You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@tomcat.apache.org by "George C. Hawkins" <ge...@yahoo.com> on 2001/03/27 22:38:43 UTC
[FOLLOWUP] Proposed Tomcat 4.0-Beta-2 Code Freeze Date?
The "Proposed Tomcat 4.0-Beta-2 Code Freeze Date" thread started on the
19th and fizzled out without any publicly visible conclusions. The
suggested tag date, i.e. the 22nd, has passed and no new one has been
proposed. Sorry to be pushy but is there a new consensus as to when to
tag. As far as I could see only Glenn strongly opposed the 22nd date and
I got the impression he could be talked out of it.
It's been quite a while since beta 1 and the viewing public anxiously
awaits a new release :-)
Yours,
George.
PS does anyone else feel that a once a week status report would be a
really good idea - nothing too big just a few lines from Craig say on
what he felt had been achieved in the last week, general feelings on
progess and non-binding comments on possible future beta/milestone
dates? This would keep nuisances like me from posting this kind of thing
and give everyone a feeling they knew how and where things were going.
_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com
[PATCH] Bugzilla #1002 (against CVS, this time)
Posted by Matthew L Daniel <md...@scdi.com>.
> * Deploy the struts-example application from Struts (includes Xerces
> in WEB-INF/lib)
That was the boost I needed. Now I got it.
Fortunately, it was the same problem as with m1, thus after you showed me
the systematic way to reproduce it, no problem.
Please see the comments in StandardClassLoader, and feel free to mail me
with questions that are not covered therein.
HTH,
-- /v\atthew
--
Matthew L Daniel Never put off until tomorrow what you can do today.
Wizard There might be a law against it by that time.
mdaniel@scdi.com -- `/usr/bin/fortune`
RE: sealing violation still exists? Was: [FOLLOWUP] Proposed Tomcat 4.0-Beta-2 Code Freeze Date?
Posted by Kevin Jones <ke...@develop.com>.
Matthew,
I'm not sure what you want here. I've just grabbed the latest CVS,
crimson.jar and jaxp.jar in the jasper directory are *not* sealed. This is
the interim fix that Craig put in place sometime over the weekend (this is
TC4). Do you think that the sealed versions should work as well now?
Kevin Jones
DevelopMentor
www.develop.com
> -----Original Message-----
> From: Matthew L Daniel [mailto:mdaniel@scdi.com]
> Sent: 29 March 2001 05:40
> To: tomcat-dev@jakarta.apache.org
> Subject: sealing violation still exists? Was: [FOLLOWUP] Proposed Tomcat
> 4.0-Beta-2 Code Freeze Date?
>
>
> > I believe that someone is working on the "sealing viloation" fix that he
> > wanted to include in beta 2.
>
> Can someone confirm (from a CVS build) that this bug still
> exists? The
> biggest thing that I need to see from someone else's machine is:
> * use the CVS jakarta-servletapi-4,
> * use the CVS jakarta-regexp-1.3-dev,
> * and make /double-backflippingly/ sure that the jars Catalina
> uses are still
> Sealed: true. The jars in your webapp shouldn't matter, but the
> ones in the
> various lib directories of Catalina must still be sealed (as they are by
> default)
>
> In unix, my favorite way of testing this is:
> cd $CATALINA_HOME;
> for j in `find . -name "*.jar"`; do
> echo $j;
> unzip -c $j META-INF/MANIFEST.MF | grep Sealed;
> done
>
> You should see that jaxp.jar and crimson.jar ARE, repeat, ARE
> sealed. If
> they are not, and you don't get the sealing violation, then you
> are a victim
> of the earlier CVS commits of unsealed jars.
>
> However, if you jump through my little hoops and you still
> don't (or do)
> get the sealing violation, PLEASE send me (and/or the list) email. The
> problem has vanished on my development machine, so it makes tracking the
> problem a nightmare.
>
> Craig knows all of this, but the more input I can get the better.
>
> -- /v\atthew
> --
> Matthew L Daniel Never put off until tomorrow what you can do today.
> Wizard There might be a law against it by that time.
> mdaniel@scdi.com -- `/usr/bin/fortune`
Re: sealing violation still exists? Was: [FOLLOWUP] Proposed Tomcat
4.0-Beta-2 Code Freeze Date?
Posted by "Craig R. McClanahan" <cr...@apache.org>.
On Wed, 28 Mar 2001, Matthew L Daniel wrote:
> > I believe that someone is working on the "sealing viloation" fix that he
> > wanted to include in beta 2.
>
> Can someone confirm (from a CVS build) that this bug still exists? The
> biggest thing that I need to see from someone else's machine is:
> * use the CVS jakarta-servletapi-4,
> * use the CVS jakarta-regexp-1.3-dev,
> * and make /double-backflippingly/ sure that the jars Catalina uses are still
> Sealed: true. The jars in your webapp shouldn't matter, but the ones in the
> various lib directories of Catalina must still be sealed (as they are by
> default)
>
> In unix, my favorite way of testing this is:
> cd $CATALINA_HOME;
> for j in `find . -name "*.jar"`; do
> echo $j;
> unzip -c $j META-INF/MANIFEST.MF | grep Sealed;
> done
>
> You should see that jaxp.jar and crimson.jar ARE, repeat, ARE sealed. If
> they are not, and you don't get the sealing violation, then you are a victim
> of the earlier CVS commits of unsealed jars.
