date:20090925

Thanks for the suggestions Chris.

Unfortunately, the memory that was exhausted was the OldGen heap area, not 
PermGen, which doesn't show up in the Catalina log.

The heap allocation is quite hefty as this is a 64-bit environment...we need to 
get our developers to look into the application behaviour, but in the meantime 
I was looking for a way of dealing with the problem.

-Original Message-
From: Christopher Schultz [mailto:ch...@christopherschultz.net] 
Sent: 24 September 2009 19:59
To: Tomcat Users List
Subject: Re: Clustering Question...

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Darren,

On 9/24/2009 10:21 AM, Darren Kukulka wrote:
 In a 2-node scenario, where both nodes are configured identically and
 load balanced via Apache based on availability, how can we configure the
 cluster to deal with situations where one node has exhausted its Old Gen
 heap allocation.

Hmm... this is a situations that is difficult to detect using remote
code (like mod_jk).

 In such situations we've observed that the application being served by
 the cluster slow down considerably.  I can understand why this would be
 the case for sessions on the degraded node, but why would sessions on
 the good node suffer?

Are you using session replication? If so, the good Tomcat may be
slowing down attempting to replicate session changes to the damaged
Tomcat that is either not responding, or responding slowly, or
responding in confusing ways.

 How can we modify our configuration to deal with such occurrences more
 effectively?

After we had some trouble with OOMEs in production (legit ones,
actually: we just needed more heap), I implemented a quick-and-dirty
OOME checker. All it does is grep OutOfMemoryError catalina.out and,
if found, sends an email to someone.

Instead of emailing, you could have your OOME checker actually shut down
(or forceably terminate) the damaged Tomcat, and then the cluster should
stabilize. With only two nodes, this might be a problem, as the good
Tomcat will take over and might, under the new load of 100% of your
traffic, experience its own pergmen exhaustion and also be shut down.

You should consider adjusting your pergmen allocation (duh!) as well as
perhaps your heap allocation as well.

- -chris
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkq7wXcACgkQ9CaO5/Lv0PDVxACfT6X4tsFOEZ0nRWpYOIfLr7lX
XMIAoJUEs5uW3tTLqeRB5wCf1bo0oi4Q
=4LWQ
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org



Connaught plc is a FTSE 250 company. We are the UK's leading provider of 
integrated services operating in the compliance, social housing and public 
sector markets.

Please visit our website to see a full list of Connaught's Registered Companies 
www.connaught.plc.uk/group/aboutconnaught/registeredcompanies

Disclaimer:

The information transmitted is intended only for the person or entity to which 
it is addressed and may contain confidential and/or privileged material. Any 
review, retransmission, dissemination or other use of, or taking of any action 
in reliance upon, this information by persons or entities other than the 
intended recipient is prohibited. If you received this in error, please contact 
the sender and delete this message.

Connaught plc, Head Office 01392 444546

Re: Apache Tomcat, IIS, ISAPI

On 25.09.2009 08:06, Buchner, Joerg [T0I] (VW Sachsen) wrote:
 Hi,
 
 i've got an Problem.
 We have an Microsoft IIS 6.0 with an Apache Tomcat 5.5.27 with IIS ISAPI 
 1.2.28.
 (AMD64).
 
 The Website use Integrated Windows Authentication.
 
 The SourceCode Files on the server are protected by ActiveDirectory Groups.
 
 All works fine.
 
 But now the problem.
 
 If a user request a html File then the IIS check the NTFS 
 Filesystempermissions.
 That means, only users have access, which are in the security group on the 
 file system.
 
 But, if the same user request a jsp File, then there is no check on the 
 Filesystempermissons.
 
 That is not OK.
 I'ld like to protect my JSP Files with NTFS Rights.
 
 Can anybody help me?

The JSPs should be served by Tomcat, right? So IIS doesn't need to be
able to see them or allowed to actually read them.

So do you want to protect them in Tomcat?

Regards,

Rainer

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: Clustering Question...

On 25.09.2009 09:01, Darren Kukulka wrote:
 Thanks for the suggestions Chris.
 
 Unfortunately, the memory that was exhausted was the OldGen heap area, not 
 PermGen, which doesn't show up in the Catalina log.
 
 The heap allocation is quite hefty as this is a 64-bit environment...we need 
 to get our developers to look into the application behaviour, but in the 
 meantime I was looking for a way of dealing with the problem.

So did you get an OutOfMemoryError, or was there only a shortage of
memory resulting in more or less continuous garbage collection runs?

Regards,

Rainer

 -Original Message-
 From: Christopher Schultz [mailto:ch...@christopherschultz.net] 
 Sent: 24 September 2009 19:59
 To: Tomcat Users List
 Subject: Re: Clustering Question...
 
 Darren,
 
 On 9/24/2009 10:21 AM, Darren Kukulka wrote:
 In a 2-node scenario, where both nodes are configured identically and
 load balanced via Apache based on availability, how can we configure the
 cluster to deal with situations where one node has exhausted its Old Gen
 heap allocation.
 
 Hmm... this is a situations that is difficult to detect using remote
 code (like mod_jk).
 
 In such situations we've observed that the application being served by
 the cluster slow down considerably.  I can understand why this would be
 the case for sessions on the degraded node, but why would sessions on
 the good node suffer?
 
 Are you using session replication? If so, the good Tomcat may be
 slowing down attempting to replicate session changes to the damaged
 Tomcat that is either not responding, or responding slowly, or
 responding in confusing ways.
 
 How can we modify our configuration to deal with such occurrences more
 effectively?
 
 After we had some trouble with OOMEs in production (legit ones,
 actually: we just needed more heap), I implemented a quick-and-dirty
 OOME checker. All it does is grep OutOfMemoryError catalina.out and,
 if found, sends an email to someone.
 
 Instead of emailing, you could have your OOME checker actually shut down
 (or forceably terminate) the damaged Tomcat, and then the cluster should
 stabilize. With only two nodes, this might be a problem, as the good
 Tomcat will take over and might, under the new load of 100% of your
 traffic, experience its own pergmen exhaustion and also be shut down.
 
 You should consider adjusting your pergmen allocation (duh!) as well as
 perhaps your heap allocation as well.
 
 -chris

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org



Connaught plc is a FTSE 250 company. We are the UK's leading provider of
integrated services operating in the compliance, social housing and
public sector markets.

Please visit our website to see a full list of Connaught's Registered
Companies www.connaught.plc.uk/group/aboutconnaught/registeredcompanies

Disclaimer:

The information transmitted is intended only for the person or entity to
which it is addressed and may contain confidential and/or privileged
material. Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by persons or
entities other than the intended recipient is prohibited. If you
received this in error, please contact the sender and delete this message.

Connaught plc, Head Office 01392 444546


-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

RE: Clustering Question...

There are no OOME messages in the logs...the application still runs,
albeit very slowly...GC is not kicking in any more frequently...I was
under the impression that GC doesn't touch OldGen though...?

-Original Message-
From: Rainer Jung [mailto:rainer.j...@kippdata.de] 
Sent: 25 September 2009 08:09
To: Tomcat Users List
Subject: Re: Clustering Question...

On 25.09.2009 09:01, Darren Kukulka wrote:
 Thanks for the suggestions Chris.

 Unfortunately, the memory that was exhausted was the OldGen heap area,
not PermGen, which doesn't show up in the Catalina log.

 The heap allocation is quite hefty as this is a 64-bit
environment...we need to get our developers to look into the application
behaviour, but in the meantime I was looking for a way of dealing with
the problem.

So did you get an OutOfMemoryError, or was there only a shortage of
memory resulting in more or less continuous garbage collection runs?

Regards,

Rainer

 -Original Message-
 From: Christopher Schultz [mailto:ch...@christopherschultz.net] 
 Sent: 24 September 2009 19:59
 To: Tomcat Users List
 Subject: Re: Clustering Question...

 Darren,

 On 9/24/2009 10:21 AM, Darren Kukulka wrote:
 In a 2-node scenario, where both nodes are configured identically and
 load balanced via Apache based on availability, how can we configure
the
 cluster to deal with situations where one node has exhausted its Old
Gen
 heap allocation.

 Hmm... this is a situations that is difficult to detect using remote
 code (like mod_jk).

 In such situations we've observed that the application being served
by
 the cluster slow down considerably.  I can understand why this would
be
 the case for sessions on the degraded node, but why would sessions on
 the good node suffer?

 Are you using session replication? If so, the good Tomcat may be
 slowing down attempting to replicate session changes to the damaged
 Tomcat that is either not responding, or responding slowly, or
 responding in confusing ways.

 How can we modify our configuration to deal with such occurrences
more
 effectively?

 After we had some trouble with OOMEs in production (legit ones,
 actually: we just needed more heap), I implemented a quick-and-dirty
 OOME checker. All it does is grep OutOfMemoryError catalina.out and,
 if found, sends an email to someone.

 Instead of emailing, you could have your OOME checker actually shut
down
 (or forceably terminate) the damaged Tomcat, and then the cluster
should
 stabilize. With only two nodes, this might be a problem, as the good
 Tomcat will take over and might, under the new load of 100% of your
 traffic, experience its own pergmen exhaustion and also be shut down.

 You should consider adjusting your pergmen allocation (duh!) as well
as
 perhaps your heap allocation as well.

