Re: gnats4.0

Re: gnats4.0

[Carl Edwards]

Hi All,

I'm starting a helpdesk project for our company's
SysAdmin group and was hoping to use GNATS.

I've got the cvs version up and running (with gnatsweb)
and now I've got to start the customizing.

My only other experience with this sort of tool is
ClearDDTS ... I am hoping this will be easier :-)

Pointers to additional documentation, or helpful
tips from previous and similar projects would be
appreciated.

Questions:

  Can I or has anyone tied the password authentication
  of gnatsd to the UNIX passwd system?
  
  If I put gnatsweb behind a secure site how do I stop
  the user from getting challenged 3 times?
    1) Web server
    2) GnatsWeb
    3) gnatsd
    
 

Thanks,
-Carl Edwards
 Vitesse Semiconductor
 

>Yes, it's available now!
>
>Best of all, it's *FREE* for your downloading pleasure!
>
>See the web page at http://sourceware.cygnus.com/gnats/
>
>which will tell you:
>
>If you would like to use the most recent version of GNATS under
>development, you can check out the source code using CVS:
>
>cvs -d :pserver:anoncvs@anoncvs.cygnus.com:/cvs/gnats login
>{enter "anoncvs" as the password}
>cvs -d :pserver:anoncvs@anoncvs.cygnus.com:/cvs/gnats co -P gnats
>
>
>We at Juniper are now using 4.0 with the latest gnatsweb in
>production.  I believe Bob is almost ready to announce 4.0 for FCS,
>however he's looking for folks to review and help update some of
>the documentation, which is *sorely* out of date.
>
>I would like to solicit all interested GNATS users to check out
>the latest code.  There are a bunch of really great new features.
>
>My personal favorite is that all GNATS fields are now fully
>configurable through a file called "field-config."  There is no
>need to edit 12 places in the source code to add or change a field.
>
>GNATS fields now can be sanity checked (regular expressions, or
>must match a valid date, or must match an enumerated type) by the
>server.
>
>The multiple database code has been cleaned up,  many MANY bug
>fixes for memory leaks and buffer overflows have been fixed (GNATS
>v3 was not exactly secure... I'd be very afraid to run GNATSD on
>an open server if I were you).
>
>The release tarball is coming right around the corner,  but folks
>can make it come faster if they will help us polish off the
>documentation.
>
>Please check out the latest code and send your feedback to either
>Bob or the list as appropriate.
>
>Paul
>
>p.s. GNATSv3 front-ends (e.g. tkgnats) may not be compatible with
>     GNATSv4 if they use GNATSD.  We have upgraded gnatsweb,  but
>     other code out there may need some cleanup and dusting off
>     to handle all the new features.
>
>
>
>
>
>
>
>On Fri, Mar 03, 2000 at 03:37:35PM -0500, Wasserman, Sherry wrote:
>> Hi There,
>> 
>> We're anxiously waiting for gnats4.0. Any idea when it will be available?
>> 
>> Sherry Wasserman
>> Comverse Network Systems
>> 1025 Briggs Rd. Suite 100
>> Mt. Laurel NJ 08054
>> (856) 608-2858
>> sherry.wasserman@comverse-in.ocm < mailto:demasi@comverse-in.com > 
>> 

gnats4.0

[Wasserman, Sherry]

Hello,

I'm wondering if there is a tar version of gnats4.0 anywhere yet?
If not, can someone please look into the possibility of making a tar version
available?

Thanks,

Sherry Wasserman
Comverse Network Systems

RE: gnats4.0

[Paul Traina]

I wouldn't try for any hybrid, at a bare minimum, one would be losing all of
the v4 feature enhancements.
The v3 and v4 index files are quite different (Bob fixed the '|' problem),
in v4, any field can now be indexed, so it's wickedly fast.

 -----Original Message-----
From: 	Rick Macdonald [ mailto:rickm@vsl.com ] 
Sent:	Sunday, April 09, 2000 6:25 PM
To:	Michael Richardson
Cc:	gnats-devel@sourceware.cygnus.com
Subject:	Re: gnats4.0 

On Sun, 9 Apr 2000, Michael Richardson wrote:

>   Would it be easier to write a v3->v4 gnatsd converter perhaps?

I don't think so. I actually have some ideas that might make the tkgnats
version much simpler; maybe it won't be so bad.

>   So, is there some way that I can run both v3 and v4 gnatsd? 

