Session Stickness problem on weblogic load balancer
Hello All
I having trouble related to session stickiness for weblogic load balancing, am using httpclusterservlet on a proxy server that forward requests to 3 managed server that are members of the same cluster.
I tried to search for a solution to this problem and I found two links that are talking about the same problem but there are still not answered!!.
Here are the links:
https://forums.oracle.com/message/11158966#11158966
https://forums.oracle.com/message/11158964#11158964
Based on oracle link Using Sessions and Session Persistence, the documentation says under the "HTTP Session Properties" section that there is a change in how the sessions are formatted and this might cause a problem
for some load balancers (the documentation did not say if weblogic httpclusterservlet is one of them), I tried to do the suggested workaround and I got "java.lang.IllegalStateException: Response already committed" error same as what happens with the user in link https://forums.oracle.com/message/11158966#11158966.
Does anyone knows how to fix this??, I search for a very long time on this problem and with no luck, please help!
Hi,
Can you please give me some more information about your setup ...
Something like ...
F5/LB --> Apache or OHS --> WebLogic Cluster..
If you are using F5/OHS directly to WebLogic cluster you can try enabling the session persistence and this problem should go away..
If you are using plain Apache or OHS try putting "DynamicServerList" off.
Let me know something more about setup so that i can contribute to help ..thanks.
Similar Messages
-
Session lost problem in Weblogic cluster and Iplanet proxy
We have an environment like this
Iplanet 4.1--> Two Weblogic servers.(WLS 6.1 sp1)
The Weblogic servers are not really clusterd, but we are using the Weblogic Cluster
attribute in the obj.conf to configure the proxy (The Weblogic servers are really
independant servers with the same application deployed).
This is working fine in our development environment. No problem with session and
load balancing. Although this architecture is not documented in Weblogic docs,
I have seen several references to this type of architecture in the web and it
seems to be working.
Now we are into our pre-production environment. The same archictecture exists.
Only the following differences
1. Both the weblogic servers run on multihomed machines. We have bound the Weblogic
servers to specific IP addresses using the Listen address option.
2. The same IP addresses are there in the proxy plug-in conf file.
3. There is a firewall between the Iplanet and Weblogic.
Now if only one weblogic is running, the application works fine. The moment we
turn the other weblogic on, the application starts misbehaving. The session seems
to get lost and proxy forwards requests randomly.
What could be the reason?
Regards
Anup
Hi
The problem got solved. There was an older version of libproxy.so in the Iplanet
proxy
Regards
Anup
Yeshwant Kamat <[email protected]> wrote:
>Anup,
>
>Is there a reason you are not clustering the WLS instances? Remove the
>firewall in your
>prod environment and see if that makes a difference.
>
>Anup wrote:
>
>> Hi Mike
>> 1. As per the documentation WebLogic Server is set up to handle session
>tracking
>> by default. So we are not doing anything special. Since the application
>works
>> fine with one Weblogic and multiple clients connecting to it through
>Iplanet ,
>> I think there is no problem with Session tracking as such.
>>
>> 2.We are using the default cookie name for session (JSESSIONID). So
>we haven't
>> done anything extra in the proxy set up or weblogic.xml
>>
>> 3. I have watched the cookie that comes on the browser
>> PAACk3iviDm4ZuMPIbB9TpTTw9slk40IEC02MKjpu14EZ9ayzqaP!-1196227542!gmbpds054!7015!7016.
>>
>> gmbpds054 is the DNS name of one of the weblogic servers. So that is
>also fine.
>>
>> What else could be the problem.
>> Regards
>> Anup
>>
>> "Mike Reiche" <[email protected]> wrote:
>> >
>> >If you have session tracking turned on in weblogic, creating a session
>> >will write
>> >a cookie back to the browser. iPlanet does sticky load balancing
>based
>> >on the
>> >IP address in this cookie. So -
>> >
>> >1) do you have session tracking turned on?
>> >
>> >2) is the cookie getting written to your browser?
