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

Re: What's the best package manager for single-package upgrades?#



> Ah, that would explain your confusion. 'apt-get upgrade' isn't what you
> want, since as documented in the apt-get(8) man page it will not install
> new packages. In particular, if you attempt to use 'apt-get upgrade' to
> upgrade from stable to testing, it will refuse to upgrade libc6 because
> of that package's new dependency on libdb1-compat, and therefore
> virtually nothing else will be upgraded because it almost all depends on
> the new libc6.
 
Actually, it does attempt that when I prefer 'unstable' .. and it fails.
I had to manually back that stuff out.

> Don't use 'apt-get upgrade' to upgrade from one version of the
> distribution to the next. That said, it should have told you that some
> big number of packages were being held back.
 
Nope.  "No updates are available" or whatever.

> > Perhaps my product selections are biased: I really could care less about
> > the latest and greatest desktop.  They are pretty.  But a browser that
> > actually works is required to do my job, for example.
> 
> Testing has a perfectly usable version of mozilla-firebird, which I'd
> argue is a much better browser than plain mozilla.

I might personally agree, but there are no production users of firebird.
So we have to keep it around in a few places at least.

> > Updates to the wireless drivers to improve device support would be
> > useful.
> 
> Kernel updates go in pretty quickly, as a rule. wireless-tools is up to
> date in testing, and linux-wlan-ng is only a fraction behind unstable.
 
Why isn't it showing me these?

> > Stuff that has been safe and stable within Sid for over a year now
> > (according to the package pages) still isn't appearing in testing.
> 
> Examples, please? I'd be happy to look at them and see what I can do; I
> can certainly explain what problems are involved.
 
Perhaps related to above?  Am I doing something wrong that I'm not seeing
this stuff?

> > In short, it appears that if one actually wants to use Debian as a
> > desktop, one has no choice but to throw the debian guidelines out the
> > window and run with unstable.
> 
> I actually use Debian testing as a desktop, eight hours a day, five days
> a week. It works great.
> 
> > This means you lose commonality with any server 'stable' systems you
> > might need to run.
> 
> As far as commonality goes (although I don't quite understand what you
> mean here), you should regard testing as closer to unstable in terms of
> versions of software than to stable, because for the most part it is,
> particularly in recent months.

The general idea being that you could have an internal policy that no
'unstable' things are deployed on servers.  I wouldn't mind running
unstable on personal desktops, but if they diverge so far that there is a
loss of commonality...

-- 
Joe Rhett                                                      Chief Geek
JRhett@Isite.Net                                      Isite Services, Inc.



Reply to: