BIND

BIND

[Andrew Carroll]

> ------------------------------
> From: "Warren W. Gay VE3WWG" <warren@ve3wwg.tk>
> Subject: Re: No call for Ada
[snip]
> That is why I have said in the past that someone needs to rewrite
> BIND (DNS) in Ada. I would sleep better at night with an Ada version
> of it exposed to the net than the C versions we use.

I am interested in this.  I don't really know if I am qualified, 
but am willing to give it my best.


Andrew Carroll
Carroll-Tech
720-273-6814
andrew@carroll-tech.net

Re: BIND

[Ludovic Brenta]

"Andrew Carroll" writes:
> > ------------------------------
> > From: "Warren W. Gay VE3WWG" > Subject: Re: No call for Ada
> [snip]
> > That is why I have said in the past that someone needs to rewrite
> > BIND (DNS) in Ada. I would sleep better at night with an Ada version
> > of it exposed to the net than the C versions we use.
> 
> I am interested in this.  I don't really know if I am qualified, 
> but am willing to give it my best.

The same idea had crossed my mind too, but I'd forgotten.  I will
help.  I can package it for Debian, do bug tracking, documentation,
code reviews, automated builds, but no design or coding for lack of
time and competence in network programming and security.  Andrew, will
you take the lead and start a project on SourceForge?

Who else will help?

-- 
Ludovic Brenta.

Re: BIND

[Lutz Donnerhacke]

* Ludovic Brenta wrote:
> Who else will help?

I created a whole framework for UDP and TCP handling.

The main problem with DNS is, that the specification does not really exists.
It is mainly specified by the BIND source code. The main part is parsing old
BIND configuration files. Futhermore a non crashing implementation is not
secure. The protocol has several weaknesses itself an can't be implemented
the na�ve way.

Before you try to implement DNS, write a MTA, it's much easier.
Better: Write an FTP server.

Do not write an Telnet server, the Telenet protocol is too complex.

But if you really want to remove a common security hole, rewrite an ASN.1
parser. ftp://ftp.iks-jena.de/pub/mitarb/lutz/standards/itu-t/x/ contains
X.208 and X.209. The available library (libsnmp) is implemented in the C
preprocessor language and therefore nor linkable. So you are really free to
implement from scratch.

Good luck.

BIND

[Andrew Carroll]

> ------------------------------
> From: Ludovic Brenta <ludovic.brenta@insalien.org>
> Subject: Re: BIND
>
> "Andrew Carroll" writes:
> > > ------------------------------
> > > From: "Warren W. Gay VE3WWG" > Subject: Re: No call for Ada
> > [snip]
> > > That is why I have said in the past that someone needs to rewrite
> > > BIND (DNS) in Ada. I would sleep better at night with an Ada version
> > > of it exposed to the net than the C versions we use.
> >
> > I am interested in this.  I don't really know if I am qualified,
> > but am willing to give it my best.
>
> The same idea had crossed my mind too, but I'd forgotten.  I will
> help.  I can package it for Debian, do bug tracking, documentation,
> code reviews, automated builds, but no design or coding for lack of
> time and competence in network programming and security.  Andrew, will
> you take the lead and start a project on SourceForge?

Alright, I can do that.  It won't be this weekend though.  We should
probably
check into what Lutz Donnerhacke had to say (below) just to give us an idea
of how feasible this would be.  I'm not real familiar with SourceForge.  Do
they
provide a discussion board and mailing list?  If not, I might be able to get
one
through the college I graduated from.


> ----------------------------------------------------------------------
> From: Lutz Donnerhacke <lutz@iks-jena.de>
> Subject: Re: BIND
>
> * Ludovic Brenta wrote:
> > Who else will help?
>
> I created a whole framework for UDP and TCP handling.

Does this mean you would like to help?


> The main problem with DNS is, that the specification does not really
exists.
> It is mainly specified by the BIND source code. The main part is parsing
old
> BIND configuration files. Futhermore a non crashing implementation is not
> secure. The protocol has several weaknesses itself an can't be implemented
> the na�ve way.

If what you say is true, then this is something that needs to be looked
into.
RFC's, books, papers, links?  Anyone?


Andrew Carroll
Carroll-Tech
720-273-6814
andrew@carroll-tech.net

Re: BIND

[Ludovic Brenta]

Andrew Carroll writes:
[Reimplementing BIND in Ada, creating a project on SourceForge]
> Alright, I can do that.  It won't be this weekend though.  We should
> probably check into what Lutz Donnerhacke had to say (below) just to
> give us an idea of how feasible this would be.  I'm not real
> familiar with SourceForge.  Do they provide a discussion board and
> mailing list?

Yes and yes.  The administrator of the project can request mailing
lists to be created.

And my id on SourceForge is lbrenta, in case you want to add me to the
list of developers for your project.