>> >
>> >3) are iPlanet and WebLogic using the same cookie? (same name)
>> >
>> >Mike
>> >
>> >"Anup Maliyackel" <[email protected]> wrote:
>> >>
>> >>We have an environment like this
>> >>
>> >>Iplanet 4.1--> Two Weblogic servers.(WLS 6.1 sp1)
>> >>
>> >>The Weblogic servers are not really clusterd, but we are using the
>Weblogic
>> >>Cluster
>> >>attribute in the obj.conf to configure the proxy (The Weblogic servers
>> >>are really
>> >>independant servers with the same application deployed).
>> >>
>> >>This is working fine in our development environment. No problem with
>> >>session and
>> >>load balancing. Although this architecture is not documented in Weblogic
>> >>docs,
>> >>I have seen several references to this type of architecture in the
>web
>> >>and it
>> >>seems to be working.
>> >>
>> >>Now we are into our pre-production environment. The same archictecture
>> >>exists.
>> >>Only the following differences
>> >>1. Both the weblogic servers run on multihomed machines. We have
>bound
>> >>the Weblogic
>> >>servers to specific IP addresses using the Listen address option.
>> >>
>> >>2. The same IP addresses are there in the proxy plug-in conf file.
>> >>
>> >>3. There is a firewall between the Iplanet and Weblogic.
>> >>
>> >>Now if only one weblogic is running, the application works fine.
>The
>> >>moment we
>> >>turn the other weblogic on, the application starts misbehaving. The
>> >session
>> >>seems
>> >>to get lost and proxy forwards requests randomly.
>> >>
>> >>What could be the reason?
>> >>
>> >>Regards
>> >>Anup
>> >
>
-
Weblogic: load balancing and RAC contention
I cannot locate any useful documentation on this: I am trying to find the optimal configuration to allow an Weblogic Application server to load balance to RAC.
I see high numbers of v$sqlstats.clwait_delta for sessions connected from weblogic servers; typically from 25-95% over the overall statement execution time.
Configuration is a three node RAC 10.2.0.4 on Linux; with and eight node weblogic cluster connected using about 150 sessions (equally divided between the rac nodes), hosting a number of applications. They use "Least Connections" load balancing which means any thread(/connection) could be used next. The weblogic sessions connect to a RAC service that is configured with LOAD_BALANCE=YES.
I suspect that contention is caused by behaviour such as
- insert on one node, then immediately query back on another node
- inserts to indexed tables being distributed across all three nodes.
There is too much going on here, and I want to simplify the configuration to eliminate contention for the same data blocks across nodes - this will likely mean new database services configured with LOAD_BALANCE=NO and some form of application partitioning on the weblogic servers. There are other load balancing schemes available for weblogic; but none seem designed to work well with rac.
Is there any best practice documentation available to configure weblogic on RAC (as opposed to single-instance, which doesnt have any such contention issues).
Any pointers appreciated!
Rgds
Mark Teehan
SingaporeHi,
see these 2 whitepapers:
http://www.oracle.com/technetwork/database/enterprise-edition/twpracwkldmgmt-132994.pdf
http://www.oracle.com/technetwork/database/enterprise-edition/bestpracticesforxaandrac-128676.pdf
they should provide you with enough information on what to do.
Regards
Sebastian -
OAM Webgate Ip validation problem caused by load balancer...
Hi all,
In my topology, i have 5 webgates on 5 OHS web servers running in reverse proxy mode . Those web servers are behind load balancer. Since load balancer is working in proxy mode, all requests seems to be coming from load balancer vip and this prevents ip validation at webgate side . Does anybody think that it is possible to solve this issue without changing load balancer configuration..
Regards,Hi,
Randat, how can i reconfigure ip validation against x-forwarded-for? A custom authz plugin, or only a configuration change ? I'll keep on searching on this solution, but if you can share your solution , it'll be appriciated..
Ambarishmitra, i want to use ip validation but since all requests are coming from single ip i can't distinguish client ip's, that's my problem..
