also cured it: Add the following to line 67 of /usr/share/hwbrowser/DeviceList.py : kudzu.CLASS_FIREWIRE : _("Firewire"), I've applied the line to cvs and built a new version that should appear in Rawhide tomorrow. hwbrowser-0.7-2 should fix the problem. QA, please verify. Note that you will need to test the fix with a machine that has a firewire port. *** Bug 75695 has been marked as a duplicate of this bug. *** *** Bug 73225 has been marked as a duplicate of this bug. *** *** Bug 78100 has been marked as a duplicate of this bug. *** I would just like to point out that the DeviceList.py change did not COMPLETELY fix the problem for me. Yes, it allowed HW Browser to start up without crashing; but HW Browser would still hang (after starting) if there was a partition from the Firewire hard drive mounted. uchima: please open a separate bug report for that. It is a related problem, but ultimately separate from the original report. This problem (the "GnomeCanvas-CRITICAL" error mentioned above) started occuring for me when I upgraded my motherboard from an Abit KT7A-raid to the Asus A7V8X-raid. I did NOT have this problem proir to the installation of my new motherboard. Also, I get a similar error when I try to run the "redhat-config-date" utility. I also have an A7V8X motherboard and the stock 8.0 hwbrowser crashes but it now works after applying the fix from comment 25. Nick: the "GnomeCanvas-Critical" errors you are seeing is from the Gnome Canvas widget, which both hwbrowser and redhat-config-date (and anaconda for that matter, although the error messages are hidden on VC1). I recommend you file a bug against gnome-python2-canvas, which is the RPM that the python gnome canvas module is in. *** Bug 81627 has been marked as a duplicate of this bug. *** I am using Red Hat 8.0 in an HP Pavilion 750n system and the hwbrowser bug also got me. However, by adding the above talked about fix to line 67 of DeviceList.py took care of things quite nicely. Thanks... *** Bug 81983 has been marked as a duplicate of this bug. *** *** Bug 82070 has been marked as a duplicate of this bug. *** Brent, you could probably answer this best. I am having this same problem during the initial install. I am on a Dell Dimension 8100. Since it is the initial install, it fails, exits abnormally and states it is safe to reboot the machine. After reboot I have a grub prompt. How can I make this change mentioned in comment 25 above? For your reference, I am installing from a DVD out of the book "The Complete Reference Red Hat Linux 8 DVD Edition" by Richard L. Petersen.ISBN 0-07-222646-3. ajwalla: hwbrowser does not run during the initial install. Are you sure you're seeing the exact same problem? Maybe I'm not so sure. Here's what was on the screen when the install failed: (anaconda:87): GnomeCanvas - CRITICAL **: file gnome-canvas-path-def.C: line 1142 (gnome_canvas_path_def_any_closed): assertion 'path != NULL' failed install exited abnormally ... at this point, abunch of shutdown message lines. last line was something like "it is now safe to reboot your system". This all occured just about a minute after that last next click before the formats started. I was unable to cut and paste, so this is from hand written notes. :^) Andy: I'm certain that this is a different problem. The messages related to GnomeCanvas don't have anything to do with either one of these crashes. Both anaconda and hwbrowser use a GnomeCanvas widget. The python bindings for GnomeCanvas for some reason writes a lot of debug messages to the command line, but it doesn't cause a crash. It's the part of the debug that starts with "Traceback (most recent last call)" that really provides information about the crash. Do you see any information like that with the installer when it crashes? Brent: I reran the install in text mode. I got an option to load the bootloader in the MBR or at the /boot mount point, there was no default selected. I don't remember getting this option in the several install runs I did in graphics mode. I selected MBR. This is the only difference in my last two install attempts other than my last attempt using text mode vrs graphics. The last install was successful. If I can help you trouble shoot this, please let me know. Andy: The thing for you to do is to file a new bug report against anaconda (the installer). The crash that you originally saw is clearly caused by something in the installer's GUI code. It will really help the installer team if you can provice the Traceback debug info. When the installer crashes, it should pop up a dialog asking you to insert a floppy and it will copy the debug data to it. Then you can attach that data to the bug report when you file it. In any case, I'm convinced that the problem you are seeing has nothing to do with hwbrowser and belongs in a separate bug report. Thanks for your help. *** Bug 83493 has been marked as a duplicate of this bug. *** *** Bug 85720 has been marked as a duplicate of this bug. *** I'm getting the same problem as Andy, redhat 8 install failing, but on a celeron 300a system. This is my 3rd time trying, same error every time, even tried installing with kde instead of gnome and are still getting this error. (anaconda:76): GnomeCanvas-CRITICAL **: file gnome-canvas-path-def.c: line 1142 (gnome_canvas_path_def_any_closed): assertion 'path != NULL' failed above is repeated 8 times, then says install exited abnormally sending termination signals...done sending kill signals...done disabling swap... (snip) unmounting filesystems... (snip) ejecting /tmp/cdrom... you may safely reboot your system *** Bug 77583 has been marked as a duplicate of this bug. *** Created attachment 91053 [details]
strace output for hwbrowser crash.
Ya, I am realy vexed!
Everything works fine for me under 9. *** Bug 89202 has been marked as a duplicate of this bug. *** There is a stack of 64 bugs that have been in Modified state for a long period of time. I am closing these as Rawhide now. If you find that the issue is not fixed, please reopen this report. |