Mladen Turk wrote:
-Original Message-
From: Costin Manolache
It's most likely a bug. I know Nacho is using JNI with IIS - and the
code looks fine.
Yeah, it was... Just fixed it in the cvs. We were sending reply msg and
the handler was feeding post msg.
Oops I fill a bit guilty
mturk 2002/11/21 01:03:18
Modified:jk/native2/common jk_worker_lb.c jk_channel_jni.c
jk_channel.c
Log:
Get rid of the one-shot worker initialization scheme.
It doesn't work as it should, and it slows down the load balacer.
Once when service channel will
hgomez 2002/11/21 01:32:17
Modified:jk/java/org/apache/jk/common ChannelSocket.java
Log:
Report timeout exception as such
Revision ChangesPath
1.27 +7 -7
jakarta-tomcat-connectors/jk/java/org/apache/jk/common/ChannelSocket.java
Index: ChannelSocket.java
hgomez 2002/11/21 01:39:41
Modified:resources/confinstall server_2.xml
Log:
No timeout to be set in AJP13 connector, it's the webserver
responsability to drop unused connections
Revision ChangesPath
1.5 +1 -1
[EMAIL PROTECTED] wrote:
hgomez 2002/11/21 01:42:36
Modified:catalina/src/conf server.xml
Log:
No timeout to be set in AJP13 connector, it's the webserver
responsability to drop unused connections
Revision ChangesPath
1.65 +1 -1
Hi,
Can we disable the server shutdown port?
I was thinking if I set the port to 0, then the StandardServer shouldn't
create the shutdown socket.
The reason for that is cause the embedded TC doesn't really need that if
the JNI is the only communication channel.
Server could be shut down from
Mladen Turk wrote:
Hi,
Can we disable the server shutdown port?
I was thinking if I set the port to 0, then the StandardServer shouldn't
create the shutdown socket.
The reason for that is cause the embedded TC doesn't really need that if
the JNI is the only communication channel.
Server could be
-Original Message-
From: Remy Maucherat [mailto:[EMAIL PROTECTED]]
You should use the BootstrapService class instead of
Bootstrap.
Silly me ... ;-(.
What about ChannelSocket can it be disabled too ?
Something like:
public void init() throws IOException {
// Find a
OK... I will put the jk2 connector on my development machine and merge the
changes at my next opportunity.
Thanks again for your help.
.-.| Steven Velez
oo|| Software Engineer
/`'\| alventive
(\_;/) | 678-202-2226
-Original Message-
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=14736.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=14165.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=14165.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
Ok,
I did some testing running 4.1.14 as a service...
Now my knowledge of memory management is not so good so this might mean
nothing.
When running as a service if I goto every example jsp page (the examples
that come with the install)
Compiled
Memory starts at 37872 and ends at 39048
Not
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=14165.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=14165.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
I think we should define memory leak.
During the precompilation of the jsp pages, the resulting class files are
also loaded by the
classloader. This means of course, that the memory size increases.
It is up to the garbage collector to remove the loaded classes from memory.
In other words, it
Hi,
There has been lots of changes since 2.0.1 like:
- Regular expressions for uri mapping
- Duplicating global directives
- Fixing POST on JNI
- etc...
I'm planning to tag the JK2 on Monday 25th.
Any objections to that?
Also we still need to do a lot of work on the docs (hope that John will
that is unusual. I've done baseline performance tests with the standard
examples and haven't seen that kind of behavior with 4.1.10-12. Can you
start tomcat with verbose:gc and see what kind of output you're
getting. Just because the memory allocated to Java is 58megs at the
end, it doesn't
Hi,
I would like to donate the following code to the Apache community.
The main reason for that is missing of decent service dispatcher for the
Tomcat.
The application can be run either as a service, console application,
dll, or windows application.
Needs some work and testing, but it can be IMO
Mladen Turk wrote:
Hi,
There has been lots of changes since 2.0.1 like:
- Regular expressions for uri mapping
- Duplicating global directives
- Fixing POST on JNI
- etc...
I'm planning to tag the JK2 on Monday 25th.
Any objections to that?