Thank you both,
Regards.. -
Oracle Weblogic Load Balancing/Clustering
Can anyone tell me what the recommendation from Oracle is on how to best set up load balancing?
We currently use the configuration.properties file to identify the 2 servers we load balance. We will be implementing additional servers in the near future and we were wondering if this is the best way to load balance 4 or more weblogic servers or if there is some other way.
We set up one cluster address to utilize these servers and use a separate hardware load balancer device.
HRDEV\webserv\hrdev\applications\peoplesoft\PORTAL.war\WEB-INF\psftdocs\hrdev\configuration.properties
# To enable jolt failover and load balancing, provide a list of application server
# domains in the format of; psserver=AppSrvr:JSLport,...
# For example: psserver=SERVER1:9000,SERVER2:9010,SERVER3:9020As this is peoplesoft specific configuration
Please try positing here:
https://community.oracle.com/community/oracle-applications/peoplesoft_enterprise/peoplesoft_general_discussion?customTheme=otn
Best Regards
Luz -
CIAC-CPSC Portal Session cookie problem on weblogic
Hi all,
I´m currently working on a CIAC project with some development made by a partner, the thing is they are using a cookie generated by the application server called “Cognoscookie”, with that they use the session ID to run some queries from a portlet.
The problem is in the customer weblogic, this cookie is not being generated, I tried on a demo environment over JBOSS and it is being generated.
Anyone with some weblogic experience have an idea on what is happening?
I know it is pretty specific question, but you never know if you don’´t ask
Thanks in advance,
Alex.Ok, got it, seems that the Cognos cookie is generated on those installs where the reporting components are installed together with the Portal, so in this case those components where missing, and the cookies too.
Thanks a lot for the help. -
Problem with HTTP load balancing
Hello Experts
I have a problem when i do loadbalancing for links like http://1.1.1.1/site/home where 1.1.1.1 is the VIP address (i got http not found), while it is working fine when the link is http://1.1.1.1,
the link is working fine on the real servers for example when i try http://2.2.2.2/site/home it works
by the way, im not doing URL loadbalancing,
any ideas
Thank you in advanceIt is generally good idea for this type of cases to get a sniffer trace (in ACE module span 10G backplane interface from supervisor or if ACE appliance take parallel span session of client and server vlan).
This case was investigated in TAC SR and this is a small summary of the traces that may help other users hitting this issue (usually it is good idea to filter by http and client IP) :
This is what we have seen for the non-working scenario.
Packet 1: Client sends HTTP GET to ACE VIP
Packet 2: ACE forwards HTTP GET to RSERVER
Packet 3: RSERVER answers ACE with HTTP 404
Packet 4: ACE forwards the real server response (HTTP 404) to the client
ACE was not changing anything in the packets that were being loadbalanced. And the HTTP 404 error sent from the server that ACE was forwarding indicates that the Web server thinks that the HTTP data stream sent by the client was correct, but simply can not provide the access to the resource specifief by URL.
Bottom line it was found that in this case the server behaves in a different way based on the hostname used to connect to the application, and this should be addressed on the application/server side. An easy way to check this is by using the server name pointing to the vip in local client hostfile. -
Weblogic cluster software load balancer
Hi,
We are currently using Weblogic domain as a Proxy Plug-In for High Availability test as it’s explained in this blog http://andrejusb.blogspot.com/2009/04/weblogic-load-balancing-for-oracle-adf.html.
Its working fine for POC project but what software load balancer would you recommended for production environment on Linux? (Assume that we don’t have a hardware load balancer).
- Oracle active-passive OHS web-tier clustering.
-Using Linux open source Linux Software(e.g. HAProxy and KeepAlived as explained here http://biemond.blogspot.com/2010/04/high-availability-load-balancer- for.html
-Using any other software load Balancer
I would appreciate if anybody can provide some recommendations/links etc.
