Please read this before reporting a bug:
https://wiki.archlinux.org/title/Bug_reporting_guidelines
Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.
REPEAT: Do NOT report bugs for outdated packages!
https://wiki.archlinux.org/title/Bug_reporting_guidelines
Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.
REPEAT: Do NOT report bugs for outdated packages!
FS#28170 - mkinitcpio doesn't deal correctly with symlinked library directories
Attached to Project:
Arch Linux
Opened by Leslie Polzer (skypher) - Monday, 30 January 2012, 09:28 GMT
Last edited by Dave Reisner (falconindy) - Tuesday, 31 January 2012, 14:52 GMT
Opened by Leslie Polzer (skypher) - Monday, 30 January 2012, 09:28 GMT
Last edited by Dave Reisner (falconindy) - Tuesday, 31 January 2012, 14:52 GMT
|
Detailskmod 4-1
mkinitcpio 0.8.2-1 HOOKS="base udev ide scsi sata usbinput encrypt filesystems shutdown fsck" An old compatibility symlink /usr/X11R6 -> . made me end up with zero-truncated versions of the libs in the initramfs image's /usr/lib directory. These were libkmod and libz which are needed by udev to set up the root device, leaving me with a broken system. I haven't debugged this in more detail but it definitely seems to be a bug, because we never want to end up with zero-truncated files in the initramfs. [ Originally from https://bbs.archlinux.org/viewtopic.php?id=134024&p=1 ] |
This task depends upon
Closed by Dave Reisner (falconindy)
Tuesday, 31 January 2012, 14:52 GMT
Reason for closing: Works for me
Additional comments about closing: Cannot reproduce, though some of this code needs reviewing.
Tuesday, 31 January 2012, 14:52 GMT
Reason for closing: Works for me
Additional comments about closing: Cannot reproduce, though some of this code needs reviewing.
Comment by Dave Reisner (falconindy) -
Monday, 30 January 2012, 20:24 GMT
I'm not sure how a symlink from /usr/X11R6 to anywhere could cause this to happen given that we're never pulling anything from /usr/X11R6. I agree that there could be a problem here (we're somewhat gullible about the return values from the add_* functions), but I need to understand why this is breaking. Simply symlinking /usr/X11R6 back to /usr isn't enough to trigger any issues. Any way you could expand on this?
Comment by Leslie Polzer (skypher) -
Monday, 30 January 2012, 21:11 GMT
Alright, I'll try to dig a little deeper this week.
Comment by Leslie Polzer (skypher) -
Tuesday, 31 January 2012, 12:57 GMT
Cannot reproduce. Must have been some other kink in my system that contributed to this but is now gone.