Lutz Donnerhacke said:
> > The main problem with DNS is, that the specification does not
> > really exists.  It is mainly specified by the BIND source
> > code. The main part is parsing old BIND configuration
> > files. Futhermore a non crashing implementation is not secure. The
> > protocol has several weaknesses itself an can't be implemented the
> > naïve way.
> 
> If what you say is true, then this is something that needs to be
> looked into.  RFC's, books, papers, links?  Anyone?

http://www.dns.net/dnsrd/ looks like a good starting point.

Lutz, do you think your framework would help implement a DNS in Ada?
If so, would you contribute it to the project?

I think that the first step if for all interested people to learn as
much as possible about the current state of the art.  The mailing list
provided for each project by SourceForge would help.

-- 
Ludovic Brenta.

BIND

[Andrew Carroll]

Okay, project is created.
I put it on sourceforge as "Ada Bind".  Unix name adabind.
I just put it up today so it may be a couple of days before we can use it.

I found some rfc's on DNS at www.dns.net/dnsrd
So I went ahead and posted the project.

Andrew Carroll
Carroll-Tech
720-273-6814
andrew@carroll-tech.net

Re: BIND

[those who know me have no need of my name]

in comp.lang.ada i read:

>The main problem with DNS is, that the specification does not really exists.

surely the rfc's haven't disappeared.

<http://www.rfc-editor.org/cgi-bin/rfcsearch.pl?searchwords=dns>

>It is mainly specified by the BIND source code. 

bind is the canonical reference implementation.  it is not intended that it
be the specification.

>The main part is parsing old BIND configuration files.

this is not terribly difficult -- each style (4.x vs 8/9) is well documented.

-- 
a signature

Re: BIND

[Jeffrey Carter]

Andrew Carroll wrote:

> Okay, project is created.
> I put it on sourceforge as "Ada Bind".  Unix name adabind.
> I just put it up today so it may be a couple of days before we can use it.

I know little about internet software, so it would be a learning 
experience for me. I would need well specified pieces of the 
application, but if I can contribute, let me know.

-- 
Jeff Carter
"All citizens will be required to change their underwear
every half hour. Underwear will be worn on the outside,
so we can check."
Bananas
29

Re: BIND

[Jeffrey Carter]

Jeffrey Carter wrote:

> I know little about internet software, so it would be a learning 
> experience for me. I would need well specified pieces of the 
> application, but if I can contribute, let me know.

And because I use a fake address here, let me give my real one:

jrcarter COMMERCIAL-AT-SIGN acm [PERIOD | FULL STOP] org

-- 
Jeff Carter
"All citizens will be required to change their underwear
every half hour. Underwear will be worn on the outside,
so we can check."
Bananas
29

Re: BIND

[Lutz Donnerhacke]

* Ludovic Brenta wrote:
> Lutz, do you think your framework would help implement a DNS in Ada?
> If so, would you contribute it to the project?

Yep.

Re: BIND

[Lutz Donnerhacke]

* those who know me have no need of my name wrote:
> in comp.lang.ada i read:
>>The main problem with DNS is, that the specification does not really exists.
>
> surely the rfc's haven't disappeared.

There is a difference between RfC and real world. In the case of DNS the
difference is huge. No only cnsidering security issues.

>>It is mainly specified by the BIND source code. 
>
> bind is the canonical reference implementation.  it is not intended that it
> be the specification.

Bind DNS handling differs from the spec. Nobody is interested in an
incompatible version. DJB did this mistake and failed miserably.

>>The main part is parsing old BIND configuration files.
>
> this is not terribly difficult -- each style (4.x vs 8/9) is well documented.

Existing documentation does not imply easy to implement.

Re: BIND

[Warren W. Gay VE3WWG]

Lutz Donnerhacke wrote:
> * Ludovic Brenta wrote:
>>Who else will help?
> I created a whole framework for UDP and TCP handling.
> 
> The main problem with DNS is, that the specification does not really exists.
> It is mainly specified by the BIND source code. The main part is parsing old
> BIND configuration files.  ...

I don't know if BIND's configuration files are part of the
standard or not, but if you want to use that format (for
familiarity at least), then the use of ayacc and Simon
Wright's Booch components can slice through that with
ease.

So if you need advice about that particular phase of
the project, I can help.  I just finished using
those tools together in a current pet project.

GO AdaBind Go! ;-)
-- 
Warren W. Gay VE3WWG
http://ve3wwg.tk

Re: BIND

[Warren W. Gay VE3WWG]

Andrew Carroll wrote:

> Okay, project is created.
> I put it on sourceforge as "Ada Bind".  Unix name adabind.
> I just put it up today so it may be a couple of days before we can use it.
> 
> I found some rfc's on DNS at www.dns.net/dnsrd
> So I went ahead and posted the project.
> 
> Andrew Carroll
> Carroll-Tech
> 720-273-6814
> andrew@carroll-tech.net

Bravo!
-- 
Warren W. Gay VE3WWG
http://ve3wwg.tk

BIND

[Andrew Carroll]

Ok,
Finally,
The Ada BIND SourceForge project is 'live'.

I know there is supposed to be a mailing list with the project but I am new
to sourceforge and I haven't figured out what the mailing list address is
yet.
So, if you want to be added/participate, post your sourceforge unix name to
this list.

