Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 25 May 2004 18:26:45 -0400
From:      Allan Fields <bsd@afields.ca>
To:        Colin Percival <colin.percival@wadham.ox.ac.uk>
Cc:        current@freebsd.org
Subject:   Re: anoncvs.FreeBSD.org gone?
Message-ID:  <20040525222645.GA34929@afields.ca>
In-Reply-To: <6.1.0.6.1.20040525091306.0370d840@popserver.sfu.ca>
References:  <40B2FEAF.7050701@alumni.rice.edu> <6.1.0.6.1.20040525091306.0370d840@popserver.sfu.ca>

next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, May 25, 2004 at 09:17:51AM +0100, Colin Percival wrote:
> At 09:07 25/05/2004, Jon Noack wrote:
> > For the last couple days I have been unable to resolve
> > anoncvs.FreeBSD.org.  It appears to have disappeared even from the
> > FreeBSD DNS servers:
>
>   I believe that anoncvs was recently taken offline due to security
> concerns.
> 
> > The handbook still mentions anoncvs.FreeBSD.org -- what is the
> > best anonymous CVS server for someone in the US?
> 
>   This may be a silly question, but why do you need anoncvs?

Hmm.. I'm of the opinion that cvsup doesn't entirely replace the
utility of anoncvs access: sure cvsup is great and I use it to sync
my sources personally, but it doesn't offer the same operational
granularity.  Although I maintain a local mirror for my site and
can use that as my CVSROOT (/home/ncvs), anoncvs access might still
be useful for those who don't want to maintain a local mirror but
still desire to use cvs to access the sources (for example: occasional
use, to get specific version of single binary sources, remote use
via slow dial-up [me a few years back], just to look at logs, etc.)

Generally I think anoncvs helps those w/o CVS accounts better
track the sources using familiar approaches: but maybe everyone
now is using webcvs?


(And, since you did ask...  ;)
Further: I've found anoncvs handy for automatically querying what
exists in the FreeBSD CVS repository from the shell: which is of
interest to those who track the sources periodically.  (Easier than
loading a browser up and webcvs.)  I'd wonder if there shouldn't
be tag entries maintained with associated metadata (but please not
XML/markup) - Either:
  (a) In a Bourne shell parsable file kept up to date by releng
      (w/ all latest tag entries) on HTTP or FTP	-or-
  (b) Using an automated process to extra tags from cvs (which I've
      implemented already)

Obviously (a) is more of a hassle to maintain but offers a cleaner
approach than the sed script I've fashioned to extract tags and form
entries automatically from the anoncvs logs.  I chose not to extract on
what's included on the releng pages as these are HTML, manually
updated(?) and don't always reflect (at the same level of detail)
everything that exists in the repository.

What my approach offers is the ability to automatically sync/update
sources (not implying actual build) and selection of any FreeBSD
branch from the command line or interactively (menu of available
branches).  But it does this currently with the help of (by default)
anoncvs.freebsd.org.

If anyone is interested, I can explain further and/or post the
script (after clean-up).  One of my BSDCan presentations (slides
online http://www.bsdcan.org/2004/papers/) looks at consistently
managing the FreeBSD sources using an automated approach to updating
sources (which I've implemented in a shell script for local use).


.. So if the main anoncvs server proves problematic, it will be
helpful to maintain at least a few mirrors still offering anoncvs
access.

Perhaps also, it is time to experiment further in transitioning to
another repository such as svn and switch over when it becomes
practical to do.  I know previously, work was done in trying to get
FreeBSD CVS and SVN to sync, which proved slightly problematic.
Anyone have any updates?


> Colin Percival

Allan Fields



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040525222645.GA34929>