I have some patches to commit. I will mail when I
Just to double check against tomcat 4.1.12, the memory usage seems
normal to me John. After tomcat starts up, it is using 23megs. My
webapp loads when tomcat starts up, so your numbers may be lower. The
system has 256megs of memory and 750mhz cpu.
after I I hit all the pages in /examples/jsp/
jfclere 2002/11/21 08:08:32
Modified:jk/native2/common jk_worker_jni.c
Log:
Allow no-threaded APR. - But no-threaded APR won't work anyway -
Revision ChangesPath
1.35 +2 -0 jakarta-tomcat-connectors/jk/native2/common/jk_worker_jni.c
Index: jk_worker_jni.c
When I run the examples with -verbose:gc this is what I get (just the
snippits at the end)
From uncompiled jsp pages
[GC 14707K-13665K(28412K), 0.0059002 secs]
[GC 15521K-15518K(28412K), 0.0128997 secs]
[GC 17374K-16086K(28412K), 0.0092266 secs]
[GC 17942K-16258K(28412K), 0.0086307 secs]
[GC
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=14741.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
jfclere 2002/11/21 08:23:59
Modified:jk/native2/common jk_channel_socket.c jk_channel_un.c
jk_logger_file.c
Log:
Add FD_CLOEXEC to prevent log file and sockets beeing opened in a son process.
Submitted by Martin Kraemer.
Revision ChangesPath
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=14742.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
That's like you said ... ;)
Is there a Java implementation withouth threads?
If not then, we should disable jni for non-threaded apr.
Make that dependent of HAVE_JNI. There is no need to compile in the JNI
for non-threaded servers.
Or simply...
#ifndef APR_HAS_THREADS
#error You will need to
This is from the examples jsp right? or are they from your webapp?
the latest optimizeit has a leak finder feature. you may want to
download the trail version and give it a shot. from the data it looks
like a lot of objects are promoted to old memory from heap. It doesn't
necessarily constitute
This is the examples with the install..
What I don't understand is why there is a difference when compiled vs
uncompiled and the diff is always there.
When we run our tests if our jsp pages are uncompiled the diff is about
170 meg.. That is what causes me to be a little nervous.
I will try
Mladen Turk wrote:
That's like you said ... ;)
Is there a Java implementation withouth threads?
I do not think so.
If not then, we should disable jni for non-threaded apr.
Make that dependent of HAVE_JNI. There is no need to compile in the JNI
for non-threaded servers.
Yes. But that is
Mladen Turk wrote:
Hi,
There has been lots of changes since 2.0.1 like:
- Regular expressions for uri mapping
- Duplicating global directives
- Fixing POST on JNI
- etc...
I'm planning to tag the JK2 on Monday 25th.
Any objections to that?
I remember that a FIONBIO was causing me some
-Original Message-
I remember that a FIONBIO was causing me some problems it
seems there is a #if 0 #endif So the code between is not used
any more. Could we remove it?
Yes... That was my invention after all (blocking blocked socket) ;-).
/* When a socket is created, it
This has been fixed, I think. There was a checkin by Costin to
jk/java/org/apache/jk/server JkMain.java - on 30/Oct/2001
So, I'm guessing that means Catalina 4.1.14 or so?
Jeff Tulley ([EMAIL PROTECTED])
(801)861-5322
Novell, Inc., the leading provider of Net business solutions
Well d/l optimizeit and realized it would take a few weeks for me to
understand what it is trying to tell me..
Just to double check against tomcat 4.1.12, the memory usage
seems normal to me John. After tomcat starts up, it is using
23megs. My webapp loads when tomcat starts up, so your
costin 2002/11/21 09:53:48
Modified:jk/native/common jk_ajp_common.c
Log:
Do not send the initial chunk for chunked encoding ( only for
regular POST ).
Ajp13 servers ( tomcat x.y, etc ) expect this initial chunk
only if a content-length is specified.
( this is just an
costin 2002/11/21 09:58:35
Modified:jk/native2/common jk_worker_ajp13.c
Log:
Same fix for jk2. Don't send initial chunk for chunked encoding.
The reason the data is sent for POST - in almost all cases tomcat
will read it ( required by the spec - to process the data ). This
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=14282.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
Looks good. Is this your code ( or code you have copyright to ) ?
IMO the best way would be to make a proposal and add it to j-t-c,
it seems very related with the jk_nt_service and mod_jk jni mode.
If you want it to be a standalone and separated program - I think
it would be better to propose
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=14741.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
+1
I'll try to add a bit more documentation ( and re-test ) the
config stuff ( reloading etc ) as well as the chunked fix.