There are some project options I haven't set yet so if you notice something
that needs to be changed let me know.


Andrew Carroll
Carroll-Tech
720-273-6814
andrew@carroll-tech.net

Re: BIND

[Craig Carey]

On Wed, 21 Apr 2004 14:34:46 -0600, "Andrew Carroll"
<andrew@carroll-tech.net> wrote:

>Ok,
>Finally,
>The Ada BIND SourceForge project is 'live'.
...

I assume that 'Bind' is a project that is too difficult. I am sure that
comp.lang.ada is prepared to say that bind project plan shuld be backed
out of.

...
>There are some project options I haven't set yet so if you notice something
>that needs to be changed let me know.

Also there is the Gdb debugger: Mr Paul Hilfinger notes that ACT has
an intent to get it running.
  http://sources.redhat.com/ml/gdb/2004-04/msg00032.html

| From: "Paul N. Hilfinger" <hilfingr at otisco dot mckusick dot com>
| Cc: gdb at sources dot redhat dot com
| Date: Mon, 5 Apr 2004 13:58:10 -0700
| Subject: Re: Building an Ada aware gdb 
...
| > I see that gdb from CVS has some exciting (for me) files
| > in it: ada-exp.y, ada-lang.h, ada-tasks.c, ada-valprint.c,
| > ada-lang.c, ada-lex.l, ada-typeprint.c.  I would like to build
| > gdb with Ada support... but how?  The only Ada related
| > configure option seems to be --enable-libada, and that didn't
| > give me an Ada aware gdb.  Does gdb support Ada now?
| > If so, how to build it...
| 
| Not yet, but we're finally getting close.  At the moment, we are
| resolving some issues with hooks we'd like to put into non-Ada sources.
| After that, it's mostly a matter of linking in the Ada modules and testing
| for regressions on every platform we can get our hands on. 
| 
| A couple of warnings: 
| 
| 1. The current CVS ada-* sources have not been synced with ACT's
| development versions.  This fact has nipped me a couple of times now
| (including your question), so I should probably update them just so
| folks can see them.
...
| On the other hand, we should have expression evaluation working pretty
| much. 
| 
| Paul Hilfinger
| ACT

It does not compile (no change there).
Also the libre.act-europe.fr version did not compile in Cygwin lat year.
It seems questionable: nearly every comp.lang.ada user being unable to
freely debug the debugger, which is due some quite major enhancements,
I guess.

Does GDB stop on properly set breakpoints after it attached to an Ada
program, e.g. if running in an Apache CGI environment (or as a DLL) (?).

Also MinGW.org GNAT Ada does not run out of the box. Environment
variables need to be set. A small project would be to get a text file
into the distribution file.

   GCC_ROOT=l:/mingw   # modify as needed. Adapted from a Linux script
   GNAT_ROOT=$GCC_ROOT
   BINUTILS_ROOT=$GCC_ROOT
   LD_LIBRARY_PATH=$GCC_ROOT/lib:$LD_LIBRARY_PATH
   PATH=$GCC_ROOT/bin:$PATH
   export GCC_ROOT GNAT_ROOT BINUTILS_ROOT PATH LD_LIBRARY_PATH
      # call gnatmake.exe

MinGW Ada is useful in lacking bugs that GNAT 3.15p NT has.

I could suggest a small project that mixed GNAT with Visual C++. But

I guess Ada would be failed if mixed with Visual C++ and then it took
at least an extra 35 secs to cycle through an attach and debug operation.
Particularly if not actually stopping on break points.

A project can be to list to comp.lang.ada, the ways in which attempts to
 compile Gdb, failed.

--  Craig Carey
  Mailing lists: http://www.ijs.co.nz/ada_95.htm

Re: BIND

[Alexei Polkhanov]

On Wed, 21 Apr 2004 14:34:46 -0600, Andrew Carroll wrote:

> Ok,
> Finally,
> The Ada BIND SourceForge project is 'live'.

Ada BIND ? What is it ? Could not find it on SourceForge, is there any
chance you can provide a link to project's home page ?
 
---
Alexei Polkhanov
Sr. Consultant/Software Systems Analyst
Tel: (604) 719-2515
E-mail: usenet@monteaureus.com
http://www.monteaureus.com/

Re: BIND

[Alexei Polkhanov]

On Wed, 21 Apr 2004 14:34:46 -0600, Andrew Carroll wrote:

> Ok,
> Finally,
> The Ada BIND SourceForge project is 'live'.
> 
> I know there is supposed to be a mailing list with the project but I am new
> to sourceforge and I haven't figured out what the mailing list address is
> yet.
> So, if you want to be added/participate, post your sourceforge unix name to
> this list.

You guys not serious, are you ? You are going t reimplement BIND in Ada ?
First I worry what is the point ? Just _write it in Ada_, and that is it ?

How do you JUSTIFY THE NEED for it ?

You don't even bother to look what kind of problem BIND has right now, and
you definitely underestimated the scope. What about things like DNSSEC -
huge undertaking with puplic key encryption protocols, have performance
problems in BIND, TSIG, national languages support (like chinese unicode DNS
names).

