Bug 517260
| Summary: | liveinst fails at partitioning screen | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Sebastian Vahl <fedora> |
| Component: | anaconda | Assignee: | David Lehman <dlehman> |
| Status: | CLOSED RAWHIDE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | medium | Docs Contact: | |
| Priority: | low | ||
| Version: | rawhide | CC: | andreasfrische+redhat, atodorov, awilliam, bugzilla, ddumas, gmureddu, harald, hicham.haouari, jlaska, JustinBard, lakshminaras2002, peher80, rwilliam, tcallawa, tomek, vanmeeuwen+fedora, wtogami |
| Target Milestone: | --- | Keywords: | Reopened |
| Target Release: | --- | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | anaconda_trace_hash:38997998bbbb9120ebbe77474fb3a0972727becc7598403e37d05570140560f6 NeedsRetesting | ||
| Fixed In Version: | anaconda-12.39-1 | Doc Type: | Bug Fix |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2009-10-31 16:06:50 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: | |||
| Bug Depends On: | |||
| Bug Blocks: | 531381 | ||
| Attachments: | |||
This should be fixed in anaconda 12.16. When will -16 hit F12Alpha/Rawhide? I've added it at as blocker for F12Alpha because atm there is no way to install live images (or test them properly). (In reply to comment #3) > I've added it at as blocker for F12Alpha because atm there is no way to install > live images (or test them properly). This is not accurate. You should be able to test live images as long as the backing device is not a partition (eg: cd, lv). (In reply to comment #3) > I've added it at as blocker for F12Alpha because atm there is no way to install > live images (or test them properly). Sebastian: I'm told that booting the live image with the argument "nompath" is sufficient to work around this issue. Are you able to confirm this workaround? (In reply to comment #5) > (In reply to comment #3) > > I've added it at as blocker for F12Alpha because atm there is no way to install > > live images (or test them properly). > > This is not accurate. You should be able to test live images as long as the > backing device is not a partition (eg: cd, lv). You were right: I've used a local partition to boot the live images. After using an usb stick, liveinst isn't failing anymore. (In reply to comment #6) > (In reply to comment #3) > > I've added it at as blocker for F12Alpha because atm there is no way to install > > live images (or test them properly). > > Sebastian: I'm told that booting the live image with the argument "nompath" is > sufficient to work around this issue. Are you able to confirm this workaround? I have only tested this when booting from a local partition and there it is failing, too. Moving this to blocks:F12Beta based on the updated feedback around using a live image with a USB stick. Should a workaround be required, once anaconda-12.16 is built you may use a rawhide nightly live image (see http://alt.fedoraproject.org/pub/alt/nightly-composes/). Thanks! *** Bug 519026 has been marked as a duplicate of this bug. *** *** Bug 521102 has been marked as a duplicate of this bug. *** I confirm this with anaconda 12.20-1.fc12 Used local partition as liveimg. passing nompath as argument when booting the liveimg didn't help Comments 11, 12 (and the dupe reports) suggest this bug still exists. Returning to ASSIGNED. -- Fedora Bugzappers volunteer triage team https://fedoraproject.org/wiki/BugZappers This bug was reviewed at the Fedora 12 Beta Blocker bug meeting on 2009-09-11. Based on a review of the comments it was determined that this bug is not fix. Moving back to ASSIGNED. This bug was reviewed at the Fedora 12 Beta Blocker bug meeting on 2009-09-18. It needs to be retested. Please help us by reporting if the latest package addresses the originally reported issue. Thanks. Created attachment 361710 [details]
anaconda 12.25 exception report
problem persists with the spins from 17.09.09
fixed in rawhide how could you possibly have tested? live builds since sept 18th don't boot on any hardware, as far as I'm aware... -- Fedora Bugzappers volunteer triage team https://fedoraproject.org/wiki/BugZappers i managed to boot the kde spin from 21.09 ( by getting into the emergency shell , unmounting the root partition , running /sbin/dmsqashfs<something> /dev/sda3, and exiting the shell ) Still broken while testing anaconda-12.32 live image install. See updated traceback attached to DUP'd bug (https://bugzilla.redhat.com/attachment.cgi?id=363500). Added anaconda trace hash from bug that was DUP'd to this issue. I'm having difficulty reproducing this failure. The problem does not appear using procedure noted in comment#0. However, when I recreated this issue, it was during the anaconda storage test day. According to hhoyer who also hit this issue, it was while testing an encrypted logical '/' volume on a RAID1 /dev/md0 physical volume. I'm not able to hit this issue during testing today. Can someone else confirm that the problem is resolved using the latest live image (see http://alt.fedoraproject.org/pub/alt/nightly-composes/desktop/). dlehman: Is there another way this defect could occur outside of the reproducer posted in comment#0? It should occur any time you use a usb-backed live install and choose custom partitioning. It will not occur with livecd, only liveusb. Discussed with dlehman and jkeating about this bug. Harald Hoyer and I have reproduced this issue while testing RAID installs, but I've no longer been able to reproduce the issue using the documented procedure in this bug. The agreement is to move this issue off the F12Beta list, and to the F12BLocker list given that the producer no longer seems to be as straight forward as noted in comment#0. I think the trick to reproducing this failure may be using a partitioned device with more than one partition as the live device (eg: usb or other disk). If it is not possible to install the liveimg from a local partition, please state it in the wiki or the errata so people can stop complaining about this problem. Just as a note, it is possible to install the live iso from the harddisk by using grub v2. Hicham: um. What? That's not what this bug is about, so far as I can tell. -- Fedora Bugzappers volunteer triage team https://fedoraproject.org/wiki/BugZappers Adam, plz see comment 0 and 7. Created attachment 365387 [details]
Attached traceback automatically from anaconda.