Thanks
AlexHi Alex,
Yes you should never use HttpClusterServlet, not even for fun ;-)
We use mod_wl (Web Server Plug-In) for Apache for several customers and that works fine.
Check this:
http://docs.oracle.com/cd/E23943_01/web.1111/e14395/toc.htm
Regards Peter -
Printer Pool load balancing problem.
We seem to be having a problem with printer load balancing working correctly. I have several computer labs. Each lab has at least 2 printers of the same type. As a result I pool the printers for each lab. We are still running ZCM10.3.4. Most of the clients are running iprint 5.82. With Zenworks 7 and 6.5 load balancing on these printers worked very well and page counts stayed very close on each printer. Now that we are at ZCM10 and have been for about 3 years now we continually see that one printer gets the majority of jobs. Most of the computers are imaged but printers are installed after imaging. Even if we alternate the default printer on the computer print jobs still mostly come out the same printer. All printers except 3 are HP and in our case they are all new HP M602's with a couple older HP4015DN's thrown in when the new HP M602's would print garbage across the wan and the 4015's would print correctly but that is another topic.
Am I misunderstanding how load balancing is supposed to work? Isn't load balancing supposed to be one of the features of printer pooling in addition to sending jobs to the printer that is not busy at the time the job is sent? How is load balancing set up?
Thank you.are you using any front end web server or hardware load balancer ? verify your load balancing if it's happening or not properly between your clustered managed servers.
Regards
Mukesh Negi
http://weblogicserveradministration.blogspot.com/ -
Load Balancing Forms Services with an effective healthcheck
I am in the process of configuring two forms 11.1.2 servers running with weblogic 10.3.5 with multiple forms applications clustered across both physical servers. We are looking to load balance to the various forms applications using a hardware load balancer. Can anyone comment on their experience with setting up effective application healthchecks using either hardware or software load balancers?
For example say that we have 3 applications clustered across 2 servers with the following
URLs:
http://server01:7777/forms/frmservlet?config=myapp1
http://server01:7777/forms/frmservlet?config=myapp2
http://server01:7777/forms/frmservlet?config=myapp3
http://server02:7777/forms/frmservlet?config=myapp1
http://server02:7777/forms/frmservlet?config=myapp2
http://server02:7777/forms/frmservlet?config=myapp3
We would need a checking mechanism on the load balancer that could tell if myapp2 was down on say server01 and therefore block traffic to that application yet keep traffic open for the other 2 apps on the same server.
A specific difficulty with forms is that when the database behind the application is down forms services will return an error message within a displayed error form. From the load balancer's point of view the forms services are "up". We need to find a way of detecting that the application is actually available and not just that the forms services themselves are available. To detect that the forms services are available we might normally use the status check:
http://server01:7777/forms/frmservlet?ifcmd=status
however this will only tell use the availability of the forms services on a physical server and not whether any actual applications are available.
I am aware that f5 do a BIGIP offering that includes some Oracle Forms specific components. Can anyone comment on how they have setup Oracle Forms healthchecks using various load balancing methods. In particular if a load balancer is limited to using WGET commands to check HTTP returned content is there a way of checking a forms application's availability and if not how have other people achieved an effective healthcheck?
Many thanks,
PhilippeDid you ever Get this to work?
I am having some problems trying to load balance with Oracle Forms, Discover and reports Oracle Application Server Release 10g (9.0.4.0.2) and I was wondering if you could help. Has any one ever got this to work consistently? We are an ERP product written mostly in forms (904) and are trying to implement are largest customer there performance issue so we need the load balancing to work. Will also accept other recommendation as cost effective as solutions.
