IPlanet - WebLogic bridge, performance

We are using iPlanet as a web server and weblogic as an application server. So, there is NSAPI plug-in,redirecting some requests coming from iPlanet to WebLogic. JSPs pages and servlets are served very fast via this plug-in, whereas static html pages and images are served very slow. Response time (in case of html pages and images) is very long. The weblogic.log file does not give us any errors, whereas /https-.../logs/errors file gives the following:
"flex log buffer overflow" - greater than 4096 characters
What can we do to decrease response time?
Actually we walked through parameters that can be passed into NSAPI plug-in, and we could not find anything in this plug-in related to buffer size.
Attached file is a copy of our obj.conf file
[obj.conf]

We were using the same configuration, and we got the same problems
And we got the solution :
you have to use the libproxy.so from the WL 51 SP8 , using this heavy
bugfixed library
the problems didn't occur.
Sasha Kretchetov wrote:
We are using iPlanet as a web server and weblogic as an application server. So, there is NSAPI plug-in,redirecting some requests coming from iPlanet to WebLogic. JSPs pages and servlets are served very fast via this plug-in, whereas static html pages and images are served very slow. Response time (in case of html pages and images) is very long. The weblogic.log file does not give us any errors, whereas /https-.../logs/errors file gives the following:
"flex log buffer overflow" - greater than 4096 characters
What can we do to decrease response time?
Actually we walked through parameters that can be passed into NSAPI plug-in, and we could not find anything in this plug-in related to buffer size.
Attached file is a copy of our obj.conf file
# Sun Netscape Alliance - obj.conf
# You can edit this file, but comments and formatting changes
# might be lost when the admin server makes changes.
Init fn="load-modules" funcs="wl-proxy,wl-init" shlib="/opt/iplanet/server4/plugins/libproxy.so"
Init fn="wl-init"
Init fn="flex-init" access="/opt/iplanet/server4/https-1212/logs/access" 3="fo" rmat.access="%Ses->client.ip% - %Req->vars.auth-user% [%SYSDATE%] \"%Req->reqpb.clf-request%\" %Req->srvhdrs.clf-status% %Req->srvhdrs.content-length%"
Init fn="load-types" mime-types="mime.types"
Init fn="load-modules" shlib="/opt/iplanet/server4/bin/https/lib/libNSServletPlugin.so" funcs="NSServletEarlyInit,NSServletLateInit,NSServletNameTrans,NSServletService" shlib_flags="(global|now)"
Init fn="NSServletEarlyInit" EarlyInit="yes"
Init fn="NSServletLateInit" LateInit="yes"
<Object name="default">
NameTrans fn="NSServletNameTrans" name="servlet"
NameTrans fn="pfx2dir" from="/servlet" dir="/opt/iplanet/server4/1212/servlet" name="ServletByExt"
NameTrans fn="pfx2dir" from="/ns-icons" dir="/opt/iplanet/server4/ns-icons" name="es-internal"
NameTrans fn="pfx2dir" from="/mc-icons" dir="/opt/iplanet/server4/ns-icons" name="es-internal"
NameTrans fn="pfx2dir" from="/help" dir="/opt/iplanet/server4/manual/https/ug" name="es-internal"
NameTrans fn="pfx2dir" from="/manual" dir="/opt/iplanet/server4/manual/https" name="es-internal"
NameTrans fn="document-root" root="/opt/iplanet/server4/1212"
PathCheck fn="unix-uri-clean"
PathCheck fn="check-acl" acl="default"
PathCheck fn="find-pathinfo"
PathCheck fn="find-index" index-names="index.html,home.html,index.htm"
ObjectType fn="type-by-extension"
ObjectType fn="force-type" type="text/plain"
Service method="(GET|HEAD)" type="magnus-internal/imagemap" fn="imagemap"
Service method="(GET|HEAD)" type="magnus-internal/directory" fn="index-common"
Service method="(GET|HEAD|POST)" type="*~magnus-internal/*" fn="send-file"
AddLog fn="flex-log" name="access"
</Object>
<Object name="cgi">
ObjectType fn="force-type" type="magnus-internal/cgi"
Service fn="send-cgi"
</Object>
<Object name="weblogic" ppath="*/Application1/*">
Service fn="wl-proxy" WebLogicHost="192.168.170.12" WebLogicPort="55205"
</Object>
<Object name="weblogic" ppath="*/Application2/*">
Service fn="wl-proxy" WebLogicHost="192.168.170.12" WebLogicPort="55205"
</Object>
<Object name="servlet">
ObjectType fn="force-type" type="text/html"
Service fn="NSServletService"
</Object>
<Object name="jsp092">
ObjectType fn="type-by-extension"
ObjectType fn="change-type" type="magnus-internal/jsp092" if-type="magnus-internal/jsp"
Service fn="NSServletService" type="magnus-internal/jsp092"
</Object>
<Object name="ServletByExt">
ObjectType fn="force-type" type="magnus-internal/servlet"
Service type="magnus-internal/servlet" fn="NSServletService"
</Object>
<Object name="es-internal">
PathCheck fn="check-acl" acl="es-internal"
</Object>
obj.conf
Content-Type:
text/plain
Content-Encoding:
base64

