Is usage of Proxy nodes expensive ?

Hi,
I would like to know how expensive is the usage of proxy node from performance
point of view.
Thanks
Girish

ABAP Proxy Runtime
http://help.sap.com/saphelp_nw04/helpdata/en/02/265c3cf311070ae10000000a114084/frameset.htm
ABAP CLIENT PROXY
/people/ravikumar.allampallam/blog/2005/03/14/abap-proxies-in-xiclient-proxy
ABAP SERVER PROXY
/people/siva.maranani/blog/2005/04/03/abap-server-proxies
This material will help you a lot.
regards
Nisar

Similar Messages

Detection of proxy node outage

Question regarding detection of proxy node outage in extend client.
My impression is that once connection established, you won't notice that proxy node was gone until next time you try to access cluster (get, put, etc. etc.).
Or there is anything like MemberListener thing that client will be notified when proxynode is gone?
This cause us some issue since we will add a map listener within extend client, but we won't receive any notification if the proxy node is gone.
To address this issue, we are implement a heartbeat approach which client will have a background thread perform periodically cache get to ensure that the connection still valid.
Above approach worked fine for single proxy node approach, since if the heartbeat check failed, we know something wrong.
However, the heartbeat approach create new issue when there are multiple proxy nodes configured in client's remote cache scheme.
This is what we observed.
An extend client configured with a remote cache scheme has proxyA and proxyB as proxy list.
1. Extend client start with connection to proxyA.
2. Extend client subscribe to key ABC for cacheX
3. Kill proxyA process.
4. The background heartbeat check thread perform a heartbeat get to cacheX. Looks like during this cacheX.get() operation, Coherence will establish a new connection to proxyB automatically. Our application doesn't even notice there is a proxy node switch.
Now the issue is if an update to key ABC occurred between step 3 and 4. Our application won't receive any notification at all. And we don't even notice there is a new connection created to a different proxynode at all.
Our business requirement require us not to miss any entry update notification in extend client (unless if all proxy nodes are down) as long as there is at least one proxy node available.
Is there any better approach for this requirement?

J.K.
Turn out that member listener approach still have chance to lost update notification.
We found out that the update occurred between MemberLeft event notification and re-add map listener will be lost.
That window is small, however, our business requirement require that no lost of update event at all as long as the update operation succeeded (the update could be performed by another extend client connect to different proxy node).
Is it even possible to ensure that requirement using extend client?
Regards,
Chen

Dynamic proxy nodes within Webflow???

Hi all,
we are trying to implement a dynamic set of webflows, i.e. we wan't to have reusable
sub webflows being used in several parent webflows. In order to this we need to
have a dynamic component at the end of a sub webflow in order to move back to
the appropriate node in the parent webflow. We do not wan't to explicitly configure
every path to every parent webflow in the sub flow. In the ideal world, the sub
webflow would be a black box that we could include in a parent webflow, however
the world is not ideal and we are therefore investigating our possibilities. Does
anyone have a great solution for this?
What we are thinking of is storing the namespace and the target node of the parent
webflow in session before entering a sub flow, the sub flow would then (when finished)
get the parent namespace and node and proxy back. However proxy nodes are not
dynamic so ???
Any ideas or suggestions are appreciated,
thanks,
Mikkel

As far as i know, you wouldn't even be able to do this by generating the byte-code to override
the methods by hand. I think the verifyer checks for the overriding of final methods - so even if
you generated such a class, it would fail class loading.I can't remember what the licence says, but it may be possible to use a modified version of java.lang.Object in which they're not final provided you only use it internally and don't distribute it.

Pls explain usage of proxy.

1) pls explain usage of proxy and what is the abap proxy.
tell me step by step.

ABAP Proxy Runtime
http://help.sap.com/saphelp_nw04/helpdata/en/02/265c3cf311070ae10000000a114084/frameset.htm
ABAP CLIENT PROXY
/people/ravikumar.allampallam/blog/2005/03/14/abap-proxies-in-xiclient-proxy
ABAP SERVER PROXY
/people/siva.maranani/blog/2005/04/03/abap-server-proxies
This material will help you a lot.
regards
Nisar

Have you had high cc usage for a planned expense that lowered your FICO scores?

In a week or so I am initiating some home improvements and repairs. I know the charges to my cc will be way over the ideal limit until I pay it(them) down, but I now have zero interest on several cards so now is the time to do it. Has someone done this and seen their credit scores sink to the bottom of the barrel? I am thinking I should just not look at scores while I am doing these necessary things and just continue paying down the balances until my credit scores recover. How badly did your credit take a hit if you have done this? Was it worth it to you? (I only just got these scores in the last weeks due to utilization and new accounts.) Oh, and what is considered 100% usage? How close to the limit can I go? Thank you for any responses. In my wallet: Venture 15k; Barclay Ring 12.5k; Discover 10k; CareCredit 7.7k; Amazon Store Card 5k. QS 5k; Band of A Travel Rewards 3.5k; Bank of A Travel Rewards 3.5k; Bank of A Platinum Visa 2k; Barclay Apple Rewards 2k. Total Credit: 66,200. 7/2-FICO: EQ - 788; TU - 777; EX - 690; Utilization: 12%; GOAL: 780 across the board then 800 across the board. Inquiries: EQ: 5 TU: 7; EX 3; Chapter 13 seven years May, 2015. Reset my garden date to 6/27. Gardening until further notice.

