Re: problems with cocoon-2.0.1 and resin-2.0.3

[Torsten Curdt]

On Thu, 28 Feb 2002, Donald Ball wrote:
> On Thu, 28 Feb 2002, Torsten Curdt wrote:
>
> > That's exactly what I had...
> >
> > ...a fresh checkout solved it. see
> >
> >  http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=101377449417740&w=2
>
> so you're saying that 2.0.1 is known to be broken with resin, but the
> latest cvs does work?

Actually I cannot really tell anything about 2.0.1 in particular because
we only use the HEAD branch - but it was very close to the latest release
version IIRC. And the symptoms are the same...

At least HEAD a few days ago worked back fine with resin.

Hope you this helps
--
Torsten


-
Please check that your question has not already been answered in the
FAQ before posting. 

To unsubscribe, e-mail: <[EMAIL PROTECTED]>
For additional commands, e-mail: <[EMAIL PROTECTED]>

RE: problems with cocoon-2.0.1 and resin-2.0.3

[Vadim Gritsenko]

> From: Donald Ball [mailto:[EMAIL PROTECTED]]
> 
> On Thu, 28 Feb 2002, Torsten Curdt wrote:
> 
> > That's exactly what I had...
> >
> > ...a fresh checkout solved it. see
> >
> >
http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=101377449417740&w=2
> 
> so you're saying that 2.0.1 is known to be broken with resin, but the
> latest cvs does work?

Haven't upgraded Resin for a while, still working on 2.0.4... But I
doubt that C2.0.1 does not work with Resin. It was working for me... And
it still does.

Vadim

> 
> - donald
> 



-
Please check that your question has not already been answered in the
FAQ before posting. 

To unsubscribe, e-mail: <[EMAIL PROTECTED]>
For additional commands, e-mail: <[EMAIL PROTECTED]>

Re: problems with cocoon-2.0.1 and resin-2.0.3

[Donald Ball]

On Thu, 28 Feb 2002, Torsten Curdt wrote:

> That's exactly what I had...
>
> ...a fresh checkout solved it. see
>
>  http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=101377449417740&w=2

so you're saying that 2.0.1 is known to be broken with resin, but the
latest cvs does work?

- donald


-
Please check that your question has not already been answered in the
FAQ before posting. 

To unsubscribe, e-mail: <[EMAIL PROTECTED]>
For additional commands, e-mail: <[EMAIL PROTECTED]>

Re: problems with cocoon-2.0.1 and resin-2.0.3

[Torsten Curdt]

On Thu, 28 Feb 2002, Donald Ball wrote:

> one more bit of potentially useful information - when i start from scratch
> with a clean webapp tree, i note that cocoon-files directory created in
> the servlet's temporary work area, the only directories that are created
> are cache-dir and upload-dir - the org directory which normally serves as
> the root of the classpath for the dynamically generated sitemap and xsp
> classes is never created.
>
> so something seems to be going wrong when cocoon reaches the end of the
> cocoon.xconf file and tries to load sitemap.xmap. i just tried copying in
> the vanilla cocoon.xconf file from cocoon-2.0.1 and saw no change in
> behavior (my only mods had been adding a client-specific logicsheet,
> adding a data source, and changing the sitemap reloading to synchronous).
>
> - donald

That's exactly what I had...

...a fresh checkout solved it. see

 http://marc.theaimsgroup.com/?l=xml-cocoon-dev&m=101377449417740&w=2

Cheers
--
Torsten


-
Please check that your question has not already been answered in the
FAQ before posting. 

To unsubscribe, e-mail: <[EMAIL PROTECTED]>
For additional commands, e-mail: <[EMAIL PROTECTED]>

Re: problems with cocoon-2.0.1 and resin-2.0.3

[Torsten Curdt]

> and it does repeat itself periodically, losing 5-10k of free memory each
> time. access, error, and sitemap logs are all blank. resin won't stop
> normally, the java processes must be killed. looks like deadlock to me,
> although why it only happens when the server is being started from scratch
> is completely beyond me...
>
> and of course, frustratingly enough, the c2.0.1 sample webapp works fine,
> even its xsp pages. does anyone have any suggestions for where i might
> begin debugging this?


I had this ones - but unfortunately cannot exactly remember how I solved
it... :(

Did you compare both cocoon.xconfs? I remember doing this...

All I can tell: HEAD works just fine with resin 2.0.5 :)
--
Torsten


-
Please check that your question has not already been answered in the
FAQ before posting. 

To unsubscribe, e-mail: <[EMAIL PROTECTED]>
For additional commands, e-mail: <[EMAIL PROTECTED]>

Re: problems with cocoon-2.0.1 and resin-2.0.3

[Donald Ball]

one more bit of potentially useful information - when i start from scratch
with a clean webapp tree, i note that cocoon-files directory created in
the servlet's temporary work area, the only directories that are created
are cache-dir and upload-dir - the org directory which normally serves as
the root of the classpath for the dynamically generated sitemap and xsp
classes is never created.

so something seems to be going wrong when cocoon reaches the end of the
cocoon.xconf file and tries to load sitemap.xmap. i just tried copying in
the vanilla cocoon.xconf file from cocoon-2.0.1 and saw no change in
behavior (my only mods had been adding a client-specific logicsheet,
adding a data source, and changing the sitemap reloading to synchronous).

- donald


-
Please check that your question has not already been answered in the
FAQ before posting. 

To unsubscribe, e-mail: <[EMAIL PROTECTED]>
For additional commands, e-mail: <[EMAIL PROTECTED]>

problems with cocoon-2.0.1 and resin-2.0.3

[Donald Ball]

hey guys. i thought i'd take the webapp that was causing the xsp engine to
hang in tomcat-4.0.2 with cocoon-2.0.1 and try it with resin-2.0.3
instead. interesting results...

when i copy the webapp into resin's webapps area while the resin server is
still running, it notices the new webapp, initializes and configures it,
and everything works fine. cool. however, when i stop and the start resin
again, the entire servlet engine hangs after it tries to initialize the
c2-based webapp.

the last line in resin's error log is:

[2002/02/28 02:11:11] DEBUG   (2002-02-28) 02:11.11:770   [cocoon  ]
(Unknown-UR
I) Unknown-thread/ThreadSafeComponentHandler: ComponentHandler initialized
for:
org.apache.avalon.excalibur.component.ExcaliburComponentSelector

nothing very interesting or out of the ordinary appears anywhere in that
log. nothing shows up in any of the cocoon logs except for the store
janitor stuff in the core log:

DEBUG   (2002-02-28) 02:12.20:621   [core.store-janitor] (Unknown-URI)
Unknown-thread/StoreJanitorImpl: JVM total Memory: 4120576
DEBUG   (2002-02-28) 02:12.20:621   [core.store-janitor] (Unknown-URI)
Unknown-thread/StoreJanitorImpl: JVM free Memory: 1575328
DEBUG   (2002-02-28) 02:12.20:621   [core.store-janitor] (Unknown-URI)
Unknown-thread/StoreJanitorImpl: Memory is low = false

and it does repeat itself periodically, losing 5-10k of free memory each
time. access, error, and sitemap logs are all blank. resin won't stop
normally, the java processes must be killed. looks like deadlock to me,
although why it only happens when the server is being started from scratch
is completely beyond me...

and of course, frustratingly enough, the c2.0.1 sample webapp works fine,
even its xsp pages. does anyone have any suggestions for where i might
begin debugging this?

- donald




-
Please check that your question has not already been answered in the
FAQ before posting. 

To unsubscribe, e-mail: <[EMAIL PROTECTED]>
For additional commands, e-mail: <[EMAIL PROTECTED]>