[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#249597: texmf: Leaves spurious tempfiles around



Jeroen van Wolffelaar <jeroen@wolffelaar.nl> wrote:

> You're right it doesn't really obviously violate the FHS -- but it does
> clutter the $TMPDIR, which by default isn't auto-cleaned. 

I think it is. /etc/rcS.d/S35mountall.sh does it on my woody and sid
systems, and on the sid system I am sure I never changed anything
there. It's the "bootclean" call right at the end. Why do you think it
is not?

> Note that the
> FHS says that /tmp is for "temporary files", and the files here don't
> fall in the category in my opinion (but it is a subtle).
>
> The purpose of this file is debugging, so in fact it is some kind of
> logfile. IMHO, /var/log is thus the right place for these files, where
> it could and should be arranged that those files are rotated away after
> a suitable time. 

If nothing goes wrong, there is no point in keeping or rotating those
files. This would clutter /var/log IMO, and make people think that
there's anything important to it. And it would make the tetex packages
even more complex, without any need as I see it.

> initex and updmap can then be made to always log there
> -- if you think that's useful in general (if it isn't, why is postinst
> saving it then?).

It is useful, bug if you run initex or updmap by hand, you'll get the
output on the screen - if you like, you can redirect it to a file of
your choice. But for sure it would be a bad idea to display it on the
screen when they are run by a maintainer script. Furthermore, when you
type "fmtutil -all" (it's not initex directly) and get an error, you
know that you can just repeat it, and you'll get the same error messages
and can safe them. Most people who get an error from postinst won't
know which command it was that failed, so they can't easily reproduce
it. 

But we should probably put a note about those tempfiles into our
reportbug template. 

Regards, Frank
-- 
Frank Küster, Biozentrum der Univ. Basel
Abt. Biophysikalische Chemie




Reply to: