On 07/11/2014 06:05 PM, Jameson Graef Rollins wrote: > On Fri, Jul 11 2014, Daniel Kahn Gillmor <d...@fifthhorseman.net> wrote: >> On 07/11/2014 08:43 AM, jhcha54008 wrote: >>> The following patch just change the owner of /var/cache/man/* >>> to man. >>> >>> Regards, >>> JH Chatenet >>> >>> diff -Naur a/usr/share/debirf/modules/z1_clean-root >>> b/usr/share/debirf/modules/z1_clean-root >>> --- a/usr/share/debirf/modules/z1_clean-root 2014-07-09 >>> 22:28:14.000000000 +0200 >>> +++ b/usr/share/debirf/modules/z1_clean-root 2014-07-10 >>> 16:17:22.000000000 +0200 >>> @@ -28,3 +28,11 @@ >>> >>> # setup dynamically updated /etc/mtab >>> ln -sf /proc/mounts "$DEBIRF_ROOT/etc/mtab" >>> + >>> +# fix owner of /var/cache/man >>> + >>> +if [ -d "$DEBIRF_ROOT/var/cache/man" ]; then >>> + chown -R man:root "$DEBIRF_ROOT/var/cache/man" >>> +fi >>> + >>> + >> >> Why should this workaround be necessary? I am pretty strongly averse to >> any extra chown -R's so i'd like to figure out *why* it's happening >> instead of slapping on a band-aid. Maybe this is a bug in debootstrap's >> fakechroot variant or something? > > JH, how was this image created? The '-r' option wasn't used, was it?
fwiw, i believe i'm able to replicate it with a debirf rescue built by
debirf 0.34 pulled from sid a week or two ago.
That is, i do see:
Unpacking bc (1.06.95-9) ...
Processing triggers for man-db (2.6.7.1-1) ...
fopen: Permission denied
Processing triggers for install-info (5.2.0.dfsg.1-4) ...
Setting up bc (1.06.95-9) ...
0 debirf-rescue:~#
this isn't a critical bug, though, since both "man bc" and "man -k bc"
behave as expected.
--dkg
signature.asc
Description: OpenPGP digital signature