Site 1:
A: SERVER
Host as1.xyzco.local
Version 10.1.2.0.2
Installation Type Identity Management and Metadata Repository
Oracle Home E:\oracle\inf_1012
Farm as1db.xyzco.net
o HTTP_Server
o Internet Directory
o OC4J_SECURITY
o Single Sign-On:orasso
o Management
B SERVER
Host as2. xyzco.local
Version 9.0.4.0.2
Installation Type Business Intelligence and Forms
Oracle Home E:\oracle\mid_904
Farm as2db. xyzco.net
o Discoverer
o Forms
o home
o HTTP_Server
o OC4J_BI_Forms
o Reports Server
o Web Cache
o Management
C SERVER
Host as3. xyzco.local
Version 9.0.4.0.2
Installation Type Business Intelligence and Forms Discoverer and Reports
Oracle Home E:\oracle\mid_904
Farm as2db. xyzco.net
o Discoverer
o Forms
o home
o HTTP_Server
o OC4J_BI_Forms
o Reports Server
o Web Cache
o Management
All servers Are:
Windows 2003 Standard Server with current service packs
Xeon Dual Processor with 4GB ram
Raid 0 drives 2 for OS and 2 for Oracle
Daniel Brody
[email protected] -
FRM-92101: Load Balancing Forms 10g with Webcache
Hi,
I´m having some problems with Oracle Forms and Webcache 10g and I was wondering if you
could help.
Machines: (Pentium IV 2.6 GHz; 2GB RAM; 80GB HD)
A. Red Hat Linux AS 2.1
Oracle AS 10g - Infrastructure Installation
B. Windows 2000 Server - Service Pack 4
Oracle AS 10g - BI and Forms Installation (Using Infrastructure on machine A)
C. Windows 2000 Server - Service Pack 4
Oracle AS 10g - BI and Forms Installation (Using Infrastructure on machine A)
I didn´t apply any Patchsets of Oracle AS 10g. My current version is 9.0.4.0 .
Goal:
Use Oracle Webcache 10g on machine B for Load Balancing an Oracle Forms Application
on machines B and C.
Facts:
The Oracle AS 10g installation on the 3 machines was made without problems.
The Forms Applications are running without problems on both Forms servers in machine B and C.
Symptoms:
I´m getting the following error when I try to access the forms application through Webcache:
FRM-92101: There was a failure in the forms server during startup.
This could happen due to invalid configuration.
Please look in the web-server log file for details.
But, this error happens ONLY when BOTH OC4J_BI_Forms instances (machines B and C) are
running.
When just one of the OC4J_BI_Forms is running everything works fine.
Webcache Configuration:
I´ve followed Metalink note 207668.1: Configuring Web Cache as a Load Balancer for Application Servers
step by step to make the configuration of the following components:
1. Configuration of the origin servers
2. Configuration of Site definitions
3. Configuration of Site to server mapping
After that, I´ve follwed the Metalink note 229900.1: Integrating and using Web Cache with Forms 9i for Load Balancing
step by step to make the configuration of the following components:
4. Session Binding
Then I try to test the Load Balancing and I got the error FRM-92101 when both OC4J_BI_Forms
instance were running.
I´ve searched Metalink and found note 268830.1: Webcache Won't Load Balance 2 Forms Instances.
Following the note instructions I did the following changes:
Added these lines at the end of HTTPD.conf (machine B):
CookieTracking On
CookieName myformscookie1
Added these lines at the end of HTTPD.conf (machine C):
CookieTracking On
CookieName myformscookie2
In Webcache configuration:
I´ve Changed the default session binding from JSESSIONID to "ANY SET COOKIE" for the
site's session binding.
But even after these changes I´m getting FRM-92101 error on the same conditions.
How can I solve this problem ?
Thank you in advance.
Marcus Santos.
([email protected])Did you ever Get this to work?
I am having some problems trying to load balance with Oracle Forms, Discover and reports Oracle Application Server Release 10g (9.0.4.0.2) and I was wondering if you could help. Has any one ever got this to work consistently? We are an ERP product written mostly in forms (904) and are trying to implement are largest customer there performance issue so we need the load balancing to work. Will also accept other recommendation as cost effective as solutions.
