Bug 371911
Summary: | Anaconda crashes on x86_64 DVD | ||||||
---|---|---|---|---|---|---|---|
Product: | [Fedora] Fedora | Reporter: | Mike Duncan <shadowace_1> | ||||
Component: | anaconda | Assignee: | Anaconda Maintenance Team <anaconda-maint-list> | ||||
Status: | CLOSED DUPLICATE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||
Severity: | high | Docs Contact: | |||||
Priority: | low | ||||||
Version: | 8 | CC: | angahith, mclayton, xeoslaenor | ||||
Target Milestone: | --- | ||||||
Target Release: | --- | ||||||
Hardware: | x86_64 | ||||||
OS: | Linux | ||||||
Whiteboard: | |||||||
Fixed In Version: | Doc Type: | Bug Fix | |||||
Doc Text: | Story Points: | --- | |||||
Clone Of: | Environment: | ||||||
Last Closed: | 2007-11-12 18:45:48 UTC | Type: | --- | ||||
Regression: | --- | Mount Type: | --- | ||||
Documentation: | --- | CRM: | |||||
Verified Versions: | Category: | --- | |||||
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||
Cloudforms Team: | --- | Target Upstream Version: | |||||
Embargoed: | |||||||
Attachments: |
|
Description
Mike Duncan
2007-11-08 21:45:13 UTC
Please attach the complete error to this bug report. I'm getting exactly the same error. Immediately after the disc check (no matter if it's executed or not) the installer seems to crash, with this message: http://img.photobucket.com/albums/v227/angahito/P1000359.jpg (In reply to comment #1) > Please attach the complete error to this bug report. Please see comment #2. This screen shot looks exactly like what I am getting. (In reply to comment #1) > Please attach the complete error to this bug report. On further research, I inserted the DVD into the drive without mounting it. I then performed a sha1sum on /dev/hda (the dvd drive). The result is: 6cbca30ea6363831c9ba429a55c265765dc3b496 which is different than the ISO image. Thinking I had a bad burn, I re-burned the iso, and performed the same sha1sum check. I received the exact same sum. I have a difficult time believing the image could have had an error at the exact same spot in both burns. Am I misunderstanding SHA1SUM in thinking that the /dev/hda sum and the iso sum should be identical? Just for clarification, this happens with both, graphical and text install option. The burn was verified and the disc check was successful, so it's highly unlikely that the media is at fault. While i haven't found any reports about incompatibility, maybe information about my system could be helpful. The machine, i'm trying to install it on, is an FSC Amilo A1630 notebook with the following specs: AMD Athlon 64 Mobile 3700+ 1024 MB PC3200 Corsair DDR RAM Uniwill 258KA0 Motherboard SIS 755 Chipset ATI Radeon 9700 Mobility Thanks Can you switch to tty3 and see what the last messages there are? (In reply to comment #6) > Can you switch to tty3 and see what the last messages there are? Here's my output: INFO : trying to mount CD device scd0 on /mnt/source INFO : mntloop loop0 on /mnt/runtime as /mnt/source/images/stage2.img fd is 8 INFO : support = 0 INFO : Loading SELinux policy INFO : getting ready to spawn shell now INFO : modules to insert ide-cd INFO : load module set done INFO : modules to insert INFO : load module set done INFO : modules to insert INFO : load module set done INFO : modules to insert pcmcia_core p* pcmcia On the last line above, the p* is actually a "p" followed immediately by a square block, indicative of an unprintable character. The two lines above it show some kind of modules being inserted, but it does not list what that module is. Those lines are transcribed as is. Any typos are mine, as I wrote these messages down and re-typed them here. Per the poster above, I will also list my equipment: Dell Latitude D830 Centrino Duo with 2G of RAM NVidia Video I use the ipw3945 wireless driver, but it sux Hope that helps! Can you boot adding 'loglevel=debug' to the kernel command line and do the same? (In reply to comment #8) > Can you boot adding 'loglevel=debug' to the kernel command line and do the same? Here we go: INFO : trying to mount CD device scd0 on /mnt/source INFO : mntloop loop0 on /mnt/runtime as /mnt/source/images/stage2.img fd is 8 DEBUG : MD%SUM -> 15055ae76b9f3a2503c7a9af9792aa05 DEBUG : skipsectors = 15 INFO : support = 0 INFO : Loading SELinux policy INFO : getting ready to spawn shell now INFO : modules to insert ide-cd DEBUG : getModuleLocation: 2.6.23.1-42.fc8/x84_64 DEBUG : module(s) ide-cd not found INFO : load module set done INFO : modules to insert DEBUG : getModuleLocation: 2.6.23.1-42.fc8/x84_64 INFO : load module set done INFO : modules to insert DEBUG : getModuleLocation: 2.6.23.1-42.fc8/x84_64 INFO : load module set done INFO : modules to insert pcmcia_core pcmcia DEBUG : getModuleLocation: 2.6.23.1-42.fc8/x84_64 It looks like Anaconda cannot find the ide CD drive, or the module isn't being loaded correctly. Fedora 7 actually failed gracefully here, asking where the location of the install media could be found. Now it's gotten worse, and just puking all over the screen. Also, I noticed in the last line that the unprintable character is not being output any more. As states above, typos are mine. For instance, x84_64 should be x86_64 I don't think the CD drive is the issue. I just tried installing it with "askmethod" and using the iso i still got on my hd. Same result. Same problem. I hadn't this problem on F8T3 With Askmethod and FTP, Anaconda start starts. I am having this same problem with the i386 DVD, Media check comes back ok, sha1sums checked ok Askmethod seems to fix the issue here, even installing from DVD Seems the same as https://bugzilla.redhat.com/show_bug.cgi?id=372291 Created attachment 254631 [details]
screenshot of error message
As already mentioned, using "askmethod" doesn't fix the bug for me, but i just
noticed that i actually get this error message, when using "askmethod" in text
mode.
Another possible fix is the "nopcmcia" option - that worked for me, see https://bugzilla.redhat.com/show_bug.cgi?id=374311 The original cause of the crash for me was "invalid pointer", not "doubly-linked list error" (which "askmethod" seems to get around?), so there are probably two different crashes we're seeing here. I tried out random things, earlier. Running the installer with resolution=1280x800, which is my screen's native resolution, made it run perfectly. Not even editing the packages (which was still reported as problematic by others) caused any trouble. I'm posting from F8 right now. *** This bug has been marked as a duplicate of 374311 *** |