CSM : Server overload and sorry-server

Hi,
I have a portal of eight real servers and one sorry server, which should redirect new user to another portal, in case of an overload condition of all eight real servers. Server load is measured on each real server using a custom developed agent, which basically measures the real CPU load. If a real server experiences an overload, the local agent uses the CSM XML interface to set the maxcons value in the CSM to stop accepting new connections. However, I want to continue accepting sticky connections (request with a valid cookie). The experience shows that the CSM does accept to create new connections to real server reaching maxcons, even if a cookie exist.
This causes a problem if we want to redirect NEW users to another portal in case of overload, but to keep EXISTING users in the server farm, even if the number of connections could increase slightly above maxconns...
How can I solve the problem ?
Thank you
Yves Haemmerli

Try this doc:
http://www.cisco.com/en/US/products/hw/modules/ps2706/products_configuration_example09186a00801a51ba.shtml

Similar Messages

Server Overloaded and Warning Status

I have a WebLogic cluster, includes 3 Managed Server running on 2 separate machines (Each has 16Gb Ram and 8 CPU - Sun Sparc). Each start with parameters
Starting WLS with line:
/usr/jdk/instances/jdk1.6.0/bin/java -server -Xms1024m -Xmx2048m -XX:MaxPermSize=128m
Some time ManagedServers have status "OverLoaded" and "Warning". When that happened , I stop and start server and server status return to OK.
Checking the status with "prstat" command, I saw that java process used just from 1Gb Ram to 2Gb Ram and CPU is under 3%, very little CPU used
What I have to do to make server return to OK status without restart?
And how to make server running with OK status everytime?
And How many Managed Server can I create on each machine with that hardware?
Thank you!

What I have to do to make server return to OK status without restart?Maybe stop some or all applications can be usefull.
And how to make server running with OK status everytime?You must tuning the enviroment, maybe should adjust JVM parameters passed to weblogic server, check thread pools, etc. For Sun HotSpot JVM, you can use -XX:+AggressiveHeap option for throughput applications, give it a try.
See [WebLogic Server Performance And Tuning|http://download.oracle.com/docs/cd/E13222_01/wls/docs92/perform/index.html]
And How many Managed Server can I create on each machine with that hardware?This decision depends on applications you want run, weblogic resources required, how many clients will be attended, and so on. If your weblogic server using 8GB achieves the goal, then you can setup another instance using the remaining available, and so on.
I hope this helps.

Does CSM 4.4 and above support server certificate ?

Dear All,
We have Cisco CSM 4.4. I want to know instead of self-signed certificate, can we import certificate from CA or internal certificate server ?
Please let me know if higher version of CSM support this feature..
Thanks & Regards
Ahmed...

Are you asking about the certificate for the CSM server itself? For that, CSM only supports the self-signed certificate generated during installation. Reference.
The same limitation applies even on the current CSM 4.7 release. I doubt it will be changed since this product will probably be end-of-sales in the next 12-18 months (in favor of the mash up of PRSM and the Defense Center product obtained via the SourceFire acquisition).

Upgrading CSM server from 3.1.1 SP 3 to CSM 3.2 SP1

I am getting an error when attempting to Upgrade my CSM server 3.1.1 SP1 to 3.2 SP1... (See attachement) (Incorrect CSM version, 3.1.1 is not compatible with this version service pack)I need to bring my CSM server to 3.2 to be able to support me IDSM-2 sensors at version 6.1
Thanks,

I am trying to Upgrade from 3.1.1 to 3.2 as recommedned. But I am getting fatal error during installation: Discard or deploy changes to DB on CSM server before upgrade....I logged into the CSM server and Submited and also Discarded all the changes
(There was no changes to be deployed)...Still getting same errors. Any suggestion out there ??
Thanks,

