Control: severity -1 minor
Control: forcemerge -1 813546

On Fri, 2016-06-03 at 21:17:11 +0800, 積丹尼 Dan Jacobson wrote:
> reopen 825918

> There is actually 1954 error groups on my machine.

I don't know really what that means, though.

> A normal user would and should kill the seemingly runaway situation
> before hit computer locks up or disks get filled.
> 
> # sort /tmp/apt_term_log.txt |uniq -c|sort -nr|head
>    1954 dpkg-query: package 'imagemagick' is not installed
>    1954 and dpkg --contents (= dpkg-deb --contents) to list their contents.
>    1954 Use dpkg --info (= dpkg-deb --info) to examine archive files,
>       2 Processing triggers for libc-bin (2.23-0experimental2) ...
>       1 dpkg: warning: unable to delete old directory 
> '/etc/etc/ImageMagick-6': Directory not empty
>       1 dpkg: warning: unable to delete old directory '/etc/etc': Directory 
> not empty
>       1 dpkg: error processing archive 
> /var/cache/apt/archives/imagemagick_8%3a6.9.2.10+dfsg-2_all.deb (--unpack):
>       1 dpkg-query: no packages found matching imagemagick:all
>       1 dpkg-maintscript-helper: error: directory 
> '/usr/share/doc/imagemagick' contains files not owned by package 
> imagemagick:all, cannot switch to symlink
>       1 Unpacking libmagickwand-6.q16-2:i386 (8:6.9.2.10+dfsg-2+b1) over 
> (8:6.9.2.10+dfsg-1) ...
>

I guess you have tried to install this package many many times? Also
you are having this problem when installing a package not even from
unstable but from experimental! This needs fixing in the affected
package.

Again there is no infinite loop in dpkg, the relevant script just takes
a list of files and scans for their ownership, this just takes some
time because I'm assuming it goes over all documentation from
imagemagick-doc. Perhaps the only possible problem here is that the
error messages are not clear on what's going on, but because that should
not happen in the first place and the problem here is really in
imagemagick, this is at most minor in dpkg, and I could perhaps improve
the error reporting, I'd rather see such errors so that they can be fixed
instead of hiding them. So I'm probably just going to close this bug
again in a bit.

Regards,
Guillem

Reply via email to