Bug 1250712
| Summary: | kernel panic when booting a LiveUsb created with liveusb-creator | ||||||
|---|---|---|---|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | Giulio 'juliuxpigface' <juliux.pigface> | ||||
| Component: | liveusb-creator | Assignee: | Luke Macken <lmacken> | ||||
| Status: | CLOSED EOL | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||
| Severity: | unspecified | Docs Contact: | |||||
| Priority: | unspecified | ||||||
| Version: | 22 | CC: | awilliam, bcl, juliux.pigface, lmacken, mbriza, pfrields, robatino, satellitgo | ||||
| Target Milestone: | --- | ||||||
| Target Release: | --- | ||||||
| Hardware: | Unspecified | ||||||
| OS: | Unspecified | ||||||
| Whiteboard: | RejectedBlocker | ||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||
| Doc Text: | Story Points: | --- | |||||
| Clone Of: | Environment: | ||||||
| Last Closed: | 2016-07-19 17:22:31 UTC | Type: | Bug | ||||
| Regression: | --- | Mount Type: | --- | ||||
| Documentation: | --- | CRM: | |||||
| Verified Versions: | Category: | --- | |||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||
| Embargoed: | |||||||
| Attachments: |
|
||||||
Sorry. It actually was "Fedora-Live-KDE-i686-23_Alpha-1.iso" and not "Fedora-Live-KDE-x86_64-23_Alpha-1.iso". I don't know if this detail should be valuable, though. Proposed as a Freeze Exception for 23-alpha by Fedora user juliuxpigface using the blocker tracking app because: The criteria "2.2.1 Release-blocking images must boot" (https://fedoraproject.org/wiki/Fedora_23_Alpha_Release_Criteria#Release-blocking_images_must_boot) links a wiki page, which lists the officially supported methods for creating a LiveUsb. I understand that live-usb-creator is one of those (https://fedoraproject.org/wiki/How_to_create_and_use_Live_USB#Using_liveusb-creator_.28Windows_and_Fedora.2C_graphical.2C_non-destructive.29). booting a USB stick attached to a VM isn't a very common case. Did you try booting it with an actual metal system? (In reply to Adam Williamson from comment #3) > booting a USB stick attached to a VM isn't a very common case. Did you try > booting it with an actual metal system? I've just tried on bare metal with the same stick. I confirm the issue: the behaviour is the same (with the addition of blinking caps lock key). I wrote the x86_64 Workstation live with luc from Fedora 22 the same way you did, and that boots fine for me. I'll try with some i686 images in a bit. I did some testing:
1. Recreated the partitioning table for the stick I was using while encountering this bug.
2. Wrote Fedora-Live-LXDE-x86_64-23_Alpha-2.iso on the stick (dd method)
3. Tried to boot; everything went fine.
4. Wrote Fedora-Live-LXDE-x86_64-23_Alpha-2.iso on the same stick (dd method)
5. Tried to boot again; everything went fine again.
So... I'm starting to think I reported this bug too early, maybe I've somehow been just unlucky and the bug isn't reproducible at all. Or... Perhaps there are some race conditions which I need to discover.
The following might be one of these:
(In reply to Giulio 'juliuxpigface' from description)
> Steps to Reproduce:
> 1. Pick a stick, In my case, it had got a working live of Fedora 23 TC (created > with live-iso-to-disk).
Adam, have you tried to create your LiveUsb with live-iso-to-disk before using live-usb-creator?