DHCP from CSM Server VLAN

Is there a way to add a helper-address (or something similiar) to the CSM server VLAN? My unix team is planning on setting up a PIXE server boot server and a few of their servers that would need the ability to boot from it are in the server VLAN that my CSM hosts, I'm not sure how a DHCP request would ever leave that VLAN is I don't have some way of adding a helper address to it.
Thanks for any help....Jeff

Hello-
The CSM does not have any DHCP Helper address equivelants. However, you can configure the server VLAN in question in a bridged mode with a vlan on the MSFC that does do DHCP and the CSM will bridge the BOOTP requests to it.
Regards,
Chris Higgins

CSM server originated connections

Hi,
I have a CSM running in router mode with one client vlan, one server vlan and two server vlans for SCAs. Load Balancing is ok, connections directly to the servers and the SCAs through the CSM are ok.
Connections from the server vlan directed to the outside via next hop into the client vlan
(the opposite direction of the usual access) don't work.
Scenario: Server - CSM - MSFC
The client VLAN has the MSFC vlan interface configured as its gateway, the cisco docu says,
that server originated connections should work by default without any NAT, but they don't here.
I also tried "static nat <ip>" and "static nat virtual", no success. The outside vlan interface of the CSM is the last point I can ping, the MSFC doesn't respond, though it is the gateway
address.
Funny: I can ping the SCAs in their vlans and the SCAs can ping the gateway on the MSFC,
it's only from the servers into the client vlan that has trouble...
Does anyone have an idea what's wrong?
Thanks for every hint
Uli

It works for me.
Do a 'sho mod csm X conn' and look for the entry created for the ping from the server to the msfc.
Make sure the response from the msfc matches the entry - same vlan !!!
Check the route from msfc to server and make sure it goes to the same vlan as traffic from server to msfc.
Also, if you did some modification in the vlans or static routes, clear the csm connections since icmp creates long lived entry.
Regards,
Gilles.

CSM Server LB across L3 Devices

I have a pair of 6513's with a CSM in each. These are port-channelled together to provide FT.
I have a number of server vlan's defined on the 6500 and 2 of these need to be LB through the CSM.
Do I define these servers as CSM server vlans (which I have now done) or define the servers as 6500 server vlans with L3 interfaces defined on the 6500's and add route commands to the CSM server VLAN's to route back to these servers via the 6500 L3. I am concerned on this last point about Nat requirements for client connections and how to guage the number of nat pool addresses I need.
My reason for this question is that I am having trouble getting ping access to these LB servers through the CSM even though I have defined server farms with a predictor of forward. Cisco give information on forwarding using TCP / UDP but give no indication of whether the CSM predictor forward supports ICMP.

I am having trouble pinging from an interface on the sup720 / switch fabric on a different subnet to a server on a vlan defined in the CSM as a server vlan.
Not seeing any hits on the csm for this serverfarm.
You said the servers can be on either the CSM or MSFC, but if put the servers to be load balanced on the MSFC, I will need to put in client Nat on the serverfarm to ensure the real server does not attempt to reply directly back to the requestee. If this is the case, does the CSS use a 1 to 1 Nat relationship or does it essentially do a many to 1 address (PAT)

ACE - sticky serverfarm and sorry servers

Primary serverfarm with stickiness (cookie insert) goes down. Backup serverfarm kicks in with sorry servers. Primary serverfarm comes back up and returning connections still get serviced by the backup farm. The reason for this is explained in the load balancing guide.
[quote]
If you want to configure a sorry server farm and you want existing connections to revert to the primary server farm after it comes back up, do not use stickiness.
[/quote]
Source: http://www.cisco.com/en/US/products/hw/modules/ps2706/products_configuration_guide_chapter09186a0080686ebf.html#wp1060156
The big questions is. How can i offer a serverfarm with stickiness and a sorry serverfarm without stickiness?
Roble