Some zones can grow up to several millions of records - how you going to
cope with that ? What about Incremental Zone transfer protocol ? New BIND
definitely need advanced support for RDBMS like ORACLE, MySQL etc and
should include "pluggable" internal in-memory database by type of Berkeley
DB. 

So you signing for a minimum 60,000 LOC project, without doing any
research on subject. Noone need another BIND which does not solve problems of
existing BIND and that is not only couple of buffer overruns. 

Cheers,
---
Alexei Polkhanov
Sr. Consultant/Software Systems Analyst
Tel: (604) 719-2515
E-mail: usenet@monteaureus.com
http://www.monteaureus.com/

Re: BIND

[Lutz Donnerhacke]

* Alexei Polkhanov wrote:
> So you signing for a minimum 60,000 LOC project, without doing any
> research on subject. Noone need another BIND which does not solve
> problems of existing BIND and that is not only couple of buffer overruns.

Ack. That's why I do not participate.

Re: BIND

[Ludovic Brenta]

"Alexei Polkhanov" writes:
> You guys not serious, are you ? You are going t reimplement BIND in
> Ada ?  First I worry what is the point ? Just _write it in Ada_, and
> that is it ?
> 
> How do you JUSTIFY THE NEED for it ?

Bugs in BIND that nobody cares to fix because of design problems or
source code that is too difficult to read and debug.  Common knowledge
that BIND is so insecure that nobody but the most inexperienced
sysadmins will run it outside a chroot jail.  Concerns with the
long-term security threats posed by BIND's inherent problems, as well
as the monoculture associated with BIND.

Heck, I don't even _need_ a reason why I should write a program, if
it's free software and if I do it on my spare time.  My reason is:
Just to have fun, okay?

> You don't even bother to look what kind of problem BIND has right
> now, and you definitely underestimated the scope.

You do not know our estimation, so you cannot accuse us or anyone of
underestimating the scope.  Apparently, do don't even bother to
consider the possibility there may be experienced software engineers
on this group.

> What about things like DNSSEC - huge undertaking with puplic key
> encryption protocols, have performance problems in BIND, TSIG,
> national languages support (like chinese unicode DNS names).

These things will naturally have to be designed into AdaBind.  Are you
trying to imply it cannot be done?  Just because _you_ cannot do it
does not imply _nobody_ can.

> Some zones can grow up to several millions of records - how you
> going to cope with that ? What about Incremental Zone transfer
> protocol ? New BIND definitely need advanced support for RDBMS like
> ORACLE, MySQL etc and should include "pluggable" internal in-memory
> database by type of Berkeley DB.

Yes indeed.  So?

With GNADE and APQ, it is already possible to write Ada programs that
interface with databases of arbitrary complexity, and even make the
database a pluggable module, leaving the choice of database to the end
user.

> So you signing for a minimum 60,000 LOC project, without doing any
> research on subject.

You do not know if any of us has done any research, so you cannot
accuse anyone of not making any.

60 kSLOC is not large, and it can be done by a single person (BIND was
indeed written mostly by a single person) or by a small team (2-3
people).  I myself have written a 120 kSLOC program with a team of 2-3
people.  Or look at PolyORB, which also was made by a small team and
is also 120 kSLOC.

> Noone need another BIND which does not solve problems of existing
> BIND and that is not only couple of buffer overruns.

Indeed no.  We need an AdaBind that does solve the problems of BIND,
and is more maintainable.  The proponents of AdaBIND (myself included)
think the situation of BIND is so hopeless that the best solution to
its problems is to rewrite it - only better.  And Ada will help.

Your post is not helping anyone, not even yourself.  Please tell us
what kind of contribution you were trying to make.  I have little
respect for destructive thinking, but I admire people who try to make
things happen.

-- 
Ludovic Brenta.

Re: BIND

[Florian Weimer]

"Alexei Polkhanov" <usenet@monteaureus.com> writes:

> Some zones can grow up to several millions of records - how you going to
> cope with that ? What about Incremental Zone transfer protocol ?

If you go for the low hanging fruit and write an authoritative-only
name server that can serve zones of medium size, it's not too
difficult.  However, getting all the details correct is another story,
and a full resolver is much more difficult to write.

-- 
Current mail filters: many dial-up/DSL/cable modem hosts, and the
following domains: atlas.cz, bigpond.com, di-ve.com, netscape.net,
postino.it, tiscali.co.uk, tiscali.cz, tiscali.it, voila.fr.

Re: BIND

[Florian Weimer]

Ludovic Brenta <ludovic.brenta@insalien.org> writes:

> Bugs in BIND that nobody cares to fix because of design problems or
> source code that is too difficult to read and debug.

The BIND 9 source code isn't too bad, actaully.

> Common knowledge that BIND is so insecure that nobody but the most
> inexperienced sysadmins will run it outside a chroot jail.

BIND 9 is quite okay.  Keep in mind that so far, no buffer overflow
bug has been discovered in the BIND 9 proper.  Compare that to the
GNAT run-time library. 8-/

> Concerns with the long-term security threats posed by BIND's
> inherent problems, as well as the monoculture associated with BIND.