Similar Messages

The iPlanet-WebLogic Server 5.1 on Solaris

Does the iPlanet-WebLogic Server plug-in is supported on Solaris platform?.


Hi Matt,
          How do you like it? Is it very stable? I had heard complaints on the
          earlier iPlanet releases (mainly on NT), but I assumed that they were
          probably working better on Solaris. What does performance look like? Did
          you use Netscape server before, and if so, are there any particular
          differences?
          Thanks,
          Cameron Purdy
          Tangosol, Inc.
          http://www.tangosol.com
          +1.617.623.5782
          WebLogic Consulting Available
          "Matt Potter" <[email protected]> wrote in message
          news:3a819419$[email protected]..
          > Trang,
          >
          > It does. I'm using iPlanet 4.1 w/WLS 5.10 sp6 on Solaris 2.6. Check the
          > platform notes at BEA http://www.weblogic.com/platforms/index.html#plugin
          > for details. -Matt
          >
          > Trang wrote in message <3a80ad07$[email protected]>...
          > >Does the iPlanet-WebLogic Server plug-in is supported on Solaris
          platform?.
          >
          >

IPlanet & Weblogic HTTP Tunnelling support

My scenerio involves the following:
- 1 java client
- 1 iplanet server forwarding requests to the weblogic server
- 1 weblogic server 6.1 with HTTP tunnelling enabled
When I get the java client to talk to the iPlanet server, the
performance is very poor and exceptions get thrown on the java client
side like so:
<Error> <HTTPClientJVMConnection> < java.net.ProtocolException:
Tunneling result not OK, result: 'DEAD', id: '195'      at
weblogic.rjvm.http.HTTPClientJVMConnection.receiveAndDispatch(HTTPClientJVMConnection.java:406)
     at weblogic.rjvm.http.HTTPClientJVMConnection.execute(HTTPClientJVMConnection.java:296)
     at weblogic.kernel.ExecuteThread.execute(ExecuteThread.java:139)      at
weblogic.kernel.ExecuteThread.run(ExecuteThread.java:120) >
Now, when i change around the client end to talk directly to the
weblogic server, everything works fine, and the tunneling bottleneck
in performance is not even noticible.
So I have come to the conclusion that iPlanet is causing these
problems. I've tried to setup the obj.conf within iPlanet using the
NSAPI-Plugin to forward tunnelling requests, but it seems to make no
difference. Here is a snip of the obj.conf file from iplanet:
<Object name="tunnel" ppath="*/HTTPClnt*">
Service fn=wl_proxy WebLogicHost=192.168.1.4 WebLogicPort=7056
</Object>
Anyone have any experience with this issue?
Thanks,
Travis

Hi,
There is patch for the webserver(Apache Plugin),thats needs to be downloaded from Metalink,dont use the default plugin that comes with the weblogic 10.3 installtion.
Regards,
Dinesh

Weblogic Bridge "failed and will reconnect later"

