Colin's fix of #575070 [0] breaks update-inetd's tests (#577299) due to
debconf not running anymore as non-root:
$ LC_ALL=C ./update-inetd --file /tmp/tmpP6yZ5F.modified --verbose --remove 'time2'
debconf: DbDriver "passwords" warning: could not open /var/cache/debconf/passwords.dat: Permission denied
Sebastian mentioned DEBCONF_DEB_REPLACE in the above bug report but I didn't
find a single reference of it in debconf's sources and docs. How's one
supposed to run debconf as non-root? I didn't need to do anything special
up to now.
Please keep the CC in responses.
Cheers,
Serafeim
[0] ``Exclusive read-write lock in file db driver breaks running debconf as
non-root''
--
debtags-organised WNPP bugs: http://members.hellug.gr/serzan/wnpp
--
To UNSUBSCRIBE, email to debian-bugs-...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listm...@lists.debian.org
On Mon, Apr 12, 2010 at 09:13:45PM +0200, Serafeim Zanikolas wrote:
> Colin's fix of #575070 [0] breaks update-inetd's tests (#577299) due to
> debconf not running anymore as non-root:
>
> $ LC_ALL=C ./update-inetd --file /tmp/tmpP6yZ5F.modified --verbose --remove 'time2'
> debconf: DbDriver "passwords" warning: could not open /var/cache/debconf/passwords.dat: Permission denied
The message there is a perfectly normal and largely harmless warning
that you've always got when running debconf as non-root. (Non-root
users can't read the password database, although it's configured in
/etc/debconf.conf. As such, debconf skips it, but warns about it.)
Earlier versions of debconf behaved the same way.
Furthermore, the warning you quote is merely a warning, and is not the
same as the error message in the bug log. Read carefully - "config.dat"
!= "passwords.dat".
Nevertheless, this is still a debconf bug; see below ...
> Sebastian mentioned DEBCONF_DEB_REPLACE in the above bug report but I didn't
> find a single reference of it in debconf's sources and docs.
It's a typo for DEBCONF_DB_REPLACE (but not relevant here).
> How's one supposed to run debconf as non-root? I didn't need to do
> anything special up to now.
Actually, the problem is that debconf now doesn't work under *fakeroot*
without configuration. (But my change wasn't really what broke this; I
merely unbroke things slightly from a previous change by Joey to fix
problems on Solaris.) It would appear that perl's -w operator doesn't
use access() by default to test whether the file is writable, but
instead simply looks at the file modes; "use filetest 'access'" changes
this to be more careful, and is what we need to fix this bug.
I've fixed this in debconf's Subversion repository, and will upload
1.5.31 shortly. I've tested that this unbreaks update-inetd's test
suite.
Regards,
--
Colin Watson [cjwa...@debian.org]
--
debtags-organised WNPP bugs: http://members.hellug.gr/serzan/wnpp
--