Should be fixed in anaconda-12.39-1. I don't see 12.39 built. *** Bug 530221 has been marked as a duplicate of this bug. *** Created attachment 365811 [details]
Attached traceback automatically from anaconda.
12.39 was tagged on 20091024. Can you please re-test with a build from that date or later and confirm the fix? Thanks. -- Fedora Bugzappers volunteer triage team https://fedoraproject.org/wiki/BugZappers Created attachment 366939 [details]
Screenshot demonstrating fix.png
Confirmed the fix with anaconda-12.41-1.fc12.i686 and a USB live image containing the following partition setup:
Model: SanDisk Cruzer (scsi)
Disk /dev/sdb: 4027MB
Sector size (logical/physical): 512B/512B
Partition Table: msdos
Number Start End Size Type File system Flags
1 512B 2020MB 2020MB primary fat32 boot, lba
2 2020MB 4027MB 2007MB primary fat32 lba
Created attachment 372922 [details]
Attached traceback automatically from anaconda.
I have this bug with the install dvd when trying to create a custom layout More information on this bug : I had a raid and another harddrive. When I unplugged the other harddrive I could go through the partitionning task. (In reply to comment #38) > More information on this bug : > I had a raid and another harddrive. > When I unplugged the other harddrive I could go through the partitionning task. Please open a separate bug report for your problem. Although it ends up causing a very similar traceback, yours is a different failure. Thanks. Created attachment 381128 [details]
Attached traceback automatically from anaconda.
Created attachment 381957 [details]
Attached traceback automatically from anaconda.
I cannot install F12 using netinstall. Is there update for anaconda available somewhere? Created attachment 381963 [details]
Attached traceback automatically from anaconda.
(In reply to comment #41) > Created an attachment (id=381957) [details] > Attached traceback automatically from anaconda. Tomasz, Can you please open a new bug and describe in some detail what you did leading to this failure? (pre-existing storage layout, steps taken in installer) Although the traceback looks the same as the others in this bug, yours is different. Thanks. Created attachment 410612 [details]
Attached traceback automatically from anaconda.
Created attachment 424307 [details]
Attached traceback automatically from anaconda.
|
Created attachment 357291 [details] full anaconda traceback saved to file Description of problem: After selecting a custom layout liveinst fails with this traceback: anaconda 12.13-1.fc12 exception report Traceback (most recent call first): File "/usr/lib/anaconda/iw/partition_gui.py", line 934, in populate devname = "%s" % device.path File "/usr/lib/anaconda/iw/partition_gui.py", line 1356, in getScreen self.populate(initial = 1) File "/usr/lib/anaconda/gui.py", line 1285, in setScreen new_screen = self.currentWindow.getScreen(anaconda) File "/usr/lib/anaconda/gui.py", line 1202, in nextClicked self.setScreen () AttributeError: 'NoneType' object has no attribute 'path' Version-Release number of selected component (if applicable): anaconda 12.13-1.fc12 How reproducible: anaconda-12.13-1.fc12.x86_64 Steps to Reproduce: 1. create a live image (KDE here) 2. install with liveinst 3. select custom layout and proceed Actual results: crash Expected results: works