The enterprise architecture is as follows:
1 Physical machine where I have 2 VMs on this machine, and each VM I have 1 domain, 1 Cluster and 1 managed server, and the second VM I have the replicated what I have in first VM.
In first domain I have the OSB, in the second domain I have SOA.
What I need ? Send information of queues (topics) of First Domain (OSB) to second domain(SOA).
At first the only thing I did was:
In the first domain I created OSB JMS Server, JMS Module, Connection Factory and Queue Topic.
In the second domain SOA I replied the same thing but with different names.
Then I created the bridge with configuration default between the first domain and the second domain, the bridge and when active it generates me a message Weblogic Bridge "failed and will reconnect later"
I did a test on my local machine windows where I created two domains, and the same procedures you did above, but the point was funcinou been activated now for the enterprise environment I'm not managing to make it work, you could tell me what is wrong . in both cases restarted the weblogic server as the machine itself the question that ran on my machine and machine company no.
If you can give me a light thank.

This is a bug (6973877, derby only) which has been fixed in 4.5 build14 and 4.4u2p1. Thanks for reporting. The problem is the missing "FOR BIT DATA" in LOG_RECORD column in derby table definition MQTMLRJMSBG41 in broker default.properties file
- LOG_RECORD VARCHAR(2048) NOT NULL,\
+ LOG_RECORD VARCHAR(2048) FOR BIT DATA NOT NULL,\

Apache-Weblogic Bridge does not work?

I've gone thru' the documentation regarding setting up of apache-weblogic bridge, but still it does not work. Does anyboday have any more details regarding this...EnvironmentOS - Solars 5.7Weblogic - 5.1 SP 8Apache 1.3.12

I've given up and had to go back to running "Double NAT" which also reports as a "problem" within the AEBS, but I just "ignore" it so the light will always be green.
It still ***** though, as "Double NAT" is also a reason for "Back to my Mac" not working properly, but how the ** am I supposed to avoid Double NAT when the wireless will not work in bridged mode?!

Aborting uploads through Apache/Weblogic bridge

Hi,
I have an application to which people should be able to upload files,
however only of a certain size. So if I detect that the file is too
large (based on Content-Length), I return an errorpage and don't read
the data from the browser. This is allowed according to rfc 1867 (which,
though it is obsoleted, is the one the HTTP1.1 rfc refers to on the
matter - also the obsoleting rfc doesn't say anything on the matter) and
all browsers seems to handle this.
However, once I go through the Apache/Weblogic bridge the behaviour
changes. When the browser sends the POST request to the bridge, the
bridge relays it on to the Weblogic server and my code. I immediately
determine that the file is too large, and returns the previously
mentioned errorpage. But, the bridge doesn't stop recieving the file
from the browser until it has got the entire file and then it returns a
standard bridge POST timeout error (and the timeout has definately not
been reached).
Any ideas?
CAPUT A/S Morten Jensen Phone +45 70 12 24 42
Nygade 6 Senior Developer Fax +45 70 11 24 42
DK-1164 Kbh K [email protected] http://www.caput.com

Add the following attributes:
<Location /re >
WLProxySSL ON
EnforceBasicConstraints false
</Location>
Original example in [http://m-button.blogspot.com/2008/11/how-to-configure-weblogic-to-use-ssl.html]
I hope this helps.

Problem with IIS- WebLogic bridge

Hi!
We have configured the IIS-WebLogic bridge for WebLogic 6.0 (SP1), with IIS
4.0 on two separate machines. This works on URLs which directly refer to
pages lying on the same machine as WebLogic, but some links which refer to
WebLogic properties file and then determines the page to be displayed (e.g.
include files). These work when the exact URL (including the IP Address and
port no. of the WebLogic Server) is pasted in the browser address window,
but fail when the proxied hyperlink is clicked from another page, or the
proxied URL is pasted in the address window.
Can anyone help. This is slightly urgent.
thanks and regards,
Prajesh
Prajesh K. Verma
Eclaton eBDU India Pvt. Ltd.
Huis van Eclaton
25, Lavelle Road
Bangalore 560 001
Tel: +91 80 2996569
Email: [email protected]
http://www.eclaton.com