There isn't quite a monoculture, BIND 8 and 9 are very different
beasts.  However, you really shouldn't run BIND 8.

You can use RIPE nsd for authoritative servers, if you want.  It's
also much smaller and aims at bug-for-bug compatibility with BIND 8.

BIND 9 on full resolvers is very hard to replace with anything else,
though.

> Heck, I don't even _need_ a reason why I should write a program, if
> it's free software and if I do it on my spare time.  My reason is:
> Just to have fun, okay?

Writing a BIND replacement is not fun.  Maybe writing a DNS server,
but certainly not a BIND replacement.

> 60 kSLOC is not large, and it can be done by a single person (BIND was
> indeed written mostly by a single person)

Which one of the BINDs?  BIND 9 had a team of several developers
working full time on it, IIRC.

I'd rather see a industry-strength Ada implementation of TLS and
X.509.  Right now, the OpenSSL monoculture worries me far more than
BIND.

-- 
Current mail filters: many dial-up/DSL/cable modem hosts, and the
following domains: atlas.cz, bigpond.com, di-ve.com, netscape.net,
postino.it, tiscali.co.uk, tiscali.cz, tiscali.it, voila.fr.

Re: BIND

[Wes Groleau]

Alexei Polkhanov wrote:
> You don't even bother to look what kind of problem BIND has right now, and
> you definitely underestimated the scope. What about things like DNSSEC -
> huge undertaking with puplic key encryption protocols, have performance
> problems in BIND, TSIG, national languages support (like chinese unicode DNS
> names).

It would be a heck of a lot easier to address
these things if it was in Ada.

-- 
Wes Groleau
http://freepages.rootsweb.com/~wgroleau/Wes

BIND

[Andrew Carroll]

Peeps,

Well, I see the Ada BIND project is already met with some 
resistance.  The only thing I can see that would bring about such 
a large resistance to this project is that there is already work 
being done on _an_ "Ada BIND" that is not Open Source.

Interesting...

You know, I used to be a big antagonist of what Microsoft 
does, like their move in the XML area for their office products. 
Calling or writing them all day long and telling them their making 
a big mistake doesn't seem to stop them.  Come to think of it, 
maybe that's why Microsoft does "that" kind of stuff every now 
and then.  Just to "shake off" those companies that have come 
up with a product that is dependent on a Microsoft API or 
interface or "something".  The product that those companies are 
making may actually hurt the bottom line of Microsoft; if they 
didn't change things.  Is it possible that the Ada BIND project 
could hurt someone's bottom line?  

Hmmm, I now have to wonder...

I also have to laugh.  The Ada BIND project, at this point, is 
nothing more than an account on sourceforge.  There are no 
documents, no designs and no source code.  There are a bunch 
of RFCs and possibly some C code to dig through.  No work 
has been done, no evaluation of scope has been done, no 
feasibility assessment has been done, no requirements gathered, 
no stakeholders identified and no vision document drawn up.  
Yet many of you are reacting to the Ada BIND proposal like 
we just squashed your sweet little wiener dog with our big Ford 
1-ton 4x4 truck.  It is such an over reaction that it is funny.

Maybe you should take a deep breath, organize your thoughts 
and then present them to us as issues to resolve instead of 
issues to kill the project?  Maybe you feel left out?  Maybe you 
feel we stepped on your toes?  I don't know what your thinking 
and maybe this is an opportunity to voice your concerns in such 
a way that will not only get your voice recognized but also help 
with the project.

What I am saying is that I'm not convinced that the project 
should be killed.  Give me some proof and then I might change 
my mind.  Well, in reality we should continue the discussion on 
the project mailing list.

adabind-public@lists.sourceforge.net
https://sourceforge.net/projects/adabind/

P.S.
That's not an invitation to send me emails directly.  Please use 
the project mailing list.

Andrew Carroll
Carroll-Tech
720-273-6814
andrew@carroll-tech.net

Re: BIND

[Björn Persson]

Alexei Polkhanov wrote:

> [...] national languages support (like chinese unicode DNS
> names).

As I understand it, Unicode DNS names must be transformed to ASCII by 
the clients, so the server doesn't even need to know that anything but 
ASCII exists.

-- 
Björn Persson

jor ers @sv ge.
b n_p son eri nu

Re: BIND

[chris]

Andrew Carroll wrote:

> I also have to laugh.  The Ada BIND project, at this point, is 
> nothing more than an account on sourceforge.  There are no 
> documents, no designs and no source code.  There are a bunch 
> of RFCs and possibly some C code to dig through.  No work 
> has been done, no evaluation of scope has been done, no 
> feasibility assessment has been done, no requirements gathered, 
> no stakeholders identified and no vision document drawn up.  
> Yet many of you are reacting to the Ada BIND proposal like 
> we just squashed your sweet little wiener dog with our big Ford 
> 1-ton 4x4 truck.  It is such an over reaction that it is funny.

