Avalon and James (was JNDI for James)

["Noel J. Bergman" <no...@devtech.com>]

> At the same time it would be a lot more helpful to the process
> in James were in sync.  While I understand that this is not on
> the James short list of priorities

Excuse me?  Where did that come from?  I thought that we were cooperating
with you.

My understanding, Stephen, was that you were working on the changes, and I
have been watching the various parts of Avalon go into a release cycle.  My
impression was that we had agreed to pickup those parts as they became
release eligible (and, as you know, there is a fix in logkit 1.2
specifically because of James), and that you were in the process of making
the necessary code changes (e.g., CM -> SM).

Do you have the code changes ready?

Oh, that reminds me ... Serge, didn't we pass a vote to give Stephen commit
rights?  Do you need to verify the vote count, and submit a request to
infrastructure or root?

	--- Noel


---------------------------------------------------------------------
To unsubscribe, e-mail: james-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: james-dev-help@jakarta.apache.org

RE: Avalon and James (was JNDI for James)

["Noel J. Bergman" <no...@devtech.com>]

> So if it's an issue, we can put in Peter's patch.  In fact I thought
> since you had indicated it worked when you included Peter's patch, your
> patch would include his change.

Right now, I've set it to use the same code that is currently in v2.1.1.
I'd like to put Peter's changes into HEAD, but haven't had time.

	--- Noel


---------------------------------------------------------------------
To unsubscribe, e-mail: james-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: james-dev-help@jakarta.apache.org

Re: Avalon and James (was JNDI for James)

[Serge Knystautas <se...@lokitech.com>]

Noel J. Bergman wrote:
> The only problem I know of is due to a change in the smtp handler.  I
> reverted that change in the v2.1 branch, but didn't revert it in v3 because
> Serge wanted to take another look at it, plus I figured that we'd take a
> look at Peter's version.  Right now, I'm prepared to either revert the
> change or upgrade to Peter's version of the handler, which adds support for
> a bunch of RFC recommended extensions; whichever enables you to move
> forward.

Yeah, this was my bug in HEAD to address, and the dozen or so test cases 
I've written can't reproduce it unfortunately, and I can't get Linux 
installed at home to use the same testing software that's creating it. 
So if it's an issue, we can put in Peter's patch.  In fact I thought 
since you had indicated it worked when you included Peter's patch, your 
patch would include his change.

> IIRC, the vote passed for you to have commit karma.  Serge is the PMC Chair,
> and should be the one to send the request.

Sorry, I thought the person who nominated the commit privilege would 
send the email, but if now that we're a TLP it's my responsibility, I'll 
get on it.  From reading the Jakarta guidelines, it's ANY committer that 
can send the email to root at apache.org to get him access 
(http://jakarta.apache.org/site/roles.html).

Anyway, I'll send the email if I'm who people are waiting on.

-- 
Serge Knystautas
President
Lokitech >> software . strategy . design >> http://www.lokitech.com
p. 301.656.5501
e. sergek@lokitech.com


---------------------------------------------------------------------
To unsubscribe, e-mail: james-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: james-dev-help@jakarta.apache.org

RE: Avalon and James (was JNDI for James)

["Noel J. Bergman" <no...@devtech.com>]

Stephen,

I did read the thread.  We can take a vote, but I believe that the official
position is that James is working with Avalon to incorporate Avalon's
upcoming *release*.  And Avalon is working with James to make the changes
necessitated.

The version of James to roll-up is the version in CVS HEAD.  Unincorporated
changes from Peter's fork notwithstanding, the CVS HEAD represents the
future direction of James.  If during the Avalon upgrade path, we pick up
additional changes in an appropriate fashion, that is fine with me.

Peter has a James variant of his own that he will either bring back to
James, or will continue to be a fork.  He has picked up some of our changes,
we have gotten some of his.  Over time, I hope to include more of his
changes.

> this process that identifies that there is a definitate problem
> in James CVS head (which I presume still exist) and the issues
> are indepent of the CM-SM migration.

