Re: 3x releases (timing input)

[David Blevins <da...@visi.com>]

On Jun 29, 2007, at 9:25 AM, David Blevins wrote:

>
> On Jun 29, 2007, at 12:52 AM, Jacek Laskowski wrote:
>
>> On 6/26/07, David Blevins <da...@visi.com> wrote:
>>> We should probably set a date on this so we're motivated to resolve
>>> issues and start getting our release gears greased and moving.
>>>
>>> What would be your preference?
>>>
>>>    - 1 week
>>>    - 2 weeks
>>>    - 3 weeks
>>>    - 1 month
>>>
>>> My preference would be (in this order):
>>>    - 3 weeks (including voting and publishing)
>>
>> +1
>>
>> Could you point out the corners we should cut before release? I'm
>> leaning towards releasing what's available now (as Geronimo
>> certification is enough to ensure OpenEJB state as far as ejb3's
>> concerned), but your question seems to imply otherwise.
>
> Server doesn't boot :(  Run assemble and try and boot openejb- 
> standalone via the shell script.

Got that issue resolved.

-David

Re: 3x releases (timing input)

[David Blevins <da...@visi.com>]

On Jul 3, 2007, at 2:31 PM, Jacek Laskowski wrote:

> On 7/3/07, David Blevins <da...@visi.com> wrote:
>
>> > It's still m1, isn't it? I've been trying to
>> > migrate it to m2, but I'm not there yet.
>>
>> The m1 stuff hasn't worked since your m2 commit, so we could do it
>> manually or try to get one or the other working.
>
> Are you implying I busted the m1-based build? I thought I'm recovering
> it not destroying. Let me see what I can do to fix the overall
> perception ;-)

:)  It may not be too far gone.  In the assembly/openejb-standalone  
module add back in the test.xml you commented out and it should  
create a test client you can run from the command line as an  
executable jar.

-David

Re: 3x releases (timing input)

[Jacek Laskowski <ja...@laskowski.net.pl>]

On 7/3/07, David Blevins <da...@visi.com> wrote:

> > It's still m1, isn't it? I've been trying to
> > migrate it to m2, but I'm not there yet.
>
> The m1 stuff hasn't worked since your m2 commit, so we could do it
> manually or try to get one or the other working.

Are you implying I busted the m1-based build? I thought I'm recovering
it not destroying. Let me see what I can do to fix the overall
perception ;-)

Jacek

-- 
Jacek Laskowski
http://www.JacekLaskowski.pl

Re: 3x releases (timing input)

[David Blevins <da...@visi.com>]

On Jul 2, 2007, at 4:52 PM, Jacek Laskowski wrote:

> On 7/2/07, David Blevins <da...@visi.com> wrote:
>
>> Jacek, you want to try running the itests on the standalone server?
>
[..]
> Just to make sure I'm on the same track about the itests and the
> standalone server. The server boots fine after your changes, doesn't
> it?

Right.

> The task would be to run deploy the itests beans and run the
> itests, wouldn't it?

The manual process is to copy the itests ear into the apps/  
directory, boot the server, then run the itests client with the right  
stuff in it's classpath.

> It's still m1, isn't it? I've been trying to
> migrate it to m2, but I'm not there yet.

The m1 stuff hasn't worked since your m2 commit, so we could do it  
manually or try to get one or the other working.

-David

Re: 3x releases (timing input)

[Jacek Laskowski <ja...@laskowski.net.pl>]

On 7/2/07, David Blevins <da...@visi.com> wrote:

> Jacek, you want to try running the itests on the standalone server?

I've been working on it, but found out that a NetBeans plugin for
Geronimo is what people asked when I played with it during the
presentations of mine. Eclipse support for Java EE is quite limited.
I'll get to the itests again in the meantime when Geronimos won't
respond to my questions about JMX Remote API vs MEJB or Timer Service
support ;-)

Just to make sure I'm on the same track about the itests and the
standalone server. The server boots fine after your changes, doesn't
it? The task would be to run deploy the itests beans and run the
itests, wouldn't it? It's still m1, isn't it? I've been trying to
migrate it to m2, but I'm not there yet.

Jacek

-- 
Jacek Laskowski
http://www.JacekLaskowski.pl

Re: 3x releases (timing input)

[Karan Malhi <ka...@gmail.com>]

Yes,

I tested it immediately after the patch was committed. works fine on
winxp2 (my machine)

On 7/2/07, David Blevins <da...@visi.com> wrote:
>
> On Jun 29, 2007, at 9:25 AM, David Blevins wrote:
>
> >
> > On Jun 29, 2007, at 12:52 AM, Jacek Laskowski wrote:
> >
> >> On 6/26/07, David Blevins <da...@visi.com> wrote:
> >>> We should probably set a date on this so we're motivated to resolve
> >>> issues and start getting our release gears greased and moving.
> >>>
> >>> What would be your preference?
> >>>
> >>>    - 1 week
> >>>    - 2 weeks
> >>>    - 3 weeks
> >>>    - 1 month
> >>>
> >>> My preference would be (in this order):
> >>>    - 3 weeks (including voting and publishing)
> >>
> >> +1
> >>
> >> Could you point out the corners we should cut before release? I'm
> >> leaning towards releasing what's available now (as Geronimo
> >> certification is enough to ensure OpenEJB state as far as ejb3's
> >> concerned), but your question seems to imply otherwise.
> >
> > Server doesn't boot :(  Run assemble and try and boot openejb-
> > standalone via the shell script.
>
> Got that issue resolved.
>
> -David
>
>


-- 
Karan Singh Malhi

Re: 3x releases (timing input)

[Karan Malhi <ka...@gmail.com>]

works like a breeze for me :).
Here is what i did:
- Disconnected the network
- I had a single EJB wrapped up in a jar and deployed it to the apps folder,
- started the server
- view the log file which showed that it had successfully deployed the ejb
- wrote a test client, looked up the ejb and called a method on it

Everything worked fine, did not see any errors or anything in the log
which mentions anything about downloading a schema or looking for a
schema.