If the v3 and v4 index files were the same (I don't think they are) and if
the locking mechanism is the same, etc, perhaps v3 and v4 could co-exist
and work on the same database, assuming that you stick to the standard
fields. Just a thought. Bob would be able to answer this.

>   Can I convince the v3 tkGnats to just do read-only access via gnatsd,
> and submit all updates via email (which would go to a v4 system)?

You mean using the v3 gnatsd? This is the same answer that I have above
regarding the index file and locking, etc.

>   Well, maybe that doesn't help, since I need to support Win32 people,
> which is why I don't just point tkGnats at the disk. I think that pointing
> at the disk is strongly discouraged in v4..

Just so people know, pointing TkGnats "at the disk" means TkGnats runs
query-pr and pr-edit instead of talking to gnatsd. Except for deleting
PRs, TkGnats doesn't access gnats files directly (except for some old
bogus code that reads the gnats-adm/config file if the env var GNATS_ROOT
is set). Same issue here again: the index file has changed ("|" replaced
by \0 last time I looked. There may be other changes and issues).


...RickM...

Re: gnats4.0

[Rick Macdonald]

On Sun, 9 Apr 2000, Michael Richardson wrote:

>   Would it be easier to write a v3->v4 gnatsd converter perhaps?

I don't think so. I actually have some ideas that might make the tkgnats
version much simpler; maybe it won't be so bad.

>   So, is there some way that I can run both v3 and v4 gnatsd? 

If the v3 and v4 index files were the same (I don't think they are) and if
the locking mechanism is the same, etc, perhaps v3 and v4 could co-exist
and work on the same database, assuming that you stick to the standard
fields. Just a thought. Bob would be able to answer this.

>   Can I convince the v3 tkGnats to just do read-only access via gnatsd,
> and submit all updates via email (which would go to a v4 system)?

You mean using the v3 gnatsd? This is the same answer that I have above
regarding the index file and locking, etc.

>   Well, maybe that doesn't help, since I need to support Win32 people,
> which is why I don't just point tkGnats at the disk. I think that pointing
> at the disk is strongly discouraged in v4..

Just so people know, pointing TkGnats "at the disk" means TkGnats runs
query-pr and pr-edit instead of talking to gnatsd. Except for deleting
PRs, TkGnats doesn't access gnats files directly (except for some old
bogus code that reads the gnats-adm/config file if the env var GNATS_ROOT
is set). Same issue here again: the index file has changed ("|" replaced
by \0 last time I looked. There may be other changes and issues).


...RickM...

Re: gnats4.0

[Michael Richardson]

>>>>> "Rick" == Rick Macdonald <rickm@vsl.com> writes:
    Rick> On Sun, 9 Apr 2000, Michael Richardson wrote:

    >> I don't care about new features via tkGnats, I just need it to do what it
    >> does now. 

    Rick> Converting TkGnats to v4 is a major undertaking that has not yet started.

  Yes, I quite understand. I wish I had time to help. 

  Would it be easier to write a v3->v4 gnatsd converter perhaps?

    >> I would very much like to move to v4 now as I need to be create new PRs
    >> via GNATSD, which I think v3 can't do (correct?).

  So, is there some way that I can run both v3 and v4 gnatsd? 

  Can I convince the v3 tkGnats to just do read-only access via gnatsd,
and submit all updates via email (which would go to a v4 system)?

  Well, maybe that doesn't help, since I need to support Win32 people,
which is why I don't just point tkGnats at the disk. I think that pointing
at the disk is strongly discouraged in v4..

  gnatsweb was more popular with some people, but I'm not one of them.

   :!mcr!:            |  Solidum Systems Corporation, http://www.solidum.com
   Michael Richardson |For a better connected world,where data flows faster<tm>
 Personal: http://www.sandelman.ottawa.on.ca/People/Michael_Richardson/Bio.html
	mailto:mcr@sandelman.ottawa.on.ca	mailto:mcr@solidum.com

Re: gnats4.0

[Rick Macdonald]

On Sun, 9 Apr 2000, Michael Richardson wrote:

>   I don't care about new features via tkGnats, I just need it to do what it
> does now. 

Converting TkGnats to v4 is a major undertaking that has not yet started.

>   I would very much like to move to v4 now as I need to be create new PRs
> via GNATSD, which I think v3 can't do (correct?).

Correct.

...RickM...

