Summary: | sys-boot/lilo-22.8-r2: Fatal: cache_add: LILO internal error (with loop device) | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Michał Górny <mgorny> |
Component: | Current packages | Assignee: | Joshua Kinard <kumba> |
Status: | CONFIRMED --- | ||
Severity: | normal | CC: | base-system, hlein |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: |
The full verbose log of lilo
The lilo.conf file |
Description
Michał Górny
![]() ![]() ![]() ![]() Created attachment 228205 [details]
The full verbose log of lilo
Created attachment 228207 [details]
The lilo.conf file
I get a similar error when loop[01] are used by loop-mounted squashfs. Michał: Can you still reproduce this on lilo-24.2? Curious if it may have been fixed upstream in the last release they put out. (In reply to Michał Górny from comment #3) > I get a similar error when loop[01] are used by loop-mounted squashfs. I don't have any machines w/floppy drives any more, but I attempted using squashfs, and could not reproduce. Using: sys-kernel/gentoo-sources-5.4.124:5.4.124 sys-boot/lilo-24.2:0 sys-fs/squashfs-tools-4.4_p1:0 # for A in 0 1 ; do mkdir -p squash_${A} && echo "$A" >squash_${A}/${A} && mksquashfs squash_${A} ${A}.squashfs -quiet && losetup /dev/loop${A} ${A}.squashfs ; done [======================================================================|] 1/1 100% [======================================================================|] 1/1 100% # losetup -a /dev/loop1: [2308]:1191840 (/root/tmp/1.squashfs) /dev/loop0: [2308]:1190815 (/root/tmp/0.squashfs) # lilo -v99 >lilo_v99.out 2>&1 # echo $? 0 # egrep -1 'Fatal|pf_hard_disk_scan:.*/dev/loop' lilo_v99.out pf_hard_disk_scan: (9,1) /dev/md1 pf_hard_disk_scan: (7,0) /dev/loop0 pf_hard_disk_scan: (7,1) /dev/loop1 0800 1DDCA02C /dev/sda So, it may be fixed w/current lilo; if you can still reproduce w/squashfs, please let us know more details from that setup. |