 -chris

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Connaught plc is a FTSE 250 company. We are the UK's leading provider of
integrated services operating in the compliance, social housing and
public sector markets.

Please visit our website to see a full list of Connaught's Registered
Companies www.connaught.plc.uk/group/aboutconnaught/registeredcompanies

Disclaimer:

The information transmitted is intended only for the person or entity to
which it is addressed and may contain confidential and/or privileged
material. Any review, retransmission, dissemination or other use of, or
taking of any action in reliance upon, this information by persons or
entities other than the intended recipient is prohibited. If you
received this in error, please contact the sender and delete this
message.

Connaught plc, Head Office 01392 444546

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: Clustering Question...

On 25.09.2009 09:12, Darren Kukulka wrote:
 There are no OOME messages in the logs...the application still runs,
 albeit very slowly...GC is not kicking in any more frequently...I was
 under the impression that GC doesn't touch OldGen though...?

It does.

So it's strange you know your memory is exhausted and that's the reason
for the slowness, although if it were exhausted, you would either get an
OutOfMemoryError or GC has to kick in.

The picture is not complete here.

Regards,

Rainer

 -Original Message-
 From: Rainer Jung [mailto:rainer.j...@kippdata.de] 
 Sent: 25 September 2009 08:09
 To: Tomcat Users List
 Subject: Re: Clustering Question...
 
 On 25.09.2009 09:01, Darren Kukulka wrote:
 Thanks for the suggestions Chris.

 Unfortunately, the memory that was exhausted was the OldGen heap area,
 not PermGen, which doesn't show up in the Catalina log.

 The heap allocation is quite hefty as this is a 64-bit
 environment...we need to get our developers to look into the application
 behaviour, but in the meantime I was looking for a way of dealing with
 the problem.
 
 So did you get an OutOfMemoryError, or was there only a shortage of
 memory resulting in more or less continuous garbage collection runs?
 
 Regards,
 
 Rainer
 
 -Original Message-
 From: Christopher Schultz [mailto:ch...@christopherschultz.net] 
 Sent: 24 September 2009 19:59
 To: Tomcat Users List
 Subject: Re: Clustering Question...

 Darren,

 On 9/24/2009 10:21 AM, Darren Kukulka wrote:
 In a 2-node scenario, where both nodes are configured identically and
 load balanced via Apache based on availability, how can we configure
 the
 cluster to deal with situations where one node has exhausted its Old
 Gen
 heap allocation.

 Hmm... this is a situations that is difficult to detect using remote
 code (like mod_jk).

 In such situations we've observed that the application being served
 by
 the cluster slow down considerably.  I can understand why this would
 be
 the case for sessions on the degraded node, but why would sessions on
 the good node suffer?

 Are you using session replication? If so, the good Tomcat may be
 slowing down attempting to replicate session changes to the damaged
 Tomcat that is either not responding, or responding slowly, or
 responding in confusing ways.

 How can we modify our configuration to deal with such occurrences
 more
 effectively?

 After we had some trouble with OOMEs in production (legit ones,
 actually: we just needed more heap), I implemented a quick-and-dirty
 OOME checker. All it does is grep OutOfMemoryError catalina.out and,
 if found, sends an email to someone.

 Instead of emailing, you could have your OOME checker actually shut
 down
 (or forceably terminate) the damaged Tomcat, and then the cluster
 should
 stabilize. With only two nodes, this might be a problem, as the good
 Tomcat will take over and might, under the new load of 100% of your
 traffic, experience its own pergmen exhaustion and also be shut down.

 You should consider adjusting your pergmen allocation (duh!) as well
 as
 perhaps your heap allocation as well.

 -chris
 
 -
 To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
 For additional commands, e-mail: users-h...@tomcat.apache.org
 
 
 
 Connaught plc is a FTSE 250 company. We are the UK's leading provider of
 integrated services operating in the compliance, social housing and
 public sector markets.
 
 Please visit our website to see a full list of Connaught's Registered
 Companies www.connaught.plc.uk/group/aboutconnaught/registeredcompanies
 
 Disclaimer:
 
 The information transmitted is intended only for the person or entity to
 which it is addressed and may contain confidential and/or privileged
 material. Any review, retransmission, dissemination or other use of, or
 taking of any action in reliance upon, this information by persons or
 entities other than the intended recipient is prohibited. If you
 received this in error, please contact the sender and delete this
 message.
 
 Connaught plc, Head Office 01392 444546

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: Virtual host configuration - best practise?

2009-09-25 Thread hepabolu

awarnier wrote:

hepabolu wrote:
...

However, from the Tomcat docs I understand that any change to server.xml
requires a restart of Tomcat which would mean that the existing
(non-dummy)
hosts which are already in production, i.e. up and running, will also be
taken offline. This is the heart of the problem...

Just as another look on the issue.

skip/

This may be one case where the added complication of having an Apache
httpd in front of Tomcat is justified.

I'm not sure if this will not result in unwanted side-effects. Each webapp
should store/retrieve information from a MySQL database (one per
host/client). There should be absolutely no possibility of a mix of
information of clients.
Each webapp serves as a webservice to an external webapp/webservice. They
communicate using the predefined domainname.
I don't think it's a good idea to have one webapp handle all requests based
on parameters alone. There are already a lot of parameters to handle a
single request, figuring out which 'client' should respond makes it all the
more complicated.

I could look into the 'multiple webapps in a single domain' setup. I just
wonder if there's (a) extra processing time due to the apache + mod_jk
configuration and (b) added complexity to figure out which webapp is the
culprit in case of troubles. I'll take (b) for granted if (a) is almost
nothing as I'm already having troubles with timeout deadlines.

Do you have any info on processing times using apache + mod_jk?

Thanks.

Bye, Helma
--
View this message in context:
http://www.nabble.com/Virtual-host-configuration---best-practise--tp25512289p25530061.html
Sent from the Tomcat - User mailing list archive at Nabble.com.

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

RE: Clustering Question...

Not sure what to tell you...

When OldGen becomes full, the system basically slows down to the point
of becoming non-responsive...but it does not report any OOME.

GC appears to have no affect, whether instigated by the JVM or manually
kicked off in JConsole...it will clear down other heap areas, but not
OldGen

-Original Message-
From: Rainer Jung [mailto:rainer.j...@kippdata.de] 
Sent: 25 September 2009 08:20
To: Tomcat Users List
Subject: Re: Clustering Question...

On 25.09.2009 09:12, Darren Kukulka wrote:
 There are no OOME messages in the logs...the application still runs,
 albeit very slowly...GC is not kicking in any more frequently...I was
 under the impression that GC doesn't touch OldGen though...?

It does.

So it's strange you know your memory is exhausted and that's the reason
for the slowness, although if it were exhausted, you would either get an
OutOfMemoryError or GC has to kick in.

The picture is not complete here.

Regards,

Rainer

 -Original Message-
 From: Rainer Jung [mailto:rainer.j...@kippdata.de] 
 Sent: 25 September 2009 08:09
 To: Tomcat Users List
 Subject: Re: Clustering Question...
 
 On 25.09.2009 09:01, Darren Kukulka wrote:
 Thanks for the suggestions Chris.

 Unfortunately, the memory that was exhausted was the OldGen heap
area,
 not PermGen, which doesn't show up in the Catalina log.

 The heap allocation is quite hefty as this is a 64-bit
 environment...we need to get our developers to look into the
application
 behaviour, but in the meantime I was looking for a way of dealing with
 the problem.
 
 So did you get an OutOfMemoryError, or was there only a shortage of
 memory resulting in more or less continuous garbage collection runs?
 
 Regards,
 
 Rainer
 
 -Original Message-
 From: Christopher Schultz [mailto:ch...@christopherschultz.net] 
 Sent: 24 September 2009 19:59
 To: Tomcat Users List
 Subject: Re: Clustering Question...

 Darren,

 On 9/24/2009 10:21 AM, Darren Kukulka wrote:
 In a 2-node scenario, where both nodes are configured identically
and
 load balanced via Apache based on availability, how can we configure
 the
 cluster to deal with situations where one node has exhausted its Old
 Gen
 heap allocation.

 Hmm... this is a situations that is difficult to detect using remote
 code (like mod_jk).

 In such situations we've observed that the application being served
 by
 the cluster slow down considerably.  I can understand why this would
 be
 the case for sessions on the degraded node, but why would sessions
on
 the good node suffer?

 Are you using session replication? If so, the good Tomcat may be
 slowing down attempting to replicate session changes to the damaged
 Tomcat that is either not responding, or responding slowly, or
 responding in confusing ways.

 How can we modify our configuration to deal with such occurrences
 more
 effectively?

 After we had some trouble with OOMEs in production (legit ones,
 actually: we just needed more heap), I implemented a quick-and-dirty
 OOME checker. All it does is grep OutOfMemoryError catalina.out
and,
 if found, sends an email to someone.

 Instead of emailing, you could have your OOME checker actually shut
 down
 (or forceably terminate) the damaged Tomcat, and then the cluster
 should
 stabilize. With only two nodes, this might be a problem, as the good
 Tomcat will take over and might, under the new load of 100% of your
 traffic, experience its own pergmen exhaustion and also be shut down.