Re: gnats4.0

[Michael Richardson]

>>>>> "Paul" == Paul Traina <pst@juniper.net> writes:
    Paul> p.s. GNATSv3 front-ends (e.g. tkgnats) may not be compatible with
    Paul>      GNATSv4 if they use GNATSD.  We have upgraded gnatsweb,  but
    Paul>      other code out there may need some cleanup and dusting off
    Paul>      to handle all the new features.

  Any advice on transitioning? Some way to run v4 on a copy of the PR 
database and propogate changes to the v3 database? (Maybe check the PR
database into CVS and run gnats on two checked out copies....?)
  That would let me put v4 through its paces without pissing off developers.
  
  I don't care about new features via tkGnats, I just need it to do what it
does now. 

  I would very much like to move to v4 now as I need to be create new PRs
via GNATSD, which I think v3 can't do (correct?).

   :!mcr!:            |  Solidum Systems Corporation, http://www.solidum.com
   Michael Richardson |For a better connected world,where data flows faster<tm>
 Personal: http://www.sandelman.ottawa.on.ca/People/Michael_Richardson/Bio.html
	mailto:mcr@sandelman.ottawa.on.ca	mailto:mcr@solidum.com


  

RE: gnats4.0

[Wasserman, Sherry]

Hi,
What kind of help do you need w/ the documentation and in what areas?  I'd
be happy to help!  I not only recognize the value and importance of good and
accurate documentation, I'm one of the rare s/w people who also likes to
write and has some experience in editing and writing user manuals for a
variety of applications.  

On the other hand, I have no experience in CVS -- we don't use it here, and
I'd rather not invest time in learning how to configure, build and use it
just so I can get the gnats 4 sources.  I assume that I'd need the programs
to work on any documentation.  Is there any way someone can tar up the
sources for me?  This would get us going on version 4 and any documentation
you want help with.

Sherry
----------------------------------------------------------------------------
---------------------------------------
		-----Original Message-----
		From:	Paul Traina [ mailto:pst@juniper.net ]
		Sent:	Friday, March 03, 2000 6:50 PM
		To:	Wasserman, Sherry
		Cc:	'gnats-devel@sourceware.cygnus.com';
bug-gnats@gnu.org
		Subject:	Re: gnats4.0

		Yes, it's available now!

		Best of all, it's *FREE* for your downloading pleasure!

		See the web page at http://sourceware.cygnus.com/gnats/

		which will tell you:

		If you would like to use the most recent version of GNATS
under
		development, you can check out the source code using CVS:

		cvs -d :pserver:anoncvs@anoncvs.cygnus.com:/cvs/gnats login
		{enter "anoncvs" as the password}
		cvs -d :pserver:anoncvs@anoncvs.cygnus.com:/cvs/gnats co -P
gnats


		We at Juniper are now using 4.0 with the latest gnatsweb in
		production.  I believe Bob is almost ready to announce 4.0
for FCS,
		however he's looking for folks to review and help update
some of
		the documentation, which is *sorely* out of date.

		I would like to solicit all interested GNATS users to check
out
		the latest code.  There are a bunch of really great new
features.

		My personal favorite is that all GNATS fields are now fully
		configurable through a file called "field-config."  There is
no
		need to edit 12 places in the source code to add or change a
field.

		GNATS fields now can be sanity checked (regular expressions,
or
		must match a valid date, or must match an enumerated type)
by the
		server.

		The multiple database code has been cleaned up,  many MANY
bug
		fixes for memory leaks and buffer overflows have been fixed
(GNATS
		v3 was not exactly secure... I'd be very afraid to run
GNATSD on
		an open server if I were you).

		The release tarball is coming right around the corner,  but
folks
		can make it come faster if they will help us polish off the
		documentation.

		Please check out the latest code and send your feedback to
either
		Bob or the list as appropriate.

		Paul

		p.s. GNATSv3 front-ends (e.g. tkgnats) may not be compatible
with
		     GNATSv4 if they use GNATSD.  We have upgraded gnatsweb,
but
		     other code out there may need some cleanup and dusting
off
		     to handle all the new features.







		On Fri, Mar 03, 2000 at 03:37:35PM -0500, Wasserman, Sherry
wrote:
		> Hi There,
		> 
		> We're anxiously waiting for gnats4.0. Any idea when it
will be available?
		> 
		> Sherry Wasserman
		> Comverse Network Systems
		> 1025 Briggs Rd. Suite 100
		> Mt. Laurel NJ 08054
		> (856) 608-2858
		> sherry.wasserman@comverse-in.ocm
< mailto:demasi@comverse-in.com > 
		> 

Re: gnats4.0

[Michael Brader]

[Trimmed to gnats-devel + Bob]

[Bob, there's a patch for mail-query.sh/Makefile.in at the end of this
message]

Paul Traina <pst@juniper.net> writes:

> Yes, it's available now!
> 
> [...]
> 
> We at Juniper are now using 4.0 with the latest gnatsweb in
> production.  I believe Bob is almost ready to announce 4.0 for FCS,
> however he's looking for folks to review and help update some of
> the documentation, which is *sorely* out of date.