Site 1:
A: SERVER
Host as1.xyzco.local
Version 10.1.2.0.2
Installation Type Identity Management and Metadata Repository
Oracle Home E:\oracle\inf_1012
Farm as1db.xyzco.net
o HTTP_Server
o Internet Directory
o OC4J_SECURITY
o Single Sign-On:orasso
o Management
B SERVER
Host as2. xyzco.local
Version 9.0.4.0.2
Installation Type Business Intelligence and Forms
Oracle Home E:\oracle\mid_904
Farm as2db. xyzco.net
o Discoverer
o Forms
o home
o HTTP_Server
o OC4J_BI_Forms
o Reports Server
o Web Cache
o Management
C SERVER
Host as3. xyzco.local
Version 9.0.4.0.2
Installation Type Business Intelligence and Forms Discoverer and Reports
Oracle Home E:\oracle\mid_904
Farm as2db. xyzco.net
o Discoverer
o Forms
o home
o HTTP_Server
o OC4J_BI_Forms
o Reports Server
o Web Cache
o Management
All servers Are:
Windows 2003 Standard Server with current service packs
Xeon Dual Processor with 4GB ram
Raid 0 drives 2 for OS and 2 for Oracle
Daniel Brody
[email protected] -
Certificate based authentication with SSL load balancer
I've been asked to implement certificate-based authentication (CBA)
on a weblogic cluster serving up web services. I've read through
Chapter 10 (security) and understand the "Identity Assertion" concept.
Environment:
Weblogic 8.1 cluster fronted by a load-balancer that handles SSL and
uses sticky-sessions.
Question:
If the load balancer is used to handle SSL, do I still need to turn
on SSL on the weblogic cluster in order to use CBA? Is there another
way to request the client's certificate?
If the above is yes, what is the minnimal level of SSL? Does it have
to be two-way?
If SSL has to be turned on is there any reason to use the load
balancer's SSL? Is there still a performance benefit?I think the simplest and most secure way is to have the servers configured for
2-way ssl, since this would ensure that the certificate they receive and use for
authentication has been validated during the ssl handshake. In this case the load
balancer itself does not need to and cannot do the handshaking, and would need
to pass the entire SSL connection through to the WLS server (ie: act similar to
a router)
Pavel.
"George Coller" <[email protected]> wrote:
>
I've been asked to implement certificate-based authentication (CBA)
on a weblogic cluster serving up web services. I've read through
Chapter 10 (security) and understand the "Identity Assertion" concept.
Environment:
Weblogic 8.1 cluster fronted by a load-balancer that handles SSL and
uses sticky-sessions.
Question:
If the load balancer is used to handle SSL, do I still need to turn
on SSL on the weblogic cluster in order to use CBA? Is there another
way to request the client's certificate?
If the above is yes, what is the minnimal level of SSL? Does it have
to be two-way?
If SSL has to be turned on is there any reason to use the load
balancer's SSL? Is there still a performance benefit? -
X.509 certificate based authentication with load balancer
I've been asked to implement certificate-based authentication (CBA)
on a weblogic cluster serving up web services. I've read through
Chapter 10 (security) and understand the "Identity Assertion" concept.
Environment:
Weblogic 8.1 cluster fronted by a load-balancer that handles SSL and
uses sticky-sessions.
Question:
If the load balancer is used to handle SSL, do I still need to turn
on SSL on the weblogic cluster in order to use CBA? Is there another
way to request the client's certificate?
If the above is yes, what is the minnimal level of SSL? Does it have
to be two-way?
If SSL has to be turned on is there any reason to use the load
balancer's SSL? Is there still a performance benefit?Hi George,
If you want the client's cert, the server has to ask for it and this
implies two-way SSL. Normal one-way SSL the server provides the cert to
the client and the client decides if it wants to continue the handshake.
If the client is OK with the server certs and two-way SSL is configured
on the server, then the server will request the client send it's certs.
If the client certs are OK, then the pipe is established.
Concerning the load balancer I'm assuming it is simply providing a
tunnel, but I don't have the experience to comment and it is something I
would suggest that you that you seek guidance from our outstanding
support team [1] or drop a note in the security newsgroup [2] for the
experts to review.