I think the problem is that

"The Ada BIND project, at this point, is nothing more than an account on 
sourceforge.  There are no documents, no designs and no source code. 
There are a bunch of RFCs and possibly some C code to dig through.  No 
work has been done, no evaluation of scope has been done, no feasibility 
assessment has been done, no requirements gathered, no stakeholders 
identified and no vision document drawn up."

?

Re: BIND

[Warren W. Gay VE3WWG]

Andrew Carroll wrote:

> I also have to laugh.  The Ada BIND project, at this point, is 
> nothing more than an account on sourceforge.  There are no 
> documents, no designs and no source code.  There are a bunch 
> of RFCs and possibly some C code to dig through.  No work 
> has been done, no evaluation of scope has been done, no 
> feasibility assessment has been done, no requirements gathered, 
> no stakeholders identified and no vision document drawn up.  
> Yet many of you are reacting to the Ada BIND proposal like 
> we just squashed your sweet little wiener dog with our big Ford 
> 1-ton 4x4 truck.  It is such an over reaction that it is funny.

I fully agree.

Certainly those that propose to work on this project are in the
best position to judge their own interest, goals and resources.

There is nothing wrong with competition either. I for one, would
very much welcome some additional choice in this area, and
would certainly welcome any Ada implementation, if it delivers
on the important goals. A reference BIND implementation is fine,
but a reference implementation need not be the production version.

Let them have a go at it. Allow them time to work out their
vision and research. This discussion seems to have attracted
the seagulls, which is unfortunate. (A seagull flies into the
room, and craps over everything)

> What I am saying is that I'm not convinced that the project 
> should be killed.  Give me some proof and then I might change 
> my mind.  Well, in reality we should continue the discussion on 
> the project mailing list.
> 
> adabind-public@lists.sourceforge.net
> https://sourceforge.net/projects/adabind/

I think instead, the discussion should be centered on how it
can succeed. It is always easier to criticize than to create.

-- 
Warren W. Gay VE3WWG
http://ve3wwg.tk

Re: BIND

[Lutz Donnerhacke]

* Warren W. Gay VE3WWG wrote:
> I think instead, the discussion should be centered on how it
> can succeed. It is always easier to criticize than to create.

 First: Downsize the goals.
Second: Implement a simple DNS proxy for NS, SOA, A, MX, CNAME and PTR.
 Third: Implement authoritive zones.
Fourth: Extent the software to other types namely to IPv6.
 Fifth: Rewrite the whole package from scratch to implement DNSSEC.
 Sixth: Restart this process for the directory protocol used now.

I'd contribute:
  Stream_Text_IO (Basic Text_IO over Ada.Streams.Root_Stream_Type'Class)
  Skip_Stream and Memory_Stream (indexable Stream_Type over Storage_Elements)
  UDP and TCP server handling packages
  Ringbuffer for Queuing

But I can't contribute it to sourcefourge, because I have enought trouble
developing at work.

Re: BIND

[Alexei Polkhanov]

On Mon, 26 Apr 2004 19:36:36 +0000, Lutz Donnerhacke wrote:

> * Warren W. Gay VE3WWG wrote:
>> I think instead, the discussion should be centered on how it
>> can succeed. It is always easier to criticize than to create.
Agree! 

50% of success is in good plan!

Requirements, good design - only rest is coding and testing. 
I am surprised why this problem was approached from another 
end - choice of programming language ??? :)

> 
>  First: Downsize the goals.
I would merge DHCP and DNS together in a single product. There is several
reasons for that.

> Second: Implement a simple DNS proxy for NS, SOA, A, MX, CNAME and PTR.
Sounds almost trivial.

>  Third: Implement authoritive zones.
I suppose support for RDBMS and HUGE ZONES is what we need here. If you 
have 2,000,000 record zone it takes 20 min to reload it. BIND has an 
"rndc" tool which reloads zone files without restarting whole 
thing. When you have such big zone you expect many changes to it too. 
This problem needs some speacial approach.

> Fourth: Extent the software to other types namely to IPv6.
IPv6 is would be a small implementation detail, don't make it look like a
major undertaking, AAAA records plus some IPv6/IPv4 migration stuff. 

National languages support, special records and extentions which need to 
be taken care of. Latest I've heard is an extention for IP Phones. 

"www.domain.com" notation is very technical, network - centric, so some 
people trying to comeup with more human - readable names, and that going
to be in next versions of BIND. Why I cannot type "Congress Library" but I
have to go and type "dsgdsgf.dsfgdsfg.hfdhdf.gov" ? So most of us ill go
to Google because they cannot remember domain name and then will type it
in browser - why can't we skip this step ?

>  Fifth: Rewrite the whole package from scratch to implement DNSSEC.
Yeah! Here is a Huge piece of code which is not the best part of BIND and
sounds like we already have some issues like encryption libraries. Nothing
I know of except thin bindings to OpenSSL. 

Also there is TSIG, and SPLIT DNS (when you have all zone records divided
into subsets with different access rules or permissions). Right now this
feature "implemented" by config files hackery and having multiple
instances of BIND running on different machines sharing only parts of zone
records.