Is there a TODO list that we can work with? I'm happy to work on doco
for 4.0 but wouldn't want to be duplicating someone else's efforts. It
seems that nothing has been done to gnats.texi and that all new doco
is in the manpages. I'll stick my hand up and volunteer to start
working on getting the TeXinfo in order, if we are going to maintain
the TeXinfo doco.

Anyway, here's a quick list of comments/suggestions/questions on the
manpages. Some of the comments might be more about gnats than the
doco. Obviously, prefix each comment with IMHO.

*** field-config

- General comments

        Can't we call it something else now as field-config is less
        and less relevant. Perhaps dbconfig or just config (it does
        supersede it after all) would be better.

        In the spirit of procmailex, a field-configex(5) containing
        examples of various types of configurations would help people
        warm to what is going to initially appear to be a very complex
        file.

- section: NAME

     "It is very helpful if the  administrator  has  a  reasonably
     good  understanding  of  the overall GNATS PR process before
     trying to create or edit a database configuration."

        We should tell them where to get this overview from (a pointer
        to Keeping Track or the SEE ALSO section would be good)

- section: Overall Database Configuration

          "If set to true, the database is placed into debug mode.
          This  causes  all  outgoing  email  to  be  sent to the
          gnats-admin user listed in the responsible adm file."

        Where can I find out what is meant by adm file? I assume it is
        defined as a file in the gnats-adm directory of a database.

          "If set to true, all of the Received:  headers  for  PRs
          submitted via email are kept in the PR; otherwise, only
          the first one is kept."

        We should probably add "Setting this option to true can take
        up a lot of disk space"

        In send-submitter-ack

          "This is in addition to the normal notification mail to the
          person(s) responsible for the new PR."

        s/mail to the/mail sent to the/

- section: Individual field configuration

        In builtin-name

          state           What's the frequency, Kenneth?

        :-) Perhaps "What state is the PR IN" or "The current state of
        the PR"

        In query-default

        The difference between exact-regexp and inexact-regexp is ?
        
- section: Field datatypes

        In enumerated-in-file:

        I understand the usefulness of this, particularly while people
        are upgrading, but should we not deprecate this in favour of
        the field-config file? Or am I missing something?

*** gnats.7

- section: DESCRIPTION

        "Many of the  primary  functions  available  with  GNATS  are
        accessible from within GNU Emacs."

        You might want to comment this out until it is true :-)

- section: STRUCTURE

        Wouldn't it make more sense with the first two paragraphs
        reversed, introducing databases then talking about what
        happens within each database

        Perhaps some of the stuff from the databases file man page
        could be put here as well, particularly the explanation of the
        default database as it seems likely to bite some people (it
        bit me :-)

     "The database name default is special; this is the entry used
     if  no database name is specified via a --database option or
     the GNATSDB environment variable."

- section: HISTORY

        Delete Elisp

*** gnatsd.8

- section: OPTIONS

          "son,  gnatsd  will immediately exit.  (It is unclear to
          the  author  of  this  man  page  why  this  option  is
          present.)"

        This is the reason (albeit somewhat dodgy) from gnats.texi

         There are three main features: 
         
         --------------------
         @example
         - security by obscurity, defined in the next paragraph
         - access level set by host name or IP address
         - access level set by userid and password
         @end example
         
         @code{gnatsd} has parameters @code{-r} or @code{--require-db}. If so
         executed, nobody gets in unless they are able to ask for a database by
         alias name with the @code{CHDB} (change database) command. If you
         specify an incorrect database alias the connection is closed. Only the
         @code{CHDB} command is allowed, and you just get one try.
         --------------------

        I think the option should be removed. It doesn't provide any
        real security but it might lead some people to believe that it
        does (a 'CHDB default' will almost always succeed in Gnats4).

- section: COMMANDS

        USER or CHDB

          maybe add at the end:

        Whenever a CHDB command is processed, the user's access level
        is set to the level for their host.