Is that what I was supposed to do to test it? If this is what I was
supposed to do then the files can be deleted and OPENEJB-555 can be
closed.

On 7/3/07, David Blevins <da...@visi.com> wrote:
>
> On Jul 3, 2007, at 10:13 AM, Karan Malhi wrote:
>
> >> Still have to clear up the sun xsd issue as mentioned earlier.
> > I deleted the xsd files from openejb-jee scheme directory and rebuilt,
> > it worked fine. I also looked into the JaxbJavaee and we are using a
> > non-validating SAXParser in there. Also looks like all the java
> > objects for the xsd files are generated, is there any other xsd for
> > which we still need java objects. Wouldnt simply deleting these files
> > fix the issue? Is there a place where we are dynamically generating
> > this tree of java objects and using them?
>
> We can definitely delete the xsds and let that be that.  I'm not sure
> running offline works as it seems some bit of parsing code is still
> trying to get the schemas from the internet.  I haven't tried yet but
> the way to test would be to unplug your network cable then boot the
> server with an app in it.
>
> -David
>
>


-- 
Karan Singh Malhi

Re: 3x releases (timing input)

[Dain Sundstrom <da...@iq80.com>]

It looks like Craig Russell is getting the license for schema's  
cleared up on the apache legal mailing list.  FWIU, the schemas are  
now available under CDDL so we can distribute them with our software.

Craig, does the license change apply to the old DTDs like EJB 1.1 and  
2.0 and xsds like EJB 2.1?

-dain

On Jul 3, 2007, at 10:57 AM, David Blevins wrote:

>
> On Jul 3, 2007, at 10:13 AM, Karan Malhi wrote:
>
>>> Still have to clear up the sun xsd issue as mentioned earlier.
>> I deleted the xsd files from openejb-jee scheme directory and  
>> rebuilt,
>> it worked fine. I also looked into the JaxbJavaee and we are using a
>> non-validating SAXParser in there. Also looks like all the java
>> objects for the xsd files are generated, is there any other xsd for
>> which we still need java objects. Wouldnt simply deleting these files
>> fix the issue? Is there a place where we are dynamically generating
>> this tree of java objects and using them?
>
> We can definitely delete the xsds and let that be that.  I'm not  
> sure running offline works as it seems some bit of parsing code is  
> still trying to get the schemas from the internet.  I haven't tried  
> yet but the way to test would be to unplug your network cable then  
> boot the server with an app in it.
>
> -David
>

Re: 3x releases (timing input)

[David Blevins <da...@visi.com>]

On Jul 3, 2007, at 10:13 AM, Karan Malhi wrote:

>> Still have to clear up the sun xsd issue as mentioned earlier.
> I deleted the xsd files from openejb-jee scheme directory and rebuilt,
> it worked fine. I also looked into the JaxbJavaee and we are using a
> non-validating SAXParser in there. Also looks like all the java
> objects for the xsd files are generated, is there any other xsd for
> which we still need java objects. Wouldnt simply deleting these files
> fix the issue? Is there a place where we are dynamically generating
> this tree of java objects and using them?

We can definitely delete the xsds and let that be that.  I'm not sure  
running offline works as it seems some bit of parsing code is still  
trying to get the schemas from the internet.  I haven't tried yet but  
the way to test would be to unplug your network cable then boot the  
server with an app in it.

-David

Re: 3x releases (timing input)

[Karan Malhi <ka...@gmail.com>]

> Still have to clear up the sun xsd issue as mentioned earlier.
I deleted the xsd files from openejb-jee scheme directory and rebuilt,
it worked fine. I also looked into the JaxbJavaee and we are using a
non-validating SAXParser in there. Also looks like all the java
objects for the xsd files are generated, is there any other xsd for
which we still need java objects. Wouldnt simply deleting these files
fix the issue? Is there a place where we are dynamically generating
this tree of java objects and using them?
-- 
Karan Singh Malhi

Re: 3x releases (timing input)

[David Blevins <da...@visi.com>]

On Jul 2, 2007, at 12:13 PM, Jacek Laskowski wrote:

> On 7/2/07, David Blevins <da...@visi.com> wrote:
>
>> Got that issue resolved.
>
> Are we ready to release it now? I would go for releasing at the same
> time as G 2.0 takes off. It's not that long, is it?

Still have to clear up the sun xsd issue as mentioned earlier.  But  
we should definitely be hammering on the server as if it were ready  
to go and flushing out issues.

Jacek, you want to try running the itests on the standalone server?

Anyone tried the tomcat integration lately?

-David

Re: 3x releases (timing input)

[Jacek Laskowski <ja...@laskowski.net.pl>]

On 7/2/07, David Blevins <da...@visi.com> wrote:

> Got that issue resolved.

Are we ready to release it now? I would go for releasing at the same
time as G 2.0 takes off. It's not that long, is it?

Jacek

-- 
Jacek Laskowski
http://www.JacekLaskowski.pl