 You should consider adjusting your pergmen allocation (duh!) as well
 as
 perhaps your heap allocation as well.

 -chris
 
 -
 To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
 For additional commands, e-mail: users-h...@tomcat.apache.org
 
 
 
 Connaught plc is a FTSE 250 company. We are the UK's leading provider
of
 integrated services operating in the compliance, social housing and
 public sector markets.
 
 Please visit our website to see a full list of Connaught's Registered
 Companies
www.connaught.plc.uk/group/aboutconnaught/registeredcompanies
 
 Disclaimer:
 
 The information transmitted is intended only for the person or entity
to
 which it is addressed and may contain confidential and/or privileged
 material. Any review, retransmission, dissemination or other use of,
or
 taking of any action in reliance upon, this information by persons or
 entities other than the intended recipient is prohibited. If you
 received this in error, please contact the sender and delete this
 message.
 
 Connaught plc, Head Office 01392 444546

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org


-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

AW: Apache Tomcat, IIS, ISAPI

2009-09-25 Thread Buchner, Joerg [T0I] (VW Sachsen)

Hi Rainer,

yes, the JSP's should be served by Tomcat! 

But Tomcat should not work as WebService,
Tomcat is only an Engine behind IIS.

The protection of the JSP Files should be realized in the NTFS ACL's...

Regards,
Jörg
 
 


-Ursprüngliche Nachricht-
Von: Rainer Jung [mailto:rainer.j...@kippdata.de] 
Gesendet: Freitag, 25. September 2009 09:07
An: Tomcat Users List
Betreff: Re: Apache Tomcat, IIS, ISAPI

On 25.09.2009 08:06, Buchner, Joerg [T0I] (VW Sachsen) wrote:
 Hi,
 
 i've got an Problem.
 We have an Microsoft IIS 6.0 with an Apache Tomcat 5.5.27 with IIS ISAPI 
 1.2.28.
 (AMD64).
 
 The Website use Integrated Windows Authentication.
 
 The SourceCode Files on the server are protected by ActiveDirectory Groups.
 
 All works fine.
 
 But now the problem.
 
 If a user request a html File then the IIS check the NTFS 
 Filesystempermissions.
 That means, only users have access, which are in the security group on the 
 file system.
 
 But, if the same user request a jsp File, then there is no check on the 
 Filesystempermissons.
 
 That is not OK.
 I'ld like to protect my JSP Files with NTFS Rights.
 
 Can anybody help me?

The JSPs should be served by Tomcat, right? So IIS doesn't need to be
able to see them or allowed to actually read them.

So do you want to protect them in Tomcat?

Regards,

Rainer

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org


-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: Virtual host configuration - best practise?

2009-09-25 Thread Pid


On 25/09/2009 08:26, hepabolu wrote:



awarnier wrote:


hepabolu wrote:
...


However, from the Tomcat docs I understand that any change to server.xml
requires a restart of Tomcat which would mean that the existing
(non-dummy)
hosts which are already in production, i.e. up and running, will also be
taken offline. This is the heart of the problem...


Just as another look on the issue.

skip/

This may be one case where the added complication of having an Apache
httpd in front of Tomcat is justified.




Do you have any info on processing times using apache + mod_jk?


Your app will be the defining factor in terms of responsivity, there is 
an overhead when using HTTPD in front of Tomcat but it's usually 
negligible compared to the time spent in your application.



OP: Chris asked you if you were using session replication, I'd also ask 
if so, which type?


If your interest is in load balancing, you are using sticky sessions and 
you are not too bothered about forcing re-login or session 
re-initialisation when the client shifts node, you could disable session 
replication.


A positive change in performance would confirm that session replication 
is causing the slow down in the one node.



You didn't specify (did you?) which algorithm you are using to load 
balance, have you experimented with the various options available?



p




Thanks.

Bye, Helma



-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

RE: Clustering Question...

 From: Darren Kukulka [mailto:darren.kuku...@connaught.plc.uk]
 Subject: RE: Clustering Question...

 When OldGen becomes full, the system basically slows down to the point
 of becoming non-responsive...but it does not report any OOME.

Then a major GC likely is running almost continuously, which will suspend the 
application threads.

What are your heap-related commmand line settings?  If you have artificially 
constrained the size of the OldGen, you could easily have a self-induced 
problem.  Setting -XX:NewRatio or any of several other options will prevent 
HotSpot from dynamically adjusting the balance between NewGen and OldGen.

Or, you may simply be out of heap space, in which case you'll need to fix your 
application so that it's not so greedy.  Excessive caching without employing a 
SoftReference technique is a frequent cause.

 - Chuck

THIS COMMUNICATION MAY CONTAIN CONFIDENTIAL AND/OR OTHERWISE PROPRIETARY 
MATERIAL and is thus for use only by the intended recipient. If you received 
this in error, please contact the sender and delete the e-mail and its 
attachments from all computers.

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: php 5 and tomcat 6

2009-09-25 Thread Josh Gooding

Chris,

I think you're right about it being a PHP question, I was just inquiring if
someone here had the PECL dll files that I need.

Here is what I was looking at:

http://tojackwu.blogspot.com/2008/10/php-5-on-tomcat-6-installation.html

Ziggy - I'm aware of liferay.  Off the top of your head do you know if it is
DLL based or is it via CGI?  If you don't it's not big deal, don't look it
up, I just wondered if you had a quick and dirty answer.

Tom, as for why would I not install httpd?  Had it, configured it, had it
running, and then... removed it.  It is nothing more than another layer in
the mix.  I have SSH configured correctly via my tomcat installation,
everything is customized to where I need it to be, and my tomcat is ironclad
as of now.  My attitude is the old KISS mentality.  Why would I want to add
another possible issue or layer in the mix?  Then I'd have 2 web servers
running, and the list of what issues could possibly happen more than
doubles.  Tomcat's performance is just as good as httpd, why would I run and
have to cross configure 2 webservers when 1 will do just fine?

Security?  Not worried, this is a pet project of mine.  If I can get a
drupal installation running via a tomcat / php port, I'll be happy.  If
someone hacks it, i'm not worried.

Ken - Thanks for the heads up on the other org.  I'll look into that as
well.

- Josh

On Thu, Sep 24, 2009 at 6:53 PM, Ken Bowen kbo...@als.com wrote:

 Argh.  Sorry for the list spam.  Was sending this to someone interested.


 On Sep 24, 2009, at 5:28 PM, Ken Bowen wrote:

  Sorry for dropping the link.
 The reference showed up in this:

 http://quercus.caucho.com/casestudies/Caucho_LiveProcess_casestudy.pdf

 The company is:  http://www.liveprocess.com/


 On Sep 24, 2009, at 1:36 PM, Hassan Schroeder wrote:

  On Thu, Sep 24, 2009 at 8:40 AM, Tommy Pham tommy...@yahoo.com wrote:

  Only way to get PHP running in Tomcat is through CGI.


 Not true: http://quercus.caucho.com/

 --
 Hassan Schroeder  hassan.schroe...@gmail.com
 twitter: @hassan

 -
 To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
 For additional commands, e-mail: users-h...@tomcat.apache.org



 -
 To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
 For additional commands, e-mail: users-h...@tomcat.apache.org



 -
 To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
 For additional commands, e-mail: users-h...@tomcat.apache.org

Re: Apache Tomcat, IIS, ISAPI

2009-09-25 Thread Josh Gooding

Joerg,

If you want your JSP's protected you can configure that in your web.xml
file.  I can't fully remember (someone jump in here) but I used role's in
tomcat and put the jsp files in a seperate directory, then I added
declarations in my web.xml file for it. Here is the snippets from my
web.xml:  (If you look down through the role-name tags certain roles can
access certain JSP's.  All the JSP's are locked from the gen-public inside
of a Private folder.

!--Constraints to control access to an entire namespace of urls--
security-constraint!--/admin/* limits access to those in admin
role.--
web-resource-collection
web-resource-nameCompanySecret/web-resource-name
url-pattern/admin/*/url-pattern
http-methodDELETE/http-method
http-methodGET/http-method
http-methodPOST/http-method
http-methodPUT/http-method
/web-resource-collection
auth-constraint
role-nameadmin/role-name
/auth-constraint
   /security-constraint

security-constraint!--/trainer/* limits access to those in trainer
role.--
web-resource-collection
web-resource-nameCompanySecret/web-resource-name
url-pattern/trainer/*/url-pattern
http-methodDELETE/http-method
http-methodGET/http-method
http-methodPOST/http-method
http-methodPUT/http-method
/web-resource-collection
auth-constraint
role-nametrainer/role-name
/auth-constraint
/security-constraint

security-constraint!--/Private/* completely blocks access except by
the webserver itself.--
web-resource-collection
web-resource-nameCompanySecret/web-resource-name
url-pattern/Private/*/url-pattern
http-methodDELETE/http-method
http-methodGET/http-method
http-methodPOST/http-method
http-methodPUT/http-method
/web-resource-collection
auth-constraint
role-name/role-name
/auth-constraint
/security-constraint

!--Constraints to limit access to individual urls which are not limited
by any namespace in their url--
security-constraint!--/Main.jsp requires login, but then is available
to everybody who can login.--
web-resource-collection
web-resource-nameCompanySecret/web-resource-name
url-pattern/Main.jsp/url-pattern
url-pattern/AutocompleteQuery/url-pattern
http-methodDELETE/http-method
http-methodGET/http-method
http-methodPOST/http-method
http-methodPUT/http-method
/web-resource-collection
auth-constraint
role-nametrainer/role-name
role-namemanager/role-name
role-nameadmin/role-name
role-nameuser/role-name
/auth-constraint
/security-constraint

security-constraint!--/CompanyOverview.jsp requires login, in any
non-user role.--
web-resource-collection
web-resource-nameCompanySecret/web-resource-name
url-pattern/CompanyOverview.jsp/url-pattern
http-methodDELETE/http-method
http-methodGET/http-method
http-methodPOST/http-method
http-methodPUT/http-method
/web-resource-collection
auth-constraint
role-nametrainer/role-name
role-namemanager/role-name
role-nameadmin/role-name
/auth-constraint
/security-constraint

Hope this helps you.