>  Sixth: Restart this process for the directory protocol used now.
> 
> I'd contribute:
>   Stream_Text_IO (Basic Text_IO over Ada.Streams.Root_Stream_Type'Class)
>   Skip_Stream and Memory_Stream (indexable Stream_Type over Storage_Elements)
>   UDP and TCP server handling packages

Would be nice to have this thing working on MOST unix platforms, so sounds
like Florist can be used here...

>   Ringbuffer for Queuing
> 
> But I can't contribute it to sourcefourge, because I have enought trouble
> developing at work.

Means it also nice to find a _sponsor_ at least for some UNIX hardware and
some network for testbed...

---
Alexei Polkhanov
Sr. Consultant/Software Systems Analyst
Tel: (604) 719-2515
E-mail: usenet@monteaureus.com
http://www.monteaureus.com/

Re: BIND

[Georg Bauhaus]

Alexei Polkhanov <usenet@monteaureus.com> wrote:

: Why I cannot type "Congress Library" but I
: have to go and type "dsgdsgf.dsfgdsfg.hfdhdf.gov" ?

<ot>It's because few *hyper*texts, home made or ready made, contain
these valuable secret things called cross-portal *links*, hidden
behind some useful humal readable word(s). A non-technical problem ;-)
If user controlled navigation weren't restricted to one place on
a typical computer screen...  OTOH, Google etc are solving this
problem.</>


-- Georg

Re: BIND

[Lutz Donnerhacke]

* Alexei Polkhanov wrote:
> On Mon, 26 Apr 2004 19:36:36 +0000, Lutz Donnerhacke wrote:
>>  First: Downsize the goals.
> I would merge DHCP and DNS together in a single product. There is several
> reasons for that.

NO! It's dumb, dumb, and dumb again.

>> Second: Implement a simple DNS proxy for NS, SOA, A, MX, CNAME and PTR.
> Sounds almost trivial.

Welcome to the real world.

>>  Third: Implement authoritive zones.
> I suppose support for RDBMS and HUGE ZONES is what we need here. If you
> have 2,000,000 record zone it takes 20 min to reload it. BIND has an
> "rndc" tool which reloads zone files without restarting whole
> thing. When you have such big zone you expect many changes to it too.

rndc is a frontend to Bind, not a solution!

> This problem needs some speacial approach.

Yep, that's why leave it out as long as possible.

>> Fourth: Extent the software to other types namely to IPv6.
> IPv6 is would be a small implementation detail, don't make it look like a
> major undertaking, AAAA records plus some IPv6/IPv4 migration stuff.

Wrong! There is no working IPv6 implementation of GNAT.Sockets.
There are years of failures in understanding and setting up IPv6.
You are on the best way to redo all this mistakes again.

> National languages support, special records and extentions which need to
> be taken care of. Latest I've heard is an extention for IP Phones.

GNA! Stop implementing anything! Please READ a specification first!
You do not understand what you are talking about. Please drop this project!

> "www.domain.com" notation is very technical, network - centric, so some
> people trying to comeup with more human - readable names, and that going
> to be in next versions of BIND. Why I cannot type "Congress Library" but I
> have to go and type "dsgdsgf.dsfgdsfg.hfdhdf.gov" ? So most of us ill go
> to Google because they cannot remember domain name and then will type it
> in browser - why can't we skip this step ?

You are looking f�r HDDB or other distributed hierarchical databases, but
this is a major step away from the Bind project. (Please note, that I
suggest as the very first step: "Downsize the goals!")

>>  Fifth: Rewrite the whole package from scratch to implement DNSSEC.
> Yeah! Here is a Huge piece of code which is not the best part of BIND and
> sounds like we already have some issues like encryption libraries.
> Nothing I know of except thin bindings to OpenSSL.

I can contribute even Hash, Symmetric, and Asymmetric Packages using generic
class wide programming, but I doubt it will help anything.

> Also there is TSIG, and SPLIT DNS (when you have all zone records divided
> into subsets with different access rules or permissions). Right now this
> feature "implemented" by config files hackery and having multiple
> instances of BIND running on different machines sharing only parts of zone
> records.

Do not touch those areas. They are braindead klugdes. Do not implement such
things. If you come to this requirement in 2007, you might consider views.

>> I'd contribute:
>>   Stream_Text_IO (Basic Text_IO over Ada.Streams.Root_Stream_Type'Class)
>>   Skip_Stream and Memory_Stream (indexable Stream_Type over Storage_Elements)
>>   UDP and TCP server handling packages
>
> Would be nice to have this thing working on MOST unix platforms, so sounds
> like Florist can be used here...

So please stick to Florist.

> Means it also nice to find a _sponsor_ at least for some UNIX hardware and
> some network for testbed...

GNA? From what system do post? Ada is for portable programming! You do not
need any other hardware than a low end development system.

(My programming, testing, and debugging enviroment is a Pentium 133Mhz with
 32MB of RAM for Unix/Linux development.)