It looks like you have the sendRedirect() problem.
We will convert the relative uri into Absolute URL in the sendRedirect().
the logic looks something like:
if (url is an absolute url) { //i.e: http://host:port/dummy.jsp
response.setHeader("Location", url);
sendError("302");
} else if (url is an absolute uri) { //i.e: /dummy.jsp
//prepend protocol, host and port info
} else if (url is a relative uri) { //i.e: dummy.jsp
//cacluate the context path based on the current request
//i.e: you are request http://iis:7001/include/index.jsp
PathTrim=/weblogic is set in the iisproxy.ini
//so the Location header looks like:
//http://iis:7001/weblogic/include/dummy.jsp
As you can see you may need to specify the redirected uri
to fall into the rules of "proxy by path" or "proxy by mime type" if you want
the iis to proxy the request properly.
"Prajesh K. Verma" <[email protected]> wrote:
Hi!
We have configured the IIS-WebLogic bridge for WebLogic 6.0 (SP1), with
IIS
4.0 on two separate machines. This works on URLs which directly refer
to
pages lying on the same machine as WebLogic, but some links which refer
to
WebLogic properties file and then determines the page to be displayed
(e.g.
include files). These work when the exact URL (including the IP Address
and
port no. of the WebLogic Server) is pasted in the browser address window,
but fail when the proxied hyperlink is clicked from another page, or
the
proxied URL is pasted in the address window.
Can anyone help. This is slightly urgent.
thanks and regards,
Prajesh
Prajesh K. Verma
Eclaton eBDU India Pvt. Ltd.
Huis van Eclaton
25, Lavelle Road
Bangalore 560 001
Tel: +91 80 2996569
Email: [email protected]
http://www.eclaton.com

VAJ3.5.3 Weblogic Bridge problems SP10?

Hi,
There seems to be some progressive memory leak in Visual-Age when looking up for
EJB on Weblogic, anybody faced such problems?
AND ...
In VAJ3.5.3 Weblogic Bridge documentation ... The docs/userguide.pdf says that
(In service-PAck for Integration-Kit) Import VAJ35weblogic510spX.zip where 'X'
is the service-pack number BUT...I don't find VAJ35weblogic510sp10.zip Anywhere
on BEA site. Please let me know where to get it from if it is ever available.
How do you upgrade the bridge to SP10?
Thanks, Harish.

Harish,
SP10 for the VAJ I-Kit is available at -
http://commerce.bea.com/downloads/weblogic_server_tools.jsp.
Thanks,
Nirav.
Harish SK wrote:
Hi,
There seems to be some progressive memory leak in Visual-Age when looking up for
EJB on Weblogic, anybody faced such problems?
AND ...
In VAJ3.5.3 Weblogic Bridge documentation ... The docs/userguide.pdf says that
(In service-PAck for Integration-Kit) Import VAJ35weblogic510spX.zip where 'X'
is the service-pack number BUT...I don't find VAJ35weblogic510sp10.zip Anywhere
on BEA site. Please let me know where to get it from if it is ever available.
How do you upgrade the bridge to SP10?
Thanks, Harish.--
Nirav Chanchani
BEA Systems, Inc.

Apache-weblogic bridge problems with virtual hosting

Hi,
Can we set apache-weblogic bridge where apache webserver is configured
as virtual hosts and each Virtual Host is mapped to a WebLogic server
instance running on separate port ?
Please provide doc or reference or your comments if you have solved this
problem
Thanks
Shriniwas

The above comments notwithstanding, there's one outstanding line in your post that seems crucial:
Also if I type on my browser the name of another secondary (or virtual) domain name in the form http://domain2.com it brings the website hosted under the primary domain which is domain1.com. In case I type http://www.domain2.com than it works as it should and displays the website hosted under this domain name.
As far as Apache is concerned, 'domain2.com' and 'www.domain2.com' are completely separate entities. They have to be otherwise if all domain2.com traffic went to this site you'd never be able to host different sites within the domain (e.g. blogs.domain2.com, wiki.domain2.com, blah.domain2.com, etc.). In other words, there is no automatic relationship between a hostname and sub-records of that hostname.
So what's happening here is Apache is seeing a request for 'domain2.com' which does not match any of its configured virtual hosts, so it handles the request via the default site. If you request www.domain2.com this matches your virtual host configuration so Apache serves the alternate content.
The fix is simple - trivial, even - you just need to tell Apache all the hostnames that match this site - in this case if you've configured the site's primary address as 'www.domain2.com' and you want it to also respond to requests to 'domain2.com' then add 'domain2.com' to the list of Server Aliases in Server Admin.