- Josh


On Fri, Sep 25, 2009 at 5:46 AM, Buchner, Joerg [T0I] (VW Sachsen) 
joerg.buch...@volkswagen.de wrote:

 Hi Rainer,

 yes, the JSP's should be served by Tomcat!

 But Tomcat should not work as WebService,
 Tomcat is only an Engine behind IIS.

 The protection of the JSP Files should be realized in the NTFS ACL's...

 Regards,
 Jörg




 -Ursprüngliche Nachricht-
 Von: Rainer Jung [mailto:rainer.j...@kippdata.de]
 Gesendet: Freitag, 25. September 2009 09:07
 An: Tomcat Users List
 Betreff: Re: Apache Tomcat, IIS, ISAPI

 On 25.09.2009 08:06, Buchner, Joerg [T0I] (VW Sachsen) wrote:
  Hi,
 
  i've got an Problem.
  We have an Microsoft IIS 6.0 with an Apache Tomcat 5.5.27 with IIS ISAPI
 1.2.28.
  (AMD64).
 
  The Website use Integrated Windows Authentication.
 
  The SourceCode Files on the server are protected by ActiveDirectory
 Groups.
 
  All works fine.
 
  But now the problem.
 
  If a user request a html File then the IIS check the NTFS
 Filesystempermissions.
  That means, only users have access, which are in the security group on
 the file system.
 
  But, if the same user request a jsp File, then there is no check on the
 Filesystempermissons.
 
  That is not OK.
  I'ld like to protect my JSP Files with NTFS Rights.
 
  Can anybody help me?

 The JSPs should be served by Tomcat, right? So IIS doesn't need to be
 able to see them or allowed to actually read them.

 So do you

RE: Clustering Question...

Thanks Chuck.  I'll relay your feedback to our development team.

To answer your question, the startup parameters are as follows;

-Xms4800m
-Xmx4800m
-XX:NewSize=900m
-XX:MaxNewSize=900m
-XX:PermSize=480m
-XX:MaxPermSize=480m
-XX:+UseParallelGC
-XX:ParallelGCThreads=3
-XX:+UseParallelOldGC
-XX:GCTimeRatio=11
-XX:+UseAdaptiveSizePolicy
-XX:ReservedCodeCacheSize=96m

-Original Message-
From: Caldarale, Charles R [mailto:chuck.caldar...@unisys.com] 
Sent: 25 September 2009 14:23
To: Tomcat Users List
Subject: RE: Clustering Question...

 From: Darren Kukulka [mailto:darren.kuku...@connaught.plc.uk]
 Subject: RE: Clustering Question...
 
 When OldGen becomes full, the system basically slows down to the point
 of becoming non-responsive...but it does not report any OOME.

Then a major GC likely is running almost continuously, which will
suspend the application threads.

What are your heap-related commmand line settings?  If you have
artificially constrained the size of the OldGen, you could easily have a
self-induced problem.  Setting -XX:NewRatio or any of several other
options will prevent HotSpot from dynamically adjusting the balance
between NewGen and OldGen.

Or, you may simply be out of heap space, in which case you'll need to
fix your application so that it's not so greedy.  Excessive caching
without employing a SoftReference technique is a frequent cause.

 - Chuck


THIS COMMUNICATION MAY CONTAIN CONFIDENTIAL AND/OR OTHERWISE PROPRIETARY
MATERIAL and is thus for use only by the intended recipient. If you
received this in error, please contact the sender and delete the e-mail
and its attachments from all computers.


-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org



Connaught plc is a FTSE 250 company. We are the UK's leading provider of 
integrated services operating in the compliance, social housing and public 
sector markets.

Please visit our website to see a full list of Connaught's Registered Companies 
www.connaught.plc.uk/group/aboutconnaught/registeredcompanies

Disclaimer:

The information transmitted is intended only for the person or entity to which 
it is addressed and may contain confidential and/or privileged material. Any 
review, retransmission, dissemination or other use of, or taking of any action 
in reliance upon, this information by persons or entities other than the 
intended recipient is prohibited. If you received this in error, please contact 
the sender and delete this message.

Connaught plc, Head Office 01392 444546

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

RE: Has anyone been successful building mod_jk on AIX 6.1 ?

2009-09-25 Thread Strickland, Lawrence P

I ran the find commands and did not find any of these files. I ran the
build with the new tools and this is what I got.

Could this one be my shared object module?
180225 1053 -rw-r--r--  1 root  system 1077292 Sep 25 08:40
./native/apache-2.0/.libs/libmod_jk.a


bash-3.00# pwd
/home/strickll/tomcat-connectors-1.2.28-new_autotools-src
bash-3.00# find . -name \*mod_jk\* -ls
176652  109 -rw---  1 51   develope111577 Mar 19  2009
./native/apache-1.3/mod_jk.c
1766538 -rw---  1 51   develope  7824 Jun 17  2007
./native/apache-1.3/mod_jk.dsp
1766551 -rw---  1 51   develope11 Sep  6  2002
./native/apache-1.3/mod_jk.exp
177030  126 -rw---  1 51   develope128597 Mar 19  2009
./native/apache-2.0/mod_jk.c
1770318 -rw---  1 51   develope  7757 Jun 17  2007
./native/apache-2.0/mod_jk.dsp
180225 1053 -rw-r--r--  1 root  system 1077292 Sep 25 08:40
./native/apache-2.0/.libs/libmod_jk.a
1802261 -rw-r--r--  1 root  system 756 Sep 25 08:40
./native/apache-2.0/.libs/mod_jk.lai
1802271 lrwxrwxrwx  1 root  system  12 Sep 25 08:40
./native/apache-2.0/.libs/mod_jk.la - ../mod_jk.la
1770541 -rw-r--r--  1 root  system 765 Sep 25 08:17
./native/apache-2.0/mod_jk.la.
177077 1053 -rw-r--r--  1 root  system 1077292 Sep 25 08:40
./native/apache-2.0/libmod_jk.a
177073  155 -rw-r--r--  1 root  system  158506 Sep 25 08:40
./native/apache-2.0/mod_jk.o
1770721 -rw-r--r--  1 root  system 297 Sep 25 08:40
./native/apache-2.0/mod_jk.lo
1770741 -rw-r--r--  1 root  system 756 Sep 25 08:40
./native/apache-2.0/mod_jk.la
bash-3.00# ./buildconf.sh
bash-3.00# pwd
bash-3.00#

-Original Message-
From: Rainer Jung [mailto:rainer.j...@kippdata.de] 
Sent: Thursday, September 24, 2009 6:06 PM
To: Tomcat Users List
Subject: Re: Has anyone been successful building mod_jk on AIX 6.1 ?

On 25.09.2009 00:55, Strickland, Lawrence P wrote:
 If the shared object module in not called mod_jk.so then what is it
called? Is it mod_jk.o?
 And, if that is the case where does it need to be? In the
apache/modules directory?

See the below find command. It could be mod_jk.so, mod_jk.a or
mod_jk.so.0. But please note what I also wrote below. If you use the
standard src download, even if the build works, it might not be a really
good binary, because the platform detection doesn't work with the old
config.guess and configure.

And yes: once you've got a good shared object file for AIX, you can put
it into the modules folder of httpd. Actually you can put it anywhere
you like, but if you use the relative path modules/... in the LoadModule
directive, then it has to be in the modules directory of httpd.

Regards,

Rainer


 
 From: Rainer Jung [mailto:rainer.j...@kippdata.de]
 Sent: Thu 9/24/2009 4:24 PM
 To: Tomcat Users List
 Subject: Re: Has anyone been successful building mod_jk on AIX 6.1 ?
 
 
 
 Hi Lawrence,
 
 On 24.09.2009 22:48, Strickland, Lawrence P wrote:
 This is the output of my build run.
 So apparently I am building version 1.2.28 with XLC 10.1 and Apache
 2.2.13

 #./httpd -v
 Server version: Apache/2.2.13 (Unix)
 Server built:   Sep 22 2009 13:27:47

 # gmake -v
 GNU Make 3.80

 /home/strickll/tomcat-connectors-1.2.28-src/native

 bash-3.00# CC=cc_r -qlanglvl=extc89 ./configure
 --with-apxs=/usr/local/apache2/bin/apxs
 
 ...
 