Regards,
Bruce
[1]
http://support.bea.com
[email protected]
[2]
http://newsgroups.bea.com/cgi-bin/dnewsweb?cmd=xover&group=weblogic.developer.interest.security
George Coller wrote:
>
I've been asked to implement certificate-based authentication (CBA)
on a weblogic cluster serving up web services. I've read through
Chapter 10 (security) and understand the "Identity Assertion" concept.
Environment:
Weblogic 8.1 cluster fronted by a load-balancer that handles SSL and
uses sticky-sessions.
Question:
If the load balancer is used to handle SSL, do I still need to turn
on SSL on the weblogic cluster in order to use CBA? Is there another
way to request the client's certificate?
If the above is yes, what is the minnimal level of SSL? Does it have
to be two-way?
If SSL has to be turned on is there any reason to use the load
balancer's SSL? Is there still a performance benefit? -
I am having some problems trying to load balance with Oracle Forms, Discoverer and reports Oracle Application Server Release 10g (9.0.4.0.2) and I was wondering if you could help. Has any one ever got this to work consistently? We are an ERP product written mostly in forms (904) and are trying to implement are largest customer there performance issue so we need the load balancing to work. Will also accept other recommendation as cost effective as solutions.
Site 1:
A: SERVER
Host as1.xyzco.local
Version 10.1.2.0.2
Installation Type Identity Management and Metadata Repository
Oracle Home E:\oracle\inf_1012
Farm as1db.xyzco.net
o HTTP_Server
o Internet Directory
o OC4J_SECURITY
o Single Sign-On:orasso
o Management
B SERVER
Host as2. xyzco.local
Version 9.0.4.0.2
Installation Type Business Intelligence and Forms
Oracle Home E:\oracle\mid_904
Farm as2db. xyzco.net
o Discoverer
o Forms
o home
o HTTP_Server
o OC4J_BI_Forms
o Reports Server
o Web Cache
o Management
C SERVER
Host as3. xyzco.local
Version 9.0.4.0.2
Installation Type Business Intelligence and Forms Discoverer and Reports
Oracle Home E:\oracle\mid_904
Farm as2db. xyzco.net
o Discoverer
o Forms
o home
o HTTP_Server
o OC4J_BI_Forms
o Reports Server
o Web Cache
o Management
All servers Are:
Windows 2003 Standard Server with current service packs
Xeon Dual Processor with 4GB ram
Raid 0 drives 2 for OS and 2 for Oracle
Daniel Brody
([email protected])I have more then half decent results with webcache as load balancer, but it still is a round robin (as mentioned before). This means that a session will not be transfered from one application server to the other once the session has been created.
So far I did not discover any benefits of a hardware loadbalancer over webcache, because (here it is again:) you are not able to transfer open sessions from one application server to the next. This has to do with the Oracle Forms architecture. In fact it is still client-server, only with the application server functioning as extra server ;) -
CSS 11503 Load Balancing Verification
Alright, so I have toiled long and hard to get this right. I think I have the config down but I am unsure on how to verify how this load balancing is working.
Here is the Content Config that I am speaking of:
content cad-rule
add service wls1-e0
add service wls1-e1
add service wls2-e0
add service wls2-e1
add service wls3-e0
add service wls3-e1
add service wls4-e0
add service wls4-e1
add service wls5-e0
add service wls5-e1
add service wls6-e0
add service wls6-e1
arrowpoint-cookie expiration 00:00:15:00
advanced-balance arrowpoint-cookie
redundant-index 2
vip address 172.30.194.195 range 2
arrowpoint-cookie name TOQ
protocol tcp
port 8001
url "/*"
active
Each service in the rule above is configured as follows:
service wls1-e1
port 8001
protocol tcp
strin ags001-e1
ip address 172.30.193.81
keepalive type http
keepalive uri "/cad/index.html"
redundant-index 12
keepalive frequency 20
keepalive maxfailure 10
keepalive retryperiod 2
active
I am using the advanced arrowpoint cookies because I need some stickiness here. Straight round-robin would not have done what I needed it to do.