>
> However, if you jump through my little hoops and you still don't (or do)
> get the sealing violation, PLEASE send me (and/or the list) email. The
> problem has vanished on my development machine, so it makes tracking the
> problem a nightmare.
>
Recent nightly builds (and the result you get when you run the unmodified
build process from CVS source) have a patched version of jaxp.jar and
crimson.jar that are not sealed. You would need to replace them with the
standard (sealed) versions, and run on a 1.3 JVM, to reproduce the problem
-- at least from my experience.
A quick way to reproduce:
* Replace the jaxp.jar and crimson.jar as mentioned above
* Deploy the struts-example application from Struts (includes Xerces
in WEB-INF/lib)
* Run on a JDK 1.3 platform (either Linux or Windows)
* Try to start Tomcat 4.0
* Experience sealing violations during the startup
> Craig knows all of this, but the more input I can get the better.
>
> -- /v\atthew
> --
> Matthew L Daniel Never put off until tomorrow what you can do today.
> Wizard There might be a law against it by that time.
> mdaniel@scdi.com -- `/usr/bin/fortune`
>
Craig
sealing violation still exists? Was: [FOLLOWUP] Proposed Tomcat 4.0-Beta-2 Code Freeze Date?
Posted by Matthew L Daniel <md...@scdi.com>.
> I believe that someone is working on the "sealing viloation" fix that he
> wanted to include in beta 2.
Can someone confirm (from a CVS build) that this bug still exists? The
biggest thing that I need to see from someone else's machine is:
* use the CVS jakarta-servletapi-4,
* use the CVS jakarta-regexp-1.3-dev,
* and make /double-backflippingly/ sure that the jars Catalina uses are still
Sealed: true. The jars in your webapp shouldn't matter, but the ones in the
various lib directories of Catalina must still be sealed (as they are by
default)
In unix, my favorite way of testing this is:
cd $CATALINA_HOME;
for j in `find . -name "*.jar"`; do
echo $j;
unzip -c $j META-INF/MANIFEST.MF | grep Sealed;
done
You should see that jaxp.jar and crimson.jar ARE, repeat, ARE sealed. If
they are not, and you don't get the sealing violation, then you are a victim
of the earlier CVS commits of unsealed jars.
However, if you jump through my little hoops and you still don't (or do)
get the sealing violation, PLEASE send me (and/or the list) email. The
problem has vanished on my development machine, so it makes tracking the
problem a nightmare.
Craig knows all of this, but the more input I can get the better.
-- /v\atthew
--
Matthew L Daniel Never put off until tomorrow what you can do today.
Wizard There might be a law against it by that time.
mdaniel@scdi.com -- `/usr/bin/fortune`
Re: [FOLLOWUP] Proposed Tomcat 4.0-Beta-2 Code Freeze Date?
Posted by Amy Roh <am...@eng.sun.com>.
----- Original Message -----
From: "Kief Morris" <ki...@bitbull.com>
To: <to...@jakarta.apache.org>
Sent: Wednesday, March 28, 2001 3:32 PM
Subject: Re: [FOLLOWUP] Proposed Tomcat 4.0-Beta-2 Code Freeze Date?
> Craig R. McClanahan typed the following on 01:23 PM 3/28/2001 -0800
> >> It's been quite a while since beta 1 and the viewing public anxiously
> >> awaits a new release :-)
> >
> >It was asked that we hold off for a bit. Would doing the release this
> >Friday meet everyone's preferences? If so, I will go ahead and submit
the
> >vote request formally.
>
> What are the issues for holding off?
I believe that someone is working on the "sealing viloation" fix that he
wanted to include in beta 2.
Amy
>
> Kief
>
Re: [FOLLOWUP] Proposed Tomcat 4.0-Beta-2 Code Freeze Date?
Posted by Kief Morris <ki...@bitbull.com>.
Craig R. McClanahan typed the following on 01:23 PM 3/28/2001 -0800
>> It's been quite a while since beta 1 and the viewing public anxiously
>> awaits a new release :-)
>
>It was asked that we hold off for a bit. Would doing the release this
>Friday meet everyone's preferences? If so, I will go ahead and submit the
>vote request formally.
What are the issues for holding off?
Kief
Re: [FOLLOWUP] Proposed Tomcat 4.0-Beta-2 Code Freeze Date?
Posted by Glenn Nielsen <gl...@voyager.apg.more.net>.