 Do the following commands find a build result:
 
 find /home/strickll/tomcat-connectors-1.2.28-src -name mod_jk.a -ls
 
 find /home/strickll/tomcat-connectors-1.2.28-src -name mod_jk.so.*
-ls
 
 I once did build for AIX but using an own toolchain based on gcc. One
of
 the things I had to fix, was that config.guess (and also configure
etc.)
 was to old to detect AIX 6. I checked 1.2.28 and unfortunately we
didn't
 use the updated autotools when doing the release. I ran buildconf.sh
 once more with newer autotools, so you might want to check whether the
 archive at
 

http://people.apache.org/~rjung/mod_jk-dev/source/jk-1.2.28-aix/tomcat-c
onnectors-1.2.28-new_autotools-src.tar.gz
 
 works better for you.
 
 But I vaguely also remember that there was also a problem with
libtool.
 I don't really remember precisely, but on AIX it wanted to build
shared
 libs as .a file instead of .so. It worked when applying the following
 ksh snipppet:
 
 
   for f in `find . -type f -name libtool`
   do
  echo Fixing $f...
  cp -p $f $f.aix_orig
  sed -e 's#hardcode_action=immediate#hardcode_action=relink#'
\
  -e 's#hardcode_direct=yes#hardcode_direct=no#' \
  -e 's#hardcode_minus_L=no#hardcode_minus_L=yes#' \
  $f.aix_orig  $f
   done
 
 but maybe you won't need it using XLC.
 
 Concerning make install: don't really use it. You can copy over the
 mod_jk.so (or however the shared object file is called on AIX) to the
 modules directory.
 
 Let us know your

RE: Clustering Question...

 From: Darren Kukulka [mailto:darren.kuku...@connaught.plc.uk]
 Subject: RE: Clustering Question...

 To answer your question, the startup parameters are as follows;
 -Xms4800m
 -Xmx4800m
 -XX:NewSize=900m
 -XX:MaxNewSize=900m
 -XX:PermSize=480m
 -XX:MaxPermSize=480m
 -XX:+UseParallelGC
 -XX:ParallelGCThreads=3
 -XX:+UseParallelOldGC
 -XX:GCTimeRatio=11
 -XX:+UseAdaptiveSizePolicy
 -XX:ReservedCodeCacheSize=96m

You might want to try removing the NewSize and MaxNewSize options.  The 
GCTimeRatio seems a bit low - the default is 99; the value of 11 will reduce 
the time available for the application threads.  UseAdaptiveSizePolicy is on by 
default in JRE/JDK 6 (but you didn't tell us what JVM you're using).

Do you have enough RAM to support that heap size without paging?  How many 
cores do you have?

Turning on PrintGC, PrintGCDetails, and PrintHeapAtGC might provide some useful 
info; use -Xloggc to direct the output to the file of choice.

Running a profiler (jmap or jhat would suffice) to find out what's actually 
taking up all of OldGen would help.

 - Chuck

THIS COMMUNICATION MAY CONTAIN CONFIDENTIAL AND/OR OTHERWISE PROPRIETARY 
MATERIAL and is thus for use only by the intended recipient. If you received 
this in error, please contact the sender and delete the e-mail and its 
attachments from all computers.

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: Virtual host configuration - best practise?

2009-09-25 Thread Pid


On 25/09/2009 14:03, Pid wrote:

On 25/09/2009 08:26, hepabolu wrote:



awarnier wrote:


hepabolu wrote:
...


However, from the Tomcat docs I understand that any change to
server.xml
requires a restart of Tomcat which would mean that the existing
(non-dummy)
hosts which are already in production, i.e. up and running, will
also be
taken offline. This is the heart of the problem...


Just as another look on the issue.

skip/

This may be one case where the added complication of having an Apache
httpd in front of Tomcat is justified.




Do you have any info on processing times using apache + mod_jk?


Your app will be the defining factor in terms of responsivity, there is
an overhead when using HTTPD in front of Tomcat but it's usually
negligible compared to the time spent in your application.


Ignore this below, it's an answer to another thread.
Teach me to rush some replies out... Doh.

p


OP: Chris asked you if you were using session replication, I'd also ask
if so, which type?

If your interest is in load balancing, you are using sticky sessions and
you are not too bothered about forcing re-login or session
re-initialisation when the client shifts node, you could disable session
replication.

A positive change in performance would confirm that session replication
is causing the slow down in the one node.


You didn't specify (did you?) which algorithm you are using to load
balance, have you experimented with the various options available?


p




Thanks.

Bye, Helma





-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: Any way to filter URL's with a dot in the directory name?

2009-09-25 Thread Christopher Schultz

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Scott,

On 9/24/2009 4:19 PM, Scott Dunbar wrote:
 No, I specified /* as my url-pattern:
 
 filter-mapping
filter-nameFileFilter/filter-name
url-pattern/*/url-pattern
 /filter-mapping
 
 and my filter is not called when the URL is
 http://www.hostname.tld/.svn/entries
 
 The servlet spec is weird here and the url-pattern in a
 web-resource-collection.  The . has special meaning that seems to
 indicate a file extension, not a directory.

The . only has special meaning when the URL pattern looks like this:

*.[anything]

All other uses of a period should be literal.

If you set up a mapping for /.svn then the matching should be exact.

Since I guess you want to protect /*/.svn then your approach of
mapping to /* and then just checking for the directory name is
probably the right one.

- -chris
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkq82ksACgkQ9CaO5/Lv0PAiGwCcC1wf++l16QB23G/a6QLqHvlj
yUkAnRATnhqF0cFxzO+DWso8L1uOjFpf
=4dkJ
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: Tomcat and Outgoing SSL

2009-09-25 Thread Christopher Schultz

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Steve,

On 9/23/2009 11:49 PM, Steve Cohen wrote:
 I have an backend application that runs under Tomcat.  It does not serve
 Web pages.  It depends on various services that use SSL in one way or
 another:
 
 1) It connects with a vendor's Web Service over https:, which depends on
 one of the certificates in the default cacerts file
 
 2) It connects with another vendor's Web Service over https: but this
 one depends on a CA certificate issued by the vendor.
 
 3) It makes SSL-encrypted connections to a MySQL database using a
 self-generated SSL certificate.
 
 I can get this to work by using keytool and importing the entire cacerts
 keystore, the self-generated CA cert for mysql, and the second vendor's
 ca cert into a single truststore, then Setting system properties to
 point at this at app startup.
 
 But this feels like a real hack.

So, you basically copy the system cacerts file and merge-in the two
other certificates? That doesn't sound too bad to me.

Another option is to simply modify the system cacerts file.

I thought that the JVM would load the system cacerts file plus
~/.cacerts or something similar automatically. Have you looked at the
documentation for SocketFactory and friends?

Another option is to simply turn-off certificate checking for SSL
connections, but I really don't recommend this except for testing
environments.

- -chris
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkq83xAACgkQ9CaO5/Lv0PBetgCfQKIyryVLRtOu3Mcr6Z2/Stnr
W5UAoLLPqNnbfHl9ZfcPwpDf2oLwEIWC
=SaIR
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

custom error page during client ssl

2009-09-25 Thread iam

Is there a way to throw a custom error page/msg when the user cancels the 
certificate pop-up on the browser during client SSL ?

Right now the browser just shows default network error.

Any prompt help is appreciated.

Thanks

Re: Has anyone been successful building mod_jk on AIX 6.1 ?

On 25.09.2009 15:46, Strickland, Lawrence P wrote:
 I ran the find commands and did not find any of these files. I ran the
 build with the new tools and this is what I got.
 
 Could this one be my shared object module?
 180225 1053 -rw-r--r--  1 root  system 1077292 Sep 25 08:40
 ./native/apache-2.0/.libs/libmod_jk.a

Yes, this could be it. It could either be a static library (not good) or
a shared library (good). Most platforms use *.a for static libs, but I
vaguely remember that AIX used it for shared.

You could try file native/apache-2.0/.libs/libmod_jk.a to see what AIX
says which type of file it is.

Did you try loading it into the web server with LoadModule?

For the build, did you use the new tarball and additionally the fixed
libtool, or only the new tarball?

Regards,

Rainer

 bash-3.00# pwd
 /home/strickll/tomcat-connectors-1.2.28-new_autotools-src
 bash-3.00# find . -name \*mod_jk\* -ls
 176652  109 -rw---  1 51   develope111577 Mar 19  2009
 ./native/apache-1.3/mod_jk.c
 1766538 -rw---  1 51   develope  7824 Jun 17  2007
 ./native/apache-1.3/mod_jk.dsp
 1766551 -rw---  1 51   develope11 Sep  6  2002
 ./native/apache-1.3/mod_jk.exp
 177030  126 -rw---  1 51   develope128597 Mar 19  2009
 ./native/apache-2.0/mod_jk.c
 1770318 -rw---  1 51   develope  7757 Jun 17  2007
 ./native/apache-2.0/mod_jk.dsp
 180225 1053 -rw-r--r--  1 root  system 1077292 Sep 25 08:40
 ./native/apache-2.0/.libs/libmod_jk.a
 1802261 -rw-r--r--  1 root  system 756 Sep 25 08:40
 ./native/apache-2.0/.libs/mod_jk.lai
 1802271 lrwxrwxrwx  1 root  system  12 Sep 25 08:40
 ./native/apache-2.0/.libs/mod_jk.la - ../mod_jk.la
 1770541 -rw-r--r--  1 root  system 765 Sep 25 08:17
 ./native/apache-2.0/mod_jk.la.
 177077 1053 -rw-r--r--  1 root  system 1077292 Sep 25 08:40
 ./native/apache-2.0/libmod_jk.a
 177073  155 -rw-r--r--  1 root  system  158506 Sep 25 08:40
 ./native/apache-2.0/mod_jk.o
 1770721 -rw-r--r--  1 root  system 297 Sep 25 08:40
 ./native/apache-2.0/mod_jk.lo
 1770741 -rw-r--r--  1 root  system 756 Sep 25 08:40
 ./native/apache-2.0/mod_jk.la
 bash-3.00# ./buildconf.sh
 bash-3.00# pwd
 bash-3.00#
 
 -Original Message-
 From: Rainer Jung [mailto:rainer.j...@kippdata.de] 
 Sent: Thursday, September 24, 2009 6:06 PM
 To: Tomcat Users List
 Subject: Re: Has anyone been successful building mod_jk on AIX 6.1 ?
 