        ADMV

        I may find it later, but there seems to be no definition of an
        adm file (a file specified in an enumerated-in-file section in
        field-config) anywhere.

*** mkdb

- section: DESCRIPTION

        "The command should be run as @GNATS_USER@."

        Perhaps: The command must be run as @GNATS_USER@. or you will
        run into problems with file/directory permissions.

*** mkcat/rmcat

- section: DESCRIPTION

        Typo: s/histerical/historical/ or possibly hysterical :-)

*** pr-edit

- section: DESCRIPTION

        first line typo: s/exit/edit/

*** queue-pr

- section: DESCRIPTION

        first line: s/should run/should be run/ ??

        Maybe there should be more information about running queue-pr
        via cron with some examples, noting that there needs to be a
        cron entry for each database (with an associated -d option)
        Though maybe this should be in Keeping Track.

*** query-pr

-section: QUERY BY MAIL

        This needs updating to remove the deprecated operators from
        the manual. The program itself needs some work. I've included
        a patch at the end of this message for mail-query.sh and
        Makefile.in which works but still seems to have a couple of
        glitches with --expr. It also uses the deprecated
        --skip-closed flag. I don't have any need for this feature so
        I'm passing the buck.

Anyway, enjoy!

Michael Brader

-- 
Michael Brader
Aurema Pty Ltd (formerly Softway Pty Ltd)
PO Box 305, Strawberry Hills, NSW 2012, Australia
Email:mbrader@aurema.com, Tel: +61 2 9698 2322, Fax: +61 2 9699 9174

Re: gnats4.0

[Rick Macdonald]

On Fri, 3 Mar 2000, Paul Traina wrote:

> p.s. GNATSv3 front-ends (e.g. tkgnats) may not be compatible with
>      GNATSv4 if they use GNATSD.

Yes, I have a big upgrade job to do on TkGnats! Sounds like it's time to
start...

...RickM...

Re: gnats4.0

[Paul Traina]

Yes, it's available now!

Best of all, it's *FREE* for your downloading pleasure!

See the web page at http://sourceware.cygnus.com/gnats/

which will tell you:

If you would like to use the most recent version of GNATS under
development, you can check out the source code using CVS:

cvs -d :pserver:anoncvs@anoncvs.cygnus.com:/cvs/gnats login
{enter "anoncvs" as the password}
cvs -d :pserver:anoncvs@anoncvs.cygnus.com:/cvs/gnats co -P gnats


We at Juniper are now using 4.0 with the latest gnatsweb in
production.  I believe Bob is almost ready to announce 4.0 for FCS,
however he's looking for folks to review and help update some of
the documentation, which is *sorely* out of date.

I would like to solicit all interested GNATS users to check out
the latest code.  There are a bunch of really great new features.

My personal favorite is that all GNATS fields are now fully
configurable through a file called "field-config."  There is no
need to edit 12 places in the source code to add or change a field.

GNATS fields now can be sanity checked (regular expressions, or
must match a valid date, or must match an enumerated type) by the
server.

The multiple database code has been cleaned up,  many MANY bug
fixes for memory leaks and buffer overflows have been fixed (GNATS
v3 was not exactly secure... I'd be very afraid to run GNATSD on
an open server if I were you).

The release tarball is coming right around the corner,  but folks
can make it come faster if they will help us polish off the
documentation.

Please check out the latest code and send your feedback to either
Bob or the list as appropriate.

Paul

p.s. GNATSv3 front-ends (e.g. tkgnats) may not be compatible with
     GNATSv4 if they use GNATSD.  We have upgraded gnatsweb,  but
     other code out there may need some cleanup and dusting off
     to handle all the new features.







On Fri, Mar 03, 2000 at 03:37:35PM -0500, Wasserman, Sherry wrote:
> Hi There,
> 
> We're anxiously waiting for gnats4.0. Any idea when it will be available?
> 
> Sherry Wasserman
> Comverse Network Systems
> 1025 Briggs Rd. Suite 100
> Mt. Laurel NJ 08054
> (856) 608-2858
> sherry.wasserman@comverse-in.ocm < mailto:demasi@comverse-in.com > 
> 

gnats4.0

[Wasserman, Sherry]

Hi There,

We're anxiously waiting for gnats4.0. Any idea when it will be available?

Sherry Wasserman
Comverse Network Systems
1025 Briggs Rd. Suite 100
Mt. Laurel NJ 08054
(856) 608-2858
sherry.wasserman@comverse-in.ocm < mailto:demasi@comverse-in.com >