Now, when I go to my show summary, this is what I see for this rule:
cad-rule Master wls1-e0 84274
wls1-e1 13144
wls2-e0 96884
wls2-e1 26374
wls3-e0 71145
wls3-e1 16592
wls4-e0 76403
wls4-e1 8657
wls5-e0 118623
wls5-e1 22760
wls6-e0 30836
wls6-e1 20464
The far right column indicates the services hits. I originally had the E1's suspended and activated them later on. So if this was true round robin, all the E0's should have the same number of service hits and all the E1's should have the same number of service hits. But as you can see, the wls5 server is getting hit the most while the wls6 server is sitting there twiddling its thumbs.
Now understanding how the arrowpoint cookies do their load balancing (inserting a cooking into the flow and then timing out after 15 mins as configured above) I would not expect a 1:1 ratio of load balancing between servers. But the distribution above seems rather extreme.
Does anyone have any suggestions on how to both A) verify that this is the right config and B) suggest to my boss that this is working the way it should be working?
Thanks!
JamesHi James,
There are several reasons of the uneven load balancing that you are seeing (based on the show summary). First
of all, the CSS is configured to do stickiness (advance-balance).
With arrowpoint-cookies (for HTTP only) method for stickiness, only the requests coming with the same cookie
are going to get stuck to the same server, since the cookie is
lost when the browser is closed (or based on the expiration), then the stickiness is going to be session
based and if the same client open a new session is going to be load balanced.
Is important to understand that when using stickiness, no real even load balancing is
going to happen since we are sticking new flows to the same server; even when layer 5 stickiness would
permit more even balancing than layer 3 stickiness (source IP based).
Also consider that the "show summary" is a command to see the hits (requests) being balanced to an specific
server, this is a good command to see the load balancing, anyway since the CSS balance
connections (flows), a persistent connection could have a lot of requests, so all those requests are
always going to the same server (incrementing the amount of hits in the counter) while a non-persistent
connection would be just one request (refer to HTTP persistence).
Also keep in mind that if a service is take out for maintenance, or is added to the load balancing later
than another, or if goes down for a period of time, then the CSS will be balancing among the remaining alive
servers. When you add the server again, the another servers are going to have connections
already established, so since the CSS is doing round robin, the server last added will
never have the same amount of connections (nor hits) that the other ones, because while one could
have 55 for example, the new one will have it first connection, and when the first one
gets the 56, the another will get the second, and so on.
Please let me know if this makes any sense.
Diego M
Maybe you are looking for
-
Im trying to get a panel to display with a text area, but it wont work. Here is the code: public ApplicationTest() frame.setContentPane(this); frame.setLocation(FRAME_UPPER_LEFT_X_COORD,FRAME_UPPER_LEFT_Y_COORD); frame.s
-
Images on Firefox come out pixelated
Hi, For some reason, Firefox on my MacBook Air displays pixelated images. When I access the same site on Google Chrome, the images come out with good resolution. I tried looking for the settings to correct this but cannot seem to find it. I do enjoy
-
Help with installing actions PSE9 on Mac
Hi everyone! Thank goodness for support forums! I've been searching all over this afternoon to see if I could just install some simple presets in PSE9--I just got it and installed it on OSX 10.6. I just can't figure out how to install actions. I've t
-
LaCie Big Disk Extreme+ Triple 2TB
I am thinking of getting a LaCie Big Disk Extreme+ Triple 2TB for use with Final Cut Express. This drive is actually 2 1-TB disks. Is there any problem using such a drive in capturing video (or anything else having to do with video editing/production
-
ITunes - when does it all end ?
It is not worth all the trouble. Never ever had problems with iTune pre 5.xxxx Never ever had software issues like I' am reading here with any ones programs. Apple does not respond to my support questions. Will Apple ever let me know when they are ba