Bug 141570
Summary: | anaconda crashes attempting to mount sysimage | ||
---|---|---|---|
Product: | [Fedora] Fedora | Reporter: | G.Wolfe Woodbury <redwolfe> |
Component: | anaconda | Assignee: | Jeremy Katz <katzj> |
Status: | CLOSED RAWHIDE | QA Contact: | Mike McLean <mikem> |
Severity: | high | Docs Contact: | |
Priority: | medium | ||
Version: | rawhide | CC: | aviro, cworth, davej, jim.cornette, jorton, nobody+pnasrat, taj, twaugh |
Target Milestone: | --- | ||
Target Release: | --- | ||
Hardware: | i386 | ||
OS: | Linux | ||
URL: | http://wolves.homeip.net/~ggw/texts/anacdump.txt | ||
Whiteboard: | |||
Fixed In Version: | Doc Type: | Bug Fix | |
Doc Text: | Story Points: | --- | |
Clone Of: | Environment: | ||
Last Closed: | 2005-08-15 16:57:55 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: |
Description
G.Wolfe Woodbury
2004-12-02 05:40:18 UTC
I see this too. *** Bug 142139 has been marked as a duplicate of this bug. *** I just hit this as well. And I've got a machine sitting here in the crashed state if that would be useful. It's clear that the bind mount fails: # mount -o bind /dev /mnt/sysimage/dev mount: Mounting /dev on /mnt/sysimage/dev failed: Invalid argument I doubt that that failure is anaconda's fault, but perhaps it would be useful for anaconda to cleanly deal with the failure at least until the underlying bug is fixed? I wouldn't know where to start digging through this but: # mount /dev /mnt/sysimage/ --bind # mount |grep sysimage /dev on /mnt/sysimage type none (rw,bind) *** Bug 142776 has been marked as a duplicate of this bug. *** As an additonal note. This problem happened on the upgrade, installation with automatic partitioning and with selecting what partition scheme to use. The attachment in duplicate bug 142776 shows the anaconda dump text from the upgrade attempt. Back to FC3 with a clean installation. taj: no, you need to run that on VT2 *during* stage2 to see it. Is this a kernel change or an anaconda problem? CCing davej.. I booted a new disk to confirm. The busybox syntax is indeed different with regards to -bind. I also notice on VT2 during stage2 that while bind mounting /dev with "mount /dev /mnt/sysimage/dev -o bind" fails, "mount /dev/pts /mnt/sysimage/dev -o bind" works. I just tried this because /dev/pts shows as a mount point while /dev does not. This is being caused by bind mounts off of the (initramfs-based) rootfs not working. Al -- is this as designed? If so, I can mount a ramfs on the rootfs to use for /dev so that I can later bind mount that instead, but I want to make sure there's not something lingering in the kernel here. You are not allowed to create bindings *to* rootfs. You can, of course, mount ramfs on rootfs's /dev and bind to it, no problems with that. Committed changes so that we should create a tmpfs /dev like on the installed system. Trying today's Raw Hide I got a traceback saying something like: "select expects exactly two arguments" context was select("kernel-devel"). Dunno whether this is before or after this bug. rawhide of 2005.01.09 works okay G.Wolfe Woodbury, Thank you for verifying this |