 On 25.09.2009 00:55, Strickland, Lawrence P wrote:
 If the shared object module in not called mod_jk.so then what is it
 called? Is it mod_jk.o?
 And, if that is the case where does it need to be? In the
 apache/modules directory?
 
 See the below find command. It could be mod_jk.so, mod_jk.a or
 mod_jk.so.0. But please note what I also wrote below. If you use the
 standard src download, even if the build works, it might not be a really
 good binary, because the platform detection doesn't work with the old
 config.guess and configure.
 
 And yes: once you've got a good shared object file for AIX, you can put
 it into the modules folder of httpd. Actually you can put it anywhere
 you like, but if you use the relative path modules/... in the LoadModule
 directive, then it has to be in the modules directory of httpd.
 
 Regards,
 
 Rainer
 
 

 From: Rainer Jung [mailto:rainer.j...@kippdata.de]
 Sent: Thu 9/24/2009 4:24 PM
 To: Tomcat Users List
 Subject: Re: Has anyone been successful building mod_jk on AIX 6.1 ?



 Hi Lawrence,

 On 24.09.2009 22:48, Strickland, Lawrence P wrote:
 This is the output of my build run.
 So apparently I am building version 1.2.28 with XLC 10.1 and Apache
 2.2.13

 #./httpd -v
 Server version: Apache/2.2.13 (Unix)
 Server built:   Sep 22 2009 13:27:47

 # gmake -v
 GNU Make 3.80

 /home/strickll/tomcat-connectors-1.2.28-src/native

 bash-3.00# CC=cc_r -qlanglvl=extc89 ./configure
 --with-apxs=/usr/local/apache2/bin/apxs

 ...

 Do the following commands find a build result:

 find /home/strickll/tomcat-connectors-1.2.28-src -name mod_jk.a -ls

 find /home/strickll/tomcat-connectors-1.2.28-src -name mod_jk.so.*
 -ls

 I once did build for AIX but using an own toolchain based on gcc. One
 of
 the things I had to fix, was that config.guess (and also configure
 etc.)
 was to old to detect AIX 6. I checked 1.2.28 and unfortunately we
 didn't
 use the updated autotools when doing the release. I ran buildconf.sh
 once more with newer autotools, so you might want to check whether the
 archive at


 http://people.apache.org/~rjung/mod_jk-dev/source/jk-1.2.28-aix/tomcat-c
 onnectors-1.2.28-new_autotools-src.tar.gz

 works better for you.

 But I vaguely also remember that there was also a problem with
 libtool.
 I don't really remember precisely, but on AIX it wanted to build
 shared
 libs as .a file instead of .so. It worked when applying the following
 ksh snipppet:

RE: Tomcat 5.5.17 stop service fails resulting error 1053 on win server 2003

2009-09-25 Thread Deepti Nigudkar

Chris, All,

Thanks for your message. Yes, we have been using Tomcat for years now
but this version of tomcat always gives error when we stop it. The process
actually stops tomcat but the error message pops-up. If we start/stop service
from tomcat5w.exe it works fine. Its only when the service is stopped from
windows service, it always ends up in error message.

Even if we run it from DOS-prompt, it works fine.

The client wants it to be stopped from services only and didn't accept
any other solution.

My guess is something in registry that needs to be fixed.

This happens on all/several installations that use tomcat 5.5.17.

I think, probably, I should install another version of tomcat and then
compare registry to fix the issue.

Please let me know if you have any other ideas.

Thanks again.

Regards,

Deepti

Date: Thu, 24 Sep 2009 12:59:35 -0400
From: ch...@christopherschultz.net
To: users@tomcat.apache.org
Subject: Re: Tomcat 5.5.17 stop service fails resulting error 1053 on win
server 2003

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Deepti,

On 9/24/2009 12:03 PM, Deepti Nigudkar wrote:
We are using Tomcat 5.5.17 and everytime we try to stop the service
from windows server 2003 we get error 1053 (It seems Microsoft uses
this code if the service start/stop timeout)

That error code is unlikely to be useful... as you suggest, it may just
mean service failed to start.

Have you used Tomcat in the past and now it suddenly does not work? Or,
is this the first time you are attempting to deploy it as a Microsoft
Windows service?

I have tried to look for the solution to this as our software use
Apache tomcat 5.5.17. I have even tried editing catalina.bat to have
conditional statement to exclude JAVA_OPTS while stopping but it
didn't help. This was one of the solution I found when I was
searching on google.

If you have any log files in \path\to\tomcat\logs, delete them (or move
them somewhere else if you want to keep them) and try to start the
service again. Any log files in that directory should be examined. IF
you see anything out of the ordinary, investigate. If you need help,
post the relevant sections of the log files (or the whole thing if
you're not sure) to the list and we'll try to help.

What happens if you try to start Tomcat from the command-line, like this:

C:\ cd \path\to\tomcat

C:\path\to\tomcat bin\catalina.bat run

[output goes here]

Does anything happen? If you get any output at all, please post that
back to the list and we can maybe help figure out what's going wrong.

We cannot upgrade to other tomcat versions as our software was
developed on this version.

Tomcat 5.5.x versions are all very compatible with a few exceptions. IF
you read the changelog, you can see any changes that might break your
application. Some significant bugs and security problems have been fixed
between 5.5.17 and 5.5.28 (the current version of 5.5.x). You should
seriously consider upgrading no matter what else you do.

I am sure there is fix for this but I was unable to find it.

The fix depends on the problem. get back to us after you've tried my
suggestions above and we'll see what's going on.

- -chris
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.9 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkq7pXcACgkQ9CaO5/Lv0PBG2wCgnXUP4IlbQFyInr3JTizShreO
/XcAn2xURKdrju3laS4jWavXQQhdDd+a
=GWj3
-END PGP SIGNATURE-

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

_
Bing™ brings you maps, menus, and reviews organized in one place. Try it now.
http://www.bing.com/search?q=restaurantsform=MLOGENpubl=WLHMTAGcrea=TEXT_MLOGEN_Core_tagline_local_1x1

RE: Tomcat 5.5.17 stop service fails resulting error 1053 on win server 2003

2009-09-25 Thread George Sexton

One option would be to replace the tomcat.exe/tomcatw.exe with the current
version from 5.5.28. This wouldn't affect your application since it would
only replace the service interface, not the actual tomcat java class files.

George Sexton
MH Software, Inc.
http://www.mhsoftware.com/
Voice: 303 438 9585
 

 -Original Message-
 From: Deepti Nigudkar [mailto:deepti_...@hotmail.com]
 Sent: Friday, September 25, 2009 12:32 PM
 To: Apache Tomcat Users
 Subject: RE: Tomcat 5.5.17 stop service fails resulting error 1053 on
 win server 2003
 
 
 Chris, All,
 
 
 
Thanks for your message. Yes, we have been using Tomcat for
 years now but this version of tomcat always gives error when we stop
 it. The process actually stops tomcat but the error message pops-up. If
 we start/stop service from tomcat5w.exe it works fine.  Its only when
 the service is stopped from windows service, it always ends up in error
 message.
 
Even if we run it from DOS-prompt, it works fine.
 
The client wants it to be stopped from services only and didn't
 accept any other solution.
 
My guess is something in registry that needs to be fixed.
 
This happens on all/several installations that use tomcat
 5.5.17.
 
I think, probably, I should install another version of tomcat
 and then compare registry to fix the issue.
 
Please let me know if you have any other ideas.
 
 Thanks again.
 
 
 
 Regards,
 
 Deepti
 
  Date: Thu, 24 Sep 2009 12:59:35 -0400
  From: ch...@christopherschultz.net
  To: users@tomcat.apache.org
  Subject: Re: Tomcat 5.5.17 stop service fails resulting error 1053 on
 win server 2003
 
  -BEGIN PGP SIGNED MESSAGE-
  Hash: SHA1
 
  Deepti,
 
  On 9/24/2009 12:03 PM, Deepti Nigudkar wrote:
   We are using Tomcat 5.5.17 and everytime we try to stop the service
   from windows server 2003 we get error 1053 (It seems Microsoft uses
   this code if the service start/stop timeout)
 
  That error code is unlikely to be useful... as you suggest, it may
 just
  mean service failed to start.
 