Overrding default Apache-Weblogic bridge error message

We are running WLS 5.1 sp5 with Apache as Webserver. I want to know how
could I display more user freindly error messages when
weblogic is down and apache is working. Currently we get the default
Weblogic-Apache down page.
I tried putting
ErrorPage http://www.mydomain.com/error.html
in apache conf file. Now I do not get the default message, but now it can
not find the above url.
I am able to this url directly in the browser. Please help ?

[email protected] (srinivasa reddy) wrote in
<39bb039e$[email protected]>:
>
hi,
I have a basic question, does webloigc support apache-weblogic
bridge with mod_ssl/any ssl installed on apache.
thanks & regards
[email protected]
There is a mod_wl_ssl.so in sp4 (i believe). mod_ssl extends Apaches API
and thereby breaks the WebLogic bridge. For raven installations you need
mod_wl_ssl_rave.so because raven extends the API even further.
Joel
E-Mail: (jbf (at) proceedo dot com)

Overriding default errror page for Apache-Weblogic bridge

I am writing this problem again as nobody has replied me with any solution.
I want to overide the default error page that comes up when weblogic is down
for some reason in apache-weblogic bridge. I looked into the weblogic
documentation and specfied the following in the apache configuration file
ErrorPage http://www.domain.com/systemerror.html
But for some reason it is not able to locate the domain as well as the page.
But if try the same url in the browser, it works. Please help.
Rakesh

I am using the plugin and using the error page successfully. Apache and WL
use the same document root, and the error page is located within the
document root:
Within the httpd.conf, I have:
<IfModule mod_weblogic.c>
# Config file for WebLogic which defines parameters for plugin
Include conf/weblogic.conf
</IfModule>
Within the weblogic.conf, I have:
ErrorPage unavailable.html
Provide your configuration settings.
"Rakesh Gupta" <[email protected]> wrote in message
news:39d3bcd3$[email protected]..
Well, we are using weblogic supplied apache plugin and I am following all
the steps that is specified in setting up apache plugins in weblogic
documentation. I do not think it is apache network problem as apache is
working fine.
The think the problem is somewhere in the weblogic-apache plugin module.
"Jesse E Tilly" <[email protected]> wrote in message
news:[email protected]..
[email protected] (Rakesh Gupta) wrote in <[email protected]>:
I am writing this problem again as nobody has replied me with any
solution.
I want to overide the default error page that comes up when weblogic is
down for some reason in apache-weblogic bridge. I looked into the
weblogic documentation and specfied the following in the apache
configuration file
ErrorPage http://www.domain.com/systemerror.html
But for some reason it is not able to locate the domain as well as the
page. But if try the same url in the browser, it works. Please help.
Rakesh
as worded, that sounds like a network setup problem on the apache server
side.
however, it seems that you want a server-connection-timeout error toreturn
an specific error page via the plug-in. This all depends on how theplugin
works. If it is how I think it works, there needs to be an INI
setting
to sepcify a page. Why? Because the plug-in effectively has control of
the response. It either needs a way hand control back to the web serveror
needs to have error pages of its own.
BEA?
Jesse

WEBLOGIC JMS performance ?

hi ,
          I tried to find some numbers , about how much messages (in the diffrerent
          categories - ptp or pub/sub) can a weblogic server
          can stand - messages / sec , or some benchmarks,
          but i didn't find any on the web ..
          any one has any idea / link for such benchmark / performance test ???
          I would appriciate any answer !
          thanks!


1) Start with the "WebLogic JMS Performance Guide" white-paper
          on dev2dev.bea.com. This guide has no numbers, but
          it has general guidelines. Performance is not just
          numbers...
          2) General note:
          In 7.0+ JMS file stores, BEA provides 3 disk write policies.
          The default is "cache-flush", as that is transactionally safe
          in all cases. The policy may have a dramatic effect
          on performance. (WL JMS defaults to "cache-flush", IBM MQ
          always uses "direct" - as far as I know, and many smaller
          JMS players default to "disabled".)
          3) The book
          "J2EE Performance Testing with BEA WebLogic Server"
          by Peter Zadrozny, Philip Aston, Ted Osborne
          contains JMS numbers, and a benchmark tool. Keep
          in mind that the numbers here are all "producer
          limited". All the benchmarks are run with
          a single producer, so the numbers often show little
          scaling. (Most apps increase producer counts
          as they scale.)
          Tom, BEA
          Amir Ben-Amots wrote:
          > hi ,
          >
          > I tried to find some numbers , about how much messages (in the diffrerent
          > categories - ptp or pub/sub) can a weblogic server
          > can stand - messages / sec , or some benchmarks,
          >
          > but i didn't find any on the web ..
          >
          > any one has any idea / link for such benchmark / performance test ???
          >
          > I would appriciate any answer !
          >
          > thanks!
          >
          >