Hi Syed,
unfortunately that's the theory. If you have cookie inserts the clients are still stuck to the sorry servers once they have hit the sorry servers.
And my sticky group looks exactly like your first example. The documentation is kind off confusing. My interpretation of the quoted text was if your primary serverfarm is sticky your backup server farm will also be sticky no matter what you configure. That is actually why i asked.
If i use dynamic cookies from the application the clients hop from one rserver to another every 2nd or 3rd connection. The behavior of the stickiness and sorry servers then works like i want it but the stickiness itself is not consistent.
1.5a still had this behavior and i think this might be a bug.
Roble

CSM: Adding 7609 and ASA

Hi,
we have a CSM server running 3.3.1 version.
We are trying to add the 2 following chassis:
7609S running 12.2(33)SRB5A
7609 running 12.2(33)SXH6
Both devices are accepting SSH and telnet connections from the CSM server. Credentials are OK.
From what I see here, those releases are fully supported by CSM 3.3.1. But, when I am trying to import any of these, I have an unknown error showing up. I attached an image showing the error message. I also included a text file, which is an extract of vmssharedsvcs.log but I don't understand a word out of it
I don't know if you guys have any idea why this is happening?
Different issue: we have a Cisco ASA running 8.2(1). This time, the import is fine, but we have a number of warnings "Policy discovery does not support the following CLI in your configuration". Why is this happening while the ASA version is supposed to be fully supported by CSM?
Many thanks,
Thomas

For the AS, CSM will have some minor commands that it will still not support, so it is normal to have these warning there even thoug that ASA version is fully supported.
For CSM I would suggest opening a case with TAC to have it looked at.
I hope it helps.
PK

OO:overloading and overriding in tandem..Interesting one

Assume Dog IS-A Animal
class test {
public static void main(String[] args){
Animal a = new Dog();
foo(a);
void foo(Dog d){
d.printme();
void foo(Animal a){
a.printme();
}//end of class test
class Animal{
void printme(){
System.out.println("i am in animal");
}//end of class animal
class Dog{
void printme(){
System.out.println("i am in Dog");
}//end of class dog
Question:what will it print and explain how briefly u arrived at that result? (I assume u fix some obvious compile time errors(missing ; etc) as this is just a rough snippet)
NOTES-----------
1.Overloading has to do with reference type
2.Overriding has to do with the object to which it actually points

ok mlk..really sorry for being so naive.I sometimes don't clearly express my doubt. I jus got a doubt in my mind regarding overriding + overloading and i constructed a sample example as a concept(not actual code).
Here's the full fledged program..it compiles fine.And IS-A is perfectly satisfied in this case.
class Animal{
void printme(){
System.out.println("In Animal");
class Dog extends Animal{
void printme(){//overriding the Animal printme() method
System.out.println("In Dog");
public class TestDog {
void foo(Animal a){
a.printme();
void foo(Dog d){
d.printme();
public static void main(String[] args) {
Animal a = new Dog();
TestDog t = new TestDog();
t.foo(a); //method overloading
And it prints In Dog..as expected.Got it.Thanks.

Overloading and bla extends bla

if class A extends class B and they have both got the same name on a method but different arguments does his make it overloading still or is it overriding?thanks in advance.

is it only overloading because they extend each other
or would it be anyway?It's just overloading because you have a class that has two methods with the same name.
and if the arguments where the
same would it still be overloading?No, that would be overriding, and only if that method is inherited.
sorry im asking so
much but i'm having trouble understanding the whole
overloading and overriding situations when its
extending a class is involved,does it make a
differance?thanksYes, it does, to overriding. X defines an implementation that you want to change. If Y doesn't inherit it, there is nochting to change. You'd just be implementing yet another method.