  Have you used Tomcat in the past and now it suddenly does not work?
 Or,
  is this the first time you are attempting to deploy it as a Microsoft
  Windows service?
 
   I have tried to look for the solution to this as our software use
   Apache tomcat 5.5.17. I have even tried editing catalina.bat to
 have
   conditional statement to exclude JAVA_OPTS while stopping but it
   didn't help. This was one of the solution I found when I was
   searching on google.
 
  If you have any log files in \path\to\tomcat\logs, delete them (or
 move
  them somewhere else if you want to keep them) and try to start the
  service again. Any log files in that directory should be examined. IF
  you see anything out of the ordinary, investigate. If you need help,
  post the relevant sections of the log files (or the whole thing if
  you're not sure) to the list and we'll try to help.
 
  What happens if you try to start Tomcat from the command-line, like
 this:
 
  C:\ cd \path\to\tomcat
 
  C:\path\to\tomcat bin\catalina.bat run
 
  [output goes here]
 
  Does anything happen? If you get any output at all, please post that
  back to the list and we can maybe help figure out what's going wrong.
 
 
   We cannot upgrade to other tomcat versions as our software was
   developed on this version.
 
  Tomcat 5.5.x versions are all very compatible with a few exceptions.
 IF
  you read the changelog, you can see any changes that might break your
  application. Some significant bugs and security problems have been
 fixed
  between 5.5.17 and 5.5.28 (the current version of 5.5.x). You should
  seriously consider upgrading no matter what else you do.
 
   I am sure there is fix for this but I was unable to find it.
 
  The fix depends on the problem. get back to us after you've tried
 my
  suggestions above and we'll see what's going on.
 
  - -chris
  -BEGIN PGP SIGNATURE-
  Version: GnuPG v1.4.9 (MingW32)
  Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
 
  iEYEARECAAYFAkq7pXcACgkQ9CaO5/Lv0PBG2wCgnXUP4IlbQFyInr3JTizShreO
  /XcAn2xURKdrju3laS4jWavXQQhdDd+a
  =GWj3
  -END PGP SIGNATURE-
 
  -
  To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
  For additional commands, e-mail: users-h...@tomcat.apache.org
 
 
 _
 BingT  brings you maps, menus, and reviews organized in one place.
 Try it now.
 http://www.bing.com/search?q=restaurantsform=MLOGENpubl=WLHMTAGcrea=
 TEXT_MLOGEN_Core_tagline_local_1x1


-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Using administration url in application

2009-09-25 Thread Zeitlin, Michael (USAEO)

Hello,

I have configured a Tomcat 5.5 server on and deployed my application
onto in the root context.   The application works successfully,  except
when I try to use anything with that starts with the  /admin url
pattern.  My guess is that Tomcat thinks I am trying to reach the Web
Administration module and it throws up a 404 error.

   I would like to tell tomcat to use a different url for its web
administration module so I can use it on my application.  I have tried
going into my admin.xml file and modifying the Context entry to use a
different path:

Context path=/admin1
docBase=/usr/share/tomcat5.5/server/webapps/admin 
 debug=0 privileged=true allowLinking=true

   I have also tried totally removing the admin.xml file.  Neither has
worked.  

  Any help would be appreciated

Michael

RE: Tomcat 5.5.17 stop service fails resulting error 1053 on win server 2003

2009-09-25 Thread Deepti Nigudkar


Thanks George. I will try that on out testing environment.

 

Regards,

Deepti
 
 From: geor...@mhsoftware.com
 To: users@tomcat.apache.org
 Subject: RE: Tomcat 5.5.17 stop service fails resulting error 1053 on win 
 server 2003
 Date: Fri, 25 Sep 2009 12:40:19 -0600
 
 One option would be to replace the tomcat.exe/tomcatw.exe with the current
 version from 5.5.28. This wouldn't affect your application since it would
 only replace the service interface, not the actual tomcat java class files.
 
 George Sexton
 MH Software, Inc.
 http://www.mhsoftware.com/
 Voice: 303 438 9585
 
 
  -Original Message-
  From: Deepti Nigudkar [mailto:deepti_...@hotmail.com]
  Sent: Friday, September 25, 2009 12:32 PM
  To: Apache Tomcat Users
  Subject: RE: Tomcat 5.5.17 stop service fails resulting error 1053 on
  win server 2003
  
  
  Chris, All,
  
  
  
  Thanks for your message. Yes, we have been using Tomcat for
  years now but this version of tomcat always gives error when we stop
  it. The process actually stops tomcat but the error message pops-up. If
  we start/stop service from tomcat5w.exe it works fine. Its only when
  the service is stopped from windows service, it always ends up in error
  message.
  
  Even if we run it from DOS-prompt, it works fine.
  
  The client wants it to be stopped from services only and didn't
  accept any other solution.
  
  My guess is something in registry that needs to be fixed.
  
  This happens on all/several installations that use tomcat
  5.5.17.
  
  I think, probably, I should install another version of tomcat
  and then compare registry to fix the issue.
  
  Please let me know if you have any other ideas.
  
  Thanks again.
  
  
  
  Regards,
  
  Deepti
  
   Date: Thu, 24 Sep 2009 12:59:35 -0400
   From: ch...@christopherschultz.net
   To: users@tomcat.apache.org
   Subject: Re: Tomcat 5.5.17 stop service fails resulting error 1053 on
  win server 2003
  
   -BEGIN PGP SIGNED MESSAGE-
   Hash: SHA1
  
   Deepti,
  
   On 9/24/2009 12:03 PM, Deepti Nigudkar wrote:
We are using Tomcat 5.5.17 and everytime we try to stop the service
from windows server 2003 we get error 1053 (It seems Microsoft uses
this code if the service start/stop timeout)
  
   That error code is unlikely to be useful... as you suggest, it may
  just
   mean service failed to start.
  
   Have you used Tomcat in the past and now it suddenly does not work?
  Or,
   is this the first time you are attempting to deploy it as a Microsoft
   Windows service?
  
I have tried to look for the solution to this as our software use
Apache tomcat 5.5.17. I have even tried editing catalina.bat to
  have
conditional statement to exclude JAVA_OPTS while stopping but it
didn't help. This was one of the solution I found when I was
searching on google.
  
   If you have any log files in \path\to\tomcat\logs, delete them (or
  move
   them somewhere else if you want to keep them) and try to start the
   service again. Any log files in that directory should be examined. IF
   you see anything out of the ordinary, investigate. If you need help,
   post the relevant sections of the log files (or the whole thing if
   you're not sure) to the list and we'll try to help.
  
   What happens if you try to start Tomcat from the command-line, like
  this:
  
   C:\ cd \path\to\tomcat
  
   C:\path\to\tomcat bin\catalina.bat run
  
   [output goes here]
  
   Does anything happen? If you get any output at all, please post that
   back to the list and we can maybe help figure out what's going wrong.
  
  
We cannot upgrade to other tomcat versions as our software was
developed on this version.
  
   Tomcat 5.5.x versions are all very compatible with a few exceptions.
  IF
   you read the changelog, you can see any changes that might break your
   application. Some significant bugs and security problems have been
  fixed
   between 5.5.17 and 5.5.28 (the current version of 5.5.x). You should
   seriously consider upgrading no matter what else you do.
  
I am sure there is fix for this but I was unable to find it.
  
   The fix depends on the problem. get back to us after you've tried
  my
   suggestions above and we'll see what's going on.
  
   - -chris
   -BEGIN PGP SIGNATURE-
   Version: GnuPG v1.4.9 (MingW32)
   Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
  
   iEYEARECAAYFAkq7pXcACgkQ9CaO5/Lv0PBG2wCgnXUP4IlbQFyInr3JTizShreO
   /XcAn2xURKdrju3laS4jWavXQQhdDd+a
   =GWj3
   -END PGP SIGNATURE-
  
   -
   To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
   For additional commands, e-mail: users-h...@tomcat.apache.org
  
  
  _
  BingT brings you maps, menus, and reviews organized in one place.
  Try it now.
  http://www.bing.com/search?q=restaurantsform=MLOGENpubl=WLHMTAGcrea=

RE: Using administration url in application

 From: Zeitlin, Michael (USAEO) [mailto:michael.zeit...@usdoj.gov]
 Subject: Using administration url in application

 I have configured a Tomcat 5.5 server on and deployed my
 application onto in the root context.

Exactly how did you do this?  The proper way is to remove the existing 
webapps/ROOT directory, and install your webapp in its place.  Any other 
procedure is probably wrong.

 I have tried going into my admin.xml file and modifying the 
 Context entry to use a different path:
 Context path=/admin1
 docBase=/usr/share/tomcat5.5/server/webapps/admin
  debug=0 privileged=true allowLinking=true

Pointless, since the path attribute is not allowed here.  You need to rename 
admin.xml to [somethingElse].xml if you want to use admin inside your default 
webapp.

 I have also tried totally removing the admin.xml file.

This makes me think you have not installed your default webapp properly.

 - Chuck

THIS COMMUNICATION MAY CONTAIN CONFIDENTIAL AND/OR OTHERWISE PROPRIETARY 
MATERIAL and is thus for use only by the intended recipient. If you received 
this in error, please contact the sender and delete the e-mail and its 
attachments from all computers.