The only problem I know of is due to a change in the smtp handler.  I
reverted that change in the v2.1 branch, but didn't revert it in v3 because
Serge wanted to take another look at it, plus I figured that we'd take a
look at Peter's version.  Right now, I'm prepared to either revert the
change or upgrade to Peter's version of the handler, which adds support for
a bunch of RFC recommended extensions; whichever enables you to move
forward.

IIRC, the vote passed for you to have commit karma.  Serge is the PMC Chair,
and should be the one to send the request.

	--- Noel


---------------------------------------------------------------------
To unsubscribe, e-mail: james-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: james-dev-help@jakarta.apache.org

Re: Avalon and James (was JNDI for James)

[Stephen McConnell <mc...@apache.org>]

Hi Noel:

Noel J. Bergman wrote:

>>At the same time it would be a lot more helpful to the process
>>in James were in sync.  While I understand that this is not on
>>the James short list of priorities
>>
>
>Excuse me?  Where did that come from?  
>
  
Read the thread ... :-)


>I thought that we were cooperating with you.
>

There are a bunch of things going on - Avalon is starting is rollout of 
release, within a few weeks we will be hitting the Cornerstone content 
relase process which concerns James more than any other project.  That 
realese process is (from my point of view) is very much about supporting 
the excalation of James. However, this, again "from my point of view" 
needs an active process so we can get James in shape before release 
condidates start comming out of Avalon.  This has nothing to do with 
what version James locks into - it about enabling a mechanisms whereby I 
can validate the candidates I'll signing off on.  To do that we need up 
rollup James from CM-to-SM, sort out an identified loop, and arrive at a 
platform against which we can do some real evaluation.

>My understanding, Stephen, was that you were working on the changes, and I
>have been watching the various parts of Avalon go into a release cycle.  My
>impression was that we had agreed to pickup those parts as they became
>release eligible (and, as you know, there is a fix in logkit 1.2
>specifically because of James), and that you were in the process of making
>the necessary code changes (e.g., CM -> SM).
>

First cut on the CM-SM transition was coplete one month ago (today). 
 That process raised issues that after a couple of weeks were resolved 
down to being a combnination of somethig strage in James CVS head (the 
looping problems), combnined with some bugs inthe James test code, + 
probably issue relative the current Excalibur candidate releases on the 
thread and pooling areas.  A second cut at the transition was applied 
against Pete' James variant which proved to be totally succesfull - in 
fact it was this process that identifies that there is a definitate 
problem in James CVS head (which I presume still exist) and the the 
issues are indepent of the CM-SM migration. To be frank, I was hoping we 
would have cleared these things before the release process started over 
on Avalon.

>
>
>Do you have the code changes ready?
>

They have been ready for at least four weeks.  But to be honest - it't 
only in the last two weeks that I've know that it isn't the code change 
that is problamatic - it's the transition/target combination that is the 
area of interest/concern - and that's the thing I've been asking for 
support on - i.e. there is something problamatic in the James CVS head 
releative to the current builds.  Resolving/identifying this IHVHO is a 
prerequisite to Avalon Excalibur and Corenerstone realese candidate 
evalutation.  I suspect there are problems low down but I've aleady been 
through the process of validating tbis on at least four seperate 
occasions.  If I run either Phoneix or Merlin using the release 
candidate (and related candidates, etc.) I end up in an infite loop 
inside James.  This is somethig wrong and it needs to be fixed before 
the Avalon RC process for Exclibur and Cornerstone can really rolls into 
place.

>
>Oh, that reminds me ... Serge, didn't we pass a vote to give Stephen commit
>rights?  Do you need to verify the vote count, and submit a request to
>infrastructure or root?
>

Nothing as yet.

Commit privs or a target to send a zip/diff to - whatever - I don't mind.

Cheers, Steve.

-- 

Stephen J. McConnell
mailto:mcconnell@apache.org
http://www.osm.net




---------------------------------------------------------------------
To unsubscribe, e-mail: james-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: james-dev-help@jakarta.apache.org