The stick I used would've had an image written with 'dd' on it previously. Ok, it seems that my guess was right. I've reproduced it. The steps are these. On a Fedora 22 x86_64 host: 1. # livecd-iso-to-disk --format --reset-mbr Fedora-Live-LXDE-x86_64-23_Alpha-2.iso /dev/sdX 2. Verify that LXDE boots normally. 3. # liveusb-creator --reset-mbr 4. Select "Fedora-Live-KDE-i686-23_Alpha-2.iso" and "dd" method. 5. Kernel panic (as described above) while booting KDE. So, live-usb-creator fails if the desired stick contains a Live previously created with livecd-iso-to-disk. But... Now I'm not sure if we should flag this as "Freeze Exception", since it's more like a corner case. this happens with any dd USB that was previously written do to the iso CD format on USB this can fixed in gparted by writing a new MBR on the USB. No point having this proposed as an Alpha FE now Alpha has been approved. So does this happen only if you use dd from liveusb-creator or does it even when you copy the image with dd directly from commandline? (Along the lines of dd if=/path/to/iso of=/dev/sdX) (In reply to Martin Bříza from comment #11) > So does this happen only if you use dd from liveusb-creator or does it even > when you copy the image with dd directly from commandline? > (Along the lines of dd if=/path/to/iso of=/dev/sdX) Sorry for the late response. I confirm it happens only from liveusb-creator. Running 'dd' directly, works as expected. Choosing it as method for liveusb-creator, causes the issue. Though 'dd' might not be relevant, to be honest I haven't tried the 'cp' one yet. (In reply to Martin Bříza from comment #11) > So does this happen only if you use dd from liveusb-creator or does it even > when you copy the image with dd directly from commandline? > (Along the lines of dd if=/path/to/iso of=/dev/sdX) This morning I've encountered this using 'dd' directly. It was a Live previously created with livecd-iso-to-disk. But, as per comment #12 and comment #5, this bug isn't always reproducible. As a side note, this time I've written a i686 image (Fedora-Live-MATE_Compiz-i686-23-TC11.iso) and the stick contained a previous i686 before this last try. It's always from Fedora 22, but a new try from Fedora 23 (with the latest updates applied) would be interesting, I guess. The issue is far far more likely to be between the image and the system than it is in the writing process. 'dd' is very, very simple and quite unlikely to behave differently between 22 and 23. Proposed as a Blocker and Freeze Exception for 23-beta by Fedora user juliuxpigface using the blocker tracking app because: Sorry, I've somehow forgotten about this one, even if it was first discovered with Alpha. Anyway... The criteria "2.2.1 Release-blocking images must boot" (https://fedoraproject.org/wiki/Fedora_23_Alpha_Release_Criteria#Release-blocking_images_must_boot) links a wiki page, which lists the officially supported methods for creating a LiveUsb. I understand that live-usb-creator is one of those (https://fedoraproject.org/wiki/How_to_create_and_use_Live_USB#Using_liveusb-creator_.28Windows_and_Fedora.2C_graphical.2C_non-destructive.29) Well, actually this is more like a corner case. I'd probably not block on this and prefer to mark it as "FE" (and list it in common bugs). But it would be nice if we we can get more testing and more discussion about this. Changing to FinalBlocker and FinalFreezeException per https://lists.fedoraproject.org/pipermail/test/2015-October/128188.html . (You don't need to use the blocker tracking app to propose a blocker, you can just enter the "Blocks:" field, if you belong to the fedorabugs group, and it appears you already do). Discussed at 2015-10-19 blocker review meeting: https://meetbot.fedoraproject.org/fedora-blocker-review/2015-10-19/f23-blocker-review.2015-10-19-16.00.html . It seems there are two somewhat different cases being discussed here. jpigface seems to have issues sometimes after a dd write to a stick that has previously had a Fedora image on it. This would be something new and unexpected. However, no-one else seems to be seeing that - multiple people in the meeting said they hadn't seen such a problem. satellit seems to be just reporting something that's already fairly well known: liveusb-creator isn't always the best at wiping existing partition tables, particularly the complex hybridized ones that dd and livecd-iso-to-disk --efi write. This is a pain but it's known and not new, and the reason why luc is recommended less strongly than dd or litd on the docs page. Overall we felt there isn't enough of a clearly demonstrated, commonly-encountered problem here to block on, so the bug was rejected as a blocker. This could be re-proposed as blocker or FE if a specific, reliable reproducer is identified. Fedora 22 changed to end-of-life (EOL) status on 2016-07-19. Fedora 22 is no longer maintained, which means that it will not receive any further security or bug fix updates. As a result we are closing this bug. If you can reproduce this bug against a currently maintained version of Fedora please feel free to reopen this bug against that version. If you are unable to reopen this bug, please file a new report against the current release. If you experience problems, please add a comment to this bug. Thank you for reporting this bug and we are sorry it could not be fixed. |
Created attachment 1059624 [details] kernel panic Description of problem: I can't boot a LiveUsb with Fedora 23 Alpha RC1. I encounter a kernel panic. Version-Release number of selected component (if applicable): liveusb-creator-3.14.2-1.fc22.noarch How reproducible: I haven't tried to reproduce it yet. Steps to Reproduce: 1. Pick a stick, In my case, it had got a working live of Fedora 23 TC (created with live-iso-to-disk). 2. Run 'liveusb-creator --reset-mbr' as root. 3. Select the ISO "Fedora-Live-KDE-x86_64-23_Alpha-1.iso" 4. Don't set persistence, use the "dd" method. 6. Create the LiveUsb. 7. Try to boot the new system on a qemu-kvm guest. Actual results: 1. I encounter the kernel panic (screenshot attached). Expected results: 1. The live image should boot.