-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

Re: custom error page during client ssl

2009-09-25 Thread Jason Brittain

Which web browser in particular are you asking about, and what's the network
error it's saying?  And, what did the pop-up say?

My guess is that this is all client-side, and browser specific, so there is
likely no way to customize the message.

--
Jason Brittain


On Fri, Sep 25, 2009 at 8:58 AM, iam techat...@yahoo.com wrote:

 Is there a way to throw a custom error page/msg when the user cancels the
 certificate pop-up on the browser during client SSL ?

 Right now the browser just shows default network error.

 Any prompt help is appreciated.

 Thanks

Re: JNDI authentication

2009-09-25 Thread herbie324


We are having the same exact issue you are speaking of.  We've gone through
logs as well.  I'd love to get some insight on this also. We are using
tomcat with basic authentication. 

daniel steel wrote:
 
 All,
   we are using jndi authentication  for authenticating the user against
 active directory. 
 the user is hitting the load balancer and then the request is forwarded to
 the tomcat.
 
  at times we are seeing slowness and on analyzing the access log, the jsps
 having high execution duration do not have any remoteusername associated
 with that call and has http response 401 - unauthorised access.
 
 the question is what stripping the remote username ?
 
 1. the load balancer is off loading the SSL.
 2. we are using basic authentication schema tied into JNDI - LDAP
 authentication.
 
 thanks
 dan
 
 
 
   
 

-- 
View this message in context: 
http://www.nabble.com/JNDI-authentication-tp25588982p25617863.html
Sent from the Tomcat - User mailing list archive at Nabble.com.


-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

RE: Has anyone been successful building mod_jk on AIX 6.1 ?

2009-09-25 Thread Strickland, Lawrence P

bash-3.00# file libmod_jk.a
libmod_jk.a: archive (big format)

What is this file? What am I looking for?

For the build, did you use the new tarball and additionally the fixed
libtool, or only the new tarball?

Yes to both of these questions. I used the new tarball and I ran the
fixed libtool.





-Original Message-
From: Rainer Jung [mailto:rainer.j...@kippdata.de] 
Sent: Friday, September 25, 2009 1:07 PM
To: Tomcat Users List
Subject: Re: Has anyone been successful building mod_jk on AIX 6.1 ?

On 25.09.2009 15:46, Strickland, Lawrence P wrote:
 I ran the find commands and did not find any of these files. I ran the
 build with the new tools and this is what I got.
 
 Could this one be my shared object module?
 180225 1053 -rw-r--r--  1 root  system 1077292 Sep 25 08:40
 ./native/apache-2.0/.libs/libmod_jk.a

Yes, this could be it. It could either be a static library (not good) or
a shared library (good). Most platforms use *.a for static libs, but I
vaguely remember that AIX used it for shared.

You could try file native/apache-2.0/.libs/libmod_jk.a to see what AIX
says which type of file it is.

Did you try loading it into the web server with LoadModule?

For the build, did you use the new tarball and additionally the fixed
libtool, or only the new tarball?

Regards,

Rainer

 bash-3.00# pwd
 /home/strickll/tomcat-connectors-1.2.28-new_autotools-src
 bash-3.00# find . -name \*mod_jk\* -ls
 176652  109 -rw---  1 51   develope111577 Mar 19  2009
 ./native/apache-1.3/mod_jk.c
 1766538 -rw---  1 51   develope  7824 Jun 17  2007
 ./native/apache-1.3/mod_jk.dsp
 1766551 -rw---  1 51   develope11 Sep  6  2002
 ./native/apache-1.3/mod_jk.exp
 177030  126 -rw---  1 51   develope128597 Mar 19  2009
 ./native/apache-2.0/mod_jk.c
 1770318 -rw---  1 51   develope  7757 Jun 17  2007
 ./native/apache-2.0/mod_jk.dsp
 180225 1053 -rw-r--r--  1 root  system 1077292 Sep 25 08:40
 ./native/apache-2.0/.libs/libmod_jk.a
 1802261 -rw-r--r--  1 root  system 756 Sep 25 08:40
 ./native/apache-2.0/.libs/mod_jk.lai
 1802271 lrwxrwxrwx  1 root  system  12 Sep 25 08:40
 ./native/apache-2.0/.libs/mod_jk.la - ../mod_jk.la
 1770541 -rw-r--r--  1 root  system 765 Sep 25 08:17
 ./native/apache-2.0/mod_jk.la.
 177077 1053 -rw-r--r--  1 root  system 1077292 Sep 25 08:40
 ./native/apache-2.0/libmod_jk.a
 177073  155 -rw-r--r--  1 root  system  158506 Sep 25 08:40
 ./native/apache-2.0/mod_jk.o
 1770721 -rw-r--r--  1 root  system 297 Sep 25 08:40
 ./native/apache-2.0/mod_jk.lo
 1770741 -rw-r--r--  1 root  system 756 Sep 25 08:40
 ./native/apache-2.0/mod_jk.la
 bash-3.00# ./buildconf.sh
 bash-3.00# pwd
 bash-3.00#
 
 -Original Message-
 From: Rainer Jung [mailto:rainer.j...@kippdata.de] 
 Sent: Thursday, September 24, 2009 6:06 PM
 To: Tomcat Users List
 Subject: Re: Has anyone been successful building mod_jk on AIX 6.1 ?
 
 On 25.09.2009 00:55, Strickland, Lawrence P wrote:
 If the shared object module in not called mod_jk.so then what is it
 called? Is it mod_jk.o?
 And, if that is the case where does it need to be? In the
 apache/modules directory?
 
 See the below find command. It could be mod_jk.so, mod_jk.a or
 mod_jk.so.0. But please note what I also wrote below. If you use the
 standard src download, even if the build works, it might not be a
really
 good binary, because the platform detection doesn't work with the old
 config.guess and configure.
 
 And yes: once you've got a good shared object file for AIX, you can
put
 it into the modules folder of httpd. Actually you can put it anywhere
 you like, but if you use the relative path modules/... in the
LoadModule
 directive, then it has to be in the modules directory of httpd.
 
 Regards,
 
 Rainer
 
 

 From: Rainer Jung [mailto:rainer.j...@kippdata.de]
 Sent: Thu 9/24/2009 4:24 PM
 To: Tomcat Users List
 Subject: Re: Has anyone been successful building mod_jk on AIX 6.1 ?



 Hi Lawrence,

 On 24.09.2009 22:48, Strickland, Lawrence P wrote:
 This is the output of my build run.
 So apparently I am building version 1.2.28 with XLC 10.1 and Apache
 2.2.13

 #./httpd -v
 Server version: Apache/2.2.13 (Unix)
 Server built:   Sep 22 2009 13:27:47

 # gmake -v
 GNU Make 3.80

 /home/strickll/tomcat-connectors-1.2.28-src/native

 bash-3.00# CC=cc_r -qlanglvl=extc89 ./configure
 --with-apxs=/usr/local/apache2/bin/apxs

 ...

 Do the following commands find a build result:

 find /home/strickll/tomcat-connectors-1.2.28-src -name mod_jk.a -ls

 find /home/strickll/tomcat-connectors-1.2.28-src -name mod_jk.so.*
 -ls

 I once did build for AIX but using an own toolchain based on gcc. One
 of
 the things I had to fix, was that config.guess (and also configure
 etc.)
 was to old to detect AIX 6. I checked 1.2.28 and unfortunately we
 didn't
 use the updated

RE: Using administration url in application

2009-09-25 Thread Zeitlin, Michael (USAEO)

Sorry,  I should have been more clear.  I did remove my existing ROOT
directory and replaced it with my application.

If the path parameter in the Context is not useful,  how does Tomcat
know what url to use for the administration app?  Will renaming the
admin.xml to [somethingElse].xml now display the admin application at
the /[somethingElse] url?  



-Original Message-
From: Caldarale, Charles R [mailto:chuck.caldar...@unisys.com] 
Sent: Friday, September 25, 2009 3:10 PM
To: Tomcat Users List
Subject: RE: Using administration url in application

 From: Zeitlin, Michael (USAEO) [mailto:michael.zeit...@usdoj.gov]
 Subject: Using administration url in application
 
 I have configured a Tomcat 5.5 server on and deployed my
 application onto in the root context.

Exactly how did you do this?  The proper way is to remove the existing
webapps/ROOT directory, and install your webapp in its place.  Any other
procedure is probably wrong.

 I have tried going into my admin.xml file and modifying the 
 Context entry to use a different path:
 Context path=/admin1
 docBase=/usr/share/tomcat5.5/server/webapps/admin
  debug=0 privileged=true allowLinking=true

Pointless, since the path attribute is not allowed here.  You need to
rename admin.xml to [somethingElse].xml if you want to use admin inside
your default webapp.

 I have also tried totally removing the admin.xml file.

This makes me think you have not installed your default webapp properly.

 - Chuck


THIS COMMUNICATION MAY CONTAIN CONFIDENTIAL AND/OR OTHERWISE PROPRIETARY
MATERIAL and is thus for use only by the intended recipient. If you
received this in error, please contact the sender and delete the e-mail
and its attachments from all computers.


-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org


-
To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org
For additional commands, e-mail: users-h...@tomcat.apache.org

RE: Using administration url in application