"Craig R. McClanahan" wrote:
>
> On Tue, 27 Mar 2001, George C. Hawkins wrote:
>
> > The "Proposed Tomcat 4.0-Beta-2 Code Freeze Date" thread started on the
> > 19th and fizzled out without any publicly visible conclusions. The
> > suggested tag date, i.e. the 22nd, has passed and no new one has been
> > proposed. Sorry to be pushy but is there a new consensus as to when to
> > tag. As far as I could see only Glenn strongly opposed the 22nd date and
> > I got the impression he could be talked out of it.
> >
> > It's been quite a while since beta 1 and the viewing public anxiously
> > awaits a new release :-)
> >
> > Yours,
> >
>
> It was asked that we hold off for a bit. Would doing the release this
> Friday meet everyone's preferences? If so, I will go ahead and submit the
> vote request formally.
>
I am trying to track down a classloading bug that prevents Jasper from
being able to translate a JSP page when it uses a taglib. This bug
is only seen when running Tomcat 4 with the SecurityManager and a policy
that grants explicit permissisons to Tocmat itself, like FilePermissions,
instead of granting Tomcat AllPermission. I should be able to get this
fixed by friday.
I have noticed one other item I would like to see fixed. On contest startup,
if a web apps web.xml is not valid, SAX fails. The error is sent to catalina.out
instead of to the host or context log. In a web hosting environment
this makes it difficult for the individual who's site is hosted to
find out about the invalid web.xml. No reason to hold up B2 for this
though.
Glenn
----------------------------------------------------------------------
Glenn Nielsen glenn@more.net | /* Spelin donut madder |
MOREnet System Programming | * if iz ina coment. |
Missouri Research and Education Network | */ |
----------------------------------------------------------------------
Re: [FOLLOWUP] Proposed Tomcat 4.0-Beta-2 Code Freeze Date?
Posted by "Craig R. McClanahan" <cr...@apache.org>.
On Tue, 27 Mar 2001, George C. Hawkins wrote:
> The "Proposed Tomcat 4.0-Beta-2 Code Freeze Date" thread started on the
> 19th and fizzled out without any publicly visible conclusions. The
> suggested tag date, i.e. the 22nd, has passed and no new one has been
> proposed. Sorry to be pushy but is there a new consensus as to when to
> tag. As far as I could see only Glenn strongly opposed the 22nd date and
> I got the impression he could be talked out of it.
>
> It's been quite a while since beta 1 and the viewing public anxiously
> awaits a new release :-)
>
> Yours,
>
It was asked that we hold off for a bit. Would doing the release this
Friday meet everyone's preferences? If so, I will go ahead and submit the
vote request formally.
>
> George.
>
> PS does anyone else feel that a once a week status report would be a
> really good idea - nothing too big just a few lines from Craig say on
> what he felt had been achieved in the last week, general feelings on
> progess and non-binding comments on possible future beta/milestone
> dates? This would keep nuisances like me from posting this kind of thing
> and give everyone a feeling they knew how and where things were going.
>
There are bad pests in the world, but you're being a good pest here
:-). I will try to keep everyone more up to date. Just as a warning,
I've got one more incredibly busy couple of weeks (O'Reilly Enterprise
Java Conference this week and ApacheCon next week) that will interfere
with the volume of my contributions.
Craig
Re: [FOLLOWUP] Proposed Tomcat 4.0-Beta-2 Code Freeze Date?
Posted by Remy Maucherat <re...@apache.org>.
> On 27 Mar, George C. Hawkins wrote:
> >
> > PS does anyone else feel that a once a week status report would be a
> > really good idea - nothing too big just a few lines from Craig say on
> > what he felt had been achieved in the last week, general feelings on
> > progess and non-binding comments on possible future beta/milestone
> > dates? This would keep nuisances like me from posting this kind of
> > thing and give everyone a feeling they knew how and where things were
> > going.
>
> You sound like a manager talking. I have lots of status from Craig. I
> see his CVS updates and all of the conversations he is involved on this
> list. To specifically write a status report would seem to me to be a bit
> much to ask. Further I can see Tomcat improve with my nightly update of
> the cvs tree.
At this point, I would recommend using the nightly builds until b2 is
released, if you're interested in trying out TC4.
Remy
Re: [FOLLOWUP] Proposed Tomcat 4.0-Beta-2 Code Freeze Date?
Posted by Ea...@EAS.San-Jose.CA.US.
On 27 Mar, George C. Hawkins wrote:
>
> PS does anyone else feel that a once a week status report would be a
> really good idea - nothing too big just a few lines from Craig say on
> what he felt had been achieved in the last week, general feelings on
> progess and non-binding comments on possible future beta/milestone
> dates? This would keep nuisances like me from posting this kind of
> thing and give everyone a feeling they knew how and where things were
> going.
You sound like a manager talking. I have lots of status from Craig. I
see his CVS updates and all of the conversations he is involved on this
list. To specifically write a status report would seem to me to be a bit
much to ask. Further I can see Tomcat improve with my nightly update of
the cvs tree.
-1 on status reports
=eas=