RMI Through Apache-Weblogic Bridge

Hi,
How to use RMI in apache-weblogic bridge
Thanks,
Siva E.

[email protected] (srinivasa reddy) wrote in
<39bb039e$[email protected]>:
>
hi,
I have a basic question, does webloigc support apache-weblogic
bridge with mod_ssl/any ssl installed on apache.
thanks & regards
[email protected]
There is a mod_wl_ssl.so in sp4 (i believe). mod_ssl extends Apaches API
and thereby breaks the WebLogic bridge. For raven installations you need
mod_wl_ssl_rave.so because raven extends the API even further.
Joel
E-Mail: (jbf (at) proceedo dot com)

Cache doesn't work with Apace-WebLogic bridge

Hi.We noticed a problem with the download performance of our Java applet with Netscape. It's a tiny applet of 300K that takes only 2 seconds to load using IE, because it is already cached, and with Netscape it is always reloaded and the cache doesn't work.The applet is located on a WebLogic server, and we access it through an Apache server that redirects all requests to the WebLogic server. We noticed that when we access the WebLogic server directly, Netscape cache works fine. However when using the bridge, Netscape ignores the cache, while in IE the cache still works.The Apache server is installed on RedHat Linux 7.1. The bridge is a WebLogic module developed by BEA (the makers of WebLogic) that extends the Apache functionality and enables it to act as a proxy and leave the client (browser) totally unaware of the redirection. The requests are sent to the Apache server and the responses appear to be coming back from it as well.We need to understand why the cache doesn't work when using the bridge. Can you please tell us if you know of this problem or not, and if not -- point us in some direction to check for the problem.Thanks.

You have to modify the password manually in OEM's target.xml.

Weblogic Bridge Batch Size setting ?

What is the per-requisite to set JMS bridge batch size in weblogic 10.3.3
Customer setting in production:
QoS --> Exactly once
Asynchronous Mode Enabled --> True
Batch Size --> 1
From weblogic console
"A messaging bridge instance provides transaction semantics when the QOS is Exactly-once. This envelops a received message and sends it within a user transaction (XA/JTA)."
From documentation
Changing the Batch Size
When the Asynchronous Mode Enabled attribute is set to false and the quality of service is Exactly-once, the Batch Size attribute can be used to reduce the number of transaction commits by increasing the number of messages per transaction (batch). The best batch size for a bridge instance depends on the combination of JMS providers used, the hardware, operating system, and other factors in the application environment. See “Configure transaction properties” in Administration Console Online Help.
Can batch size be set when Asynchronous Mode Enabled --> True ? If yes, what can be the optimal batch size in a production environment ?
Right now application works fines for non-batch orders from CRM. When customers submit batch of 500 orders during specific hours of the day, lot of messages get queued up affecting the order completion rate.
Thanks,
Murali
Edited by: murali_ora123 on May 22, 2013 6:28 AM
Edited by: murali_ora123 on May 22, 2013 6:34 AM

BatchSize does not take effect in async mode.
What is the provider of your source destination? If it is not WLS JMS destination, your bridge may actually work in sync mode although you have configured it to work in async mode. In order to work in async mode for exactly-once QOS, a WLS proprietary feature is needed. We should see a log message in your server log file when the switch to the sync mode happens.
If your bridge is indeed work in sync mode, you need to tune your batch size and batch interval to find the best performance for your application load condition. BatchInterval helps you send a batch before the batch is filled with the number of messages defined by batch size.
Hope this helps.

IPlanet - WebLogic bridge, performance

Similar Messages

Maybe you are looking for