Costin
Mladen Turk wrote:
Hi,
There has been lots of changes since 2.0.1 like:
- Regular expressions for uri mapping
- Duplicating global directives
- Fixing POST
The solution is not yet very well defined.
Right now catalina can be extended with Valves or connectors or listeners.
Some of them (like jk connector ) can define MBeans. The major problem
is having the admin interface recognize them - it should work
but I never tried.
I already committed some
+1
I don't know if this is known - the socket channel has a feature
to specify a range - if the port is taken it'll try the next one.
This allows a single config file to be used for a pool of tomcats
( i.e. simpler config for load-balancing ).
Costin
Mladen Turk wrote:
-Original
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=14741.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
Alexander Leyke wrote:
The JNI channel is special because it is single-threaded ( a doors channel
will use the same type of code ), and that has some implications in
how the request is processed - but the data encoding is the same.
Does it mean that in-process worker is a performance
-Original Message-
From: news [mailto:[EMAIL PROTECTED]] On Behalf Of Costin Manolache
Sent: Thursday, November 21, 2002 7:24 PM
To: [EMAIL PROTECTED]
Subject: Re: [DONATION] procrun
Looks good. Is this your code ( or code you have copyright to ) ?
My own.
IMO the best way
Am Don, 2002-11-21 um 19.29 schrieb Costin Manolache:
The solution is not yet very well defined.
yup. I have experienced the same.
Right now catalina can be extended with Valves or connectors or listeners.
Some of them (like jk connector ) can define MBeans. The major problem
is having
Am Don, 2002-11-21 um 19.29 schrieb Costin Manolache:
The solution is not yet very well defined.
yup. I have experienced the same.
Right now catalina can be extended with Valves or connectors or listeners.
Some of them (like jk connector ) can define MBeans. The major problem
is having
Jakob Praher wrote:
Right now catalina can be extended with Valves or connectors or
listeners.
Some of them (like jk connector ) can define MBeans. The major problem
is having the admin interface recognize them - it should work
but I never tried.
I'll have to look at them.
I only
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=14736.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
Costin Manolache wrote:
Looks good. Is this your code ( or code you have copyright to ) ?
IMO the best way would be to make a proposal and add it to j-t-c,
it seems very related with the jk_nt_service and mod_jk jni mode.
If you want it to be a standalone and separated program - I think
it
Costin Manolache wrote:
In jk2, all those strings are eliminated ( from C side ), we also eliminate
the buffer allocation ( we reuse the same jbyteArray and C array, with
code to eventually support NIO - that cuts 2 memcpy and few other small
things ). We also reduce the number of JNI calls to
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=14722.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
Remy Maucherat wrote:
Costin Manolache wrote:
Looks good. Is this your code ( or code you have copyright to ) ?
IMO the best way would be to make a proposal and add it to j-t-c,
it seems very related with the jk_nt_service and mod_jk jni mode.
If you want it to be a standalone and
Alexander Leyke wrote:
You can configure a different starter ( TomcatStarter did work
with an older version of BootstrapService - now you can't use
the 4.1 version of BS without daemon ). You can probably try
to just configure it to use o.a.catalina.startup.Bootstrap.
( or a different starter
Costin Manolache wrote:
I used BootstrapService because it didn't create the 8005 shutdown
socket. My tests were done with 4.0.
You can create your own wrapper, use CatalinaService directly or
call Bootstrap ( if you don't mind 8005 - in your case it doesn't
matter too much since you'll have a
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=14282.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
Alexander Leyke wrote:
I used BootstrapService because it didn't create the 8005 shutdown
socket. My tests were done with 4.0.
You can create your own wrapper, use CatalinaService directly or
call Bootstrap ( if you don't mind 8005 - in your case it doesn't
matter too much since you'll have a
billbarker2002/11/21 20:44:10
Modified:catalina/src/share/org/apache/catalina/mbeans
MBeanUtils.java
Log:
Relaxing the restrictions on creating Valve MBeans.
There is no functional change for Catalina Valves. For custom Valves, they only
need to
DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
http://nagoya.apache.org/bugzilla/show_bug.cgi?id=5952.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND
INSERTED IN THE BUG DATABASE.
59 matches
Mail list logo