Re: BIND

[David Starner]

On Mon, 26 Apr 2004 18:58:32 -0400, Alexei Polkhanov wrote:

> Why I cannot type "Congress Library" but I
> have to go and type "dsgdsgf.dsfgdsfg.hfdhdf.gov" ? 

Which Congress Library? The Trades Unions Congress Library, or the 
Palau Congress Library? Or do you want the American Library Associations
Congress on Library and Information Services? Or the World Library and
Information Congress?

As a human in context, I would guess you mean the Library of Congress, but
my guesses aren't always right. What should "Steve Jackson" take you to?
Steve Jackson Games? The personal webpage of Steve Jackson (of SJG)?
The Steve Jackson who's a professor of biology at Cambridge? The Steve
Jackson who's a professor of botany at the University of Wyoming? The
page of the Steve Jackson that started Games Workshop?

Secondly, I've see few web address like "dsgdsgf.dsfgdsfg.hfdhdf.gov". The
Library of Congress is www.loc.gov. Most people who need that site more
than once or twice can remember that 11 character, especially as it's www
(terribly common) + loc (Library Of Congress) + gov (since it's a US
government institution.) Likewise www.sjgames.com isn't terribly hard to
remember either. Much easier than a 10 digit phone number. The hard ones
tend to be obscure and transitory, which would be terribly hard to fit
into your system. 

Finally, just like people record the phone numbers important to them,
people can bookmark pages they want to go back to. If you want "Congress
Library" to go to www.loc.gov, you can do so, and no one will complain. If
you want Steve Jackson to link to your grandson's webpage, no problem.

Re: BIND

[Alexei Polkhanov]

On Wed, 28 Apr 2004 02:34:12 +0000, David Starner wrote:

> On Mon, 26 Apr 2004 18:58:32 -0400, Alexei Polkhanov wrote:
> 
>> Why I cannot type "Congress Library" but I
>> have to go and type "dsgdsgf.dsfgdsfg.hfdhdf.gov" ? 
> 
> Which Congress Library? The Trades Unions Congress Library, or the 
> Palau Congress Library? Or do you want the American Library Associations
> Congress on Library and Information Services? Or the World Library and
> Information Congress?
> 
> As a human in context, I would guess you mean the Library of Congress, but
> my guesses aren't always right. What should "Steve Jackson" take you to?
> Steve Jackson Games? The personal webpage of Steve Jackson (of SJG)?
> The Steve Jackson who's a professor of biology at Cambridge? The Steve
> Jackson who's a professor of botany at the University of Wyoming? The
> page of the Steve Jackson that started Games Workshop?
> 
> Secondly, I've see few web address like "dsgdsgf.dsfgdsfg.hfdhdf.gov". The
> Library of Congress is www.loc.gov. Most people who need that site more
> than once or twice can remember that 11 character, especially as it's www
> (terribly common) + loc (Library Of Congress) + gov (since it's a US
> government institution.) Likewise www.sjgames.com isn't terribly hard to
> remember either. Much easier than a 10 digit phone number. The hard ones
> tend to be obscure and transitory, which would be terribly hard to fit
> into your system. 
> 

I am not imagining this things. There an RFC draft already, protocol called XRP. 
http://www.icann.org/tlds/agreements/biz/draft-brunner-xrp-00.txt 

> Finally, just like people record the phone numbers important to them,
> people can bookmark pages they want to go back to. If you want "Congress
> Library" to go to www.loc.gov, you can do so, and no one will complain. If
> you want Steve Jackson to link to your grandson's webpage, no problem.

from my understanding XRP is including DNS in its recent form as a subset.

We should really change subject from BIND to XRP then. Leave the BIND alone
and think about implementation of XRP server which has little to do with
BIND.

Sorry, whole discussion now sounds like an offtopic here, this is Ada 
newsgroup after all.

---
Alexei Polkhanov
Sr. Consultant/Software Systems Analyst
Tel: (604) 719-2515
E-mail: usenet@monteaureus.com
http://www.monteaureus.com/

Re: BIND

[Marius Amado Alves]

> > Why I cannot type "Congress Library" but I
> > have to go and type "dsgdsgf.dsfgdsfg.hfdhdf.gov" ?
>
> Which Congress Library? The Trades Unions Congress Library, or the
> Palau Congress Library?...

You don't need cryptism to have identity. The string "Trades Unions Congress 
Library" is just as unique as "dsgdsgf.dsfgdsfg.hfdhdf.gov".

Now, of course a smart system would help the user pinpoint the entity, using 
name similarity heuristics. And this makes much more sense for 'real' names 
(e.g "Congress Library") than to (current state) URLs.

BIND

[Andrew Carroll]

> ------------------------------
> On Mon, 26 Apr 2004 18:58:32 -0400, Alexei Polkhanov wrote:
> 
> > Why I cannot type "Congress Library" but I
> > have to go and type "dsgdsgf.dsfgdsfg.hfdhdf.gov" ? 
> 
Isn't this the CNRP?  Common Name Resolution Protocol?  RFC 3367.


Andrew Carroll
Carroll-Tech
720-273-6814
andrew@carroll-tech.net