What's the difference between "overloading" and "overriding" in Java

What's the difference between "overloading" and "overriding" in Java

hashdata wrote:
What is the real-time usage of these concepts...?Overriding is used when two classes react differently to the same method call. A good example is toString(). For Object it just returns the class name and the identityHashCode, for String it returns the String itself and for a List it (usually) returns a String representation of the content of the list.
Overloading is used when similar functionality is provided for different arguments. A good example is [Arrays.sort()|http://java.sun.com/javase/6/docs/api/java/util/Arrays.html#sort(byte%5B%5D)]: all the sort() methods do the same thing (they sort arrays), but one sorts byte-arrays, another one sorts int-arrays, yet another one sorts Object-arrays.
By the way, you almost certainly mean "real-world" usage. "real-time" (and thus "real-time usage) means something entirely unrelated to your question.

Can we use overload and overwrite concept in OO-abap

hi
can we use overload and overwrite concept in OO-abap

Hi
CLASS zl_lcl_vehicle DEFINITION.
PUBLIC SECTION.
Signature of method
METHODS: set_make
IMPORTING value(im_make) TYPE string " Pass by value
im_model TYPE string," Pass by reference
ENDCLASS. "zl_lcl_vehicle DEFINITION
CLASS zl_lcl_vehicle IMPLEMENTATION.
Implementation of method.
METHOD set_make.
IF im_make IS NOT INITIAL
AND im_model IS NOT INITIAL.
gv_make = im_make.
gv_model = im_model.
ENDIF.
ENDMETHOD. "set_make
ENDCLASS. "zl_lcl_vehicle
Overloading means changing signature as well as implementation of a method.
Overriding is changing only implementation of method with signature unchanged.
From ABAP perspective, only the CONSTRUCTOR method can be overloaded in a subclass i.e both the signature and implementation can be adapted in subclass.
Any other method can't be overloaded. It can only be redefined/overridden i.e implementation changed with signature unchanged.
In ABAP there is something called a redefinition.
When you inherit a class from a super class, you can redifne a method. You cannot chnage the signature( Interface) of the method. It will remain the same as that of the super class.You must redefine a method in the same visibility section in which it appears in the superclass.
Eg.
CLASS C_SUPER_CLASS DEFINITION .
PUBLIC SECTION.
METHODS: DRIVE ,
STOP.
PROTECTED SECTION.
DATA SPEED TYPE I.
ENDCLASS.
CLASS C_SUPER_CLASS IMPLEMENTATION.
METHOD DRIVE.
SPEED = 0.
WRITE: / 'Bike speed =', SPEED.
ENDMETHOD.
ENDCLASS.
CLASS C_SUB_CLASS DEFINITION INHERITING FROM C_SUPER_CLASS.
PUBLIC SECTION.
METHODS DRIVE REDEFINITION.
ENDCLASS
CLASS C_SUB_CLASS IMPLEMENTATION.
METHOD DRIVE.
SPEED = SPEED + 10.
WRITE: / 'Bicycle speed =', SPEED.
ENDMETHOD.
ENDCLASS.
Regards
Vasu

BAPI Overloading and overriding

Hi Friends,
I just want to know Whether we can use BAPI Overloading and overiding concept is possible in BAPI?
Regards
Dinesh

Hello,
As far as I understand It is not possible but can you please explain little bit more about the requirement so we may think of some alternative.
Enjoy SAP!
Augustin.

Overloading and Overriding

Why do we need Overloading and Overriding?

EJP, I'm just curious to know why you mention "*Overloading gives us the power to call a single method and pass it different parameters and the compiler at runtime decided which suitable method to call*" as a wrong answer.Because it is a wrong answer. The action occurs at compile time, not at runtime. The compiler isn't even present at runtime, by definition, so the statement is not just wrong but a contradiction in terms.
Is there any other reason for having overloading in java?The statement isn't a reason at all. It's just a definition, and an incorrect one, of what method overloading is. A definition is not a reason.
So why you're asking me for another reason when there hasn't been one provided at all in the entire thread, other than by me, is a complete, utter, and baffling mystery.
You seem to have the same problem ram has, of confusing 'what' with 'why', and of confusing definitions with reasons. One is bad enough, please let's not have two of you.

CSM : Server overload and sorry-server

Similar Messages

Maybe you are looking for