> This was probably a result of the botched pkg_admin rebuild or
> something.  Usually, things should work automatically.
>
> > as i can see, there was some problem with pkg_install? well, i had the
> > version ~4 months old, so it probably couldn't manage new version. as it
> > seems to me i should firstly update pkg_install if i get new pkgsrc
through
> > cvs, right?
>
> No.  The problem was in pkg_install, that is true.  And you happened to
> have updated pkgsrc when it determined that it needs a newwer
> pkg_install, thus giving you the faulty one.  In general, pkgsrc knows
> when it needs to update the pkgtools and it will automatically do so, so
> there's no need for you to always rebuild/reinstall pkg_install each
> time you update pkgsrc.
>
> Sorry for the problems.
>
> > and now it seems to be ok. why doesn't make update remove the old
version as
> > it SHOULD do (it's obviously the same problem as for the x11-links
package)?
> > is there still something wrong?
>
> I suspect that the fluxbox package was removed by the old pkg_install
> tools, but these changes could not be recorded due to the bug with the
> pkgdb-file.

> Now that you have all things back to normal, you should not
> see these problems.

ok, hope you're right :o) thank you VERY VERY much for your help, you saved
me A LOT of time :o)))

by the way, do you think that my problems with getting new version of the
pkgsrc using cvs can be somehow related to problems i had with the update of
the package? i'm not sure about how cvs knows which directory to update. i
was doing
cvs update -dP
in /usr/pkgsrc so it should update all packages + checkout the new ones. it
didn't work very well - the cvs itself repored
...
cvs server: Updating bsetroot
cvs server: Updating bsetroot/patches
...
but there was no bsetroot directory in pkgsrc/wm/... :o( i had to checkout
the whole pkgsrc tree and replace the old one (i  renamed it do
pkgsrc.old...)...

lubos
_@_:


-
#################################################
Lubos Vrbka
National Centre for Biomolecular Research
Masaryk university, Brno, Czech Republic

shnek@chemi.muni.cz
http://www.chemi.muni.cz/~shnek
tel. +420 541 129 508
#################################################