Musiclover wrote:
Have you had high cc usage for a planned expense that lowered your FICO scores?I don't recall but I have been up to 76% recently. However, the balance was immediately paid off. I don't recall if reported or not. The impact didn't really matter since it was going to be paid off immediately. Musiclover wrote:
Oh, and what is considered 100% usage?
Utilization is balance/limit. If the balance equals the limit it's 100%. Musiclover wrote:
How close to the limit can I go?Depending on the card you may be able to exceed the limit. However, such cards would require that you immediately pay any balance exceeding the limit. For scoring purposes it is recommended to not exceed 30%. However, short term high utilization generally isn't an issue. It's prolonged high utilization that can get you into trouble. Be aware that there is a double hit from having high utilization on a card and having itmaxed out (generally >90%). Musiclover wrote:
I know the charges to my cc will be way over the ideal limit until I pay it(them) down,How long will that take? What utilization would you be at initially? How long would it take to get it down to 50%? Under 30%?

External component usage with interface node

Hi everyone,
I have component A , which has an interface node with Input element checked:
Nodes
Node Name     IF_HROBJID
Interface Node     1
Input Element (Ext.)     1
Dictionary structure     ZHR_HROBJID_LIST
Cardinality     0..n
I have another component: Component B Where I want to bind a table to this interface node.
In component B :
1. I definied the Componenet Usage of component B in the componenet controller.
2. I drag & drop the IF_HROBJID node to the component controller's context (the node is mapped)
But when I try to bind a table, I always get the " The Mapping to Node COMPONENTCONTROLLER.1.IF_HROBJID Has Not Been Completed. " error.
nd_if_hrobjid = wd_context->path_get_node( path = 'IF_HROBJID' ).
nd_if_hrobjid->bind_table( lt_if_hrobjid ).
Can someone tell me what is the problem, and how can I solve it?
Thanks
N.

Hi,
Here you are trying to use component A in component B right?
Then what you need to do is:
1. Declare the component usage of A say A_USAGE in properties tab of component B.
2. Create a node say 'DATA' in component controller of component B. The node 'DATA' would have the same structure as the node in your component A.
3. Fill the node 'DATA' in component controller's method.
4. Under your webdynpro component B, you will see Component Usages -> A_USAGE. open interface controller inside that.
    Goto context tab and drag the 'DATA' node from component controller of B on to the node in interface controller.
Hope this is clear!
Best Regards,
Srilatha

How to add proxy node in message monitoring RWB

Hi all,
in RWB message monitoring , i can see two options in drop down list those are integration engine and adapter engine. i want to add one node for proxy run time also.
can anybody let me know how to add a node for proxy runtime?
any document or help will be appreciated
Thanks
sandeep sharma

Hi,
connect your SAP application system (like ERP, etc.)
to SLD - the same one as PI uses (you need to have PIRWBUSER on ERP too)
then start transaction SPROXY and build and interface on it
start it and check if you see the new node in message monitoring
Regards,
Michal Krawczyk

Apache proxy, Node.js and Default Websites

Hello everyone !
I'm currently using a mac mini server under OS X Mavericks (10.9.1) and Server 3.0.2
I've been very happy with it, and found it was quite nice to use... until I decided to host a Ghost blog (https://ghost.org) on it. Ghost runs on Node.js (http://nodejs.org).
I have a domain (domain.com), and two subdomains (www.domain.com and sub.domain.com).
The server and it's default websites use sub.domain.com, with http://sub.domain.com redirecting to https://sub.domain.com
I want to use both domain.com and www.domain.com for the blog.
Using cloudflare for the DNS, I created a CNAME for domain.com, pointing to www.domain.com (domain.com CNAME www.domain.com).
Node.js and Ghost install both went well, with Node.js listening on port 2368.
I then started configuring a virtual host (www.domain.com) in the Server.app for the Ghost blog , along with an Apache Reverse Proxy, by editing www.domain.com's .conf and replacing everything in it by
<VirtualHost 10.0.1.2:80>
     ServerName www.domain.com
     ProxyPreserveHost on
     ProxyPass / http://localhost:2368/
</VirtualHost>
Going to www.domain.com, I correctly end up on the Ghost blog.
But, amusingly and annoyingly, when trying to access the default website through sub.domain.com... it's also the blog that shows up !
After setting up another website (sub2.domain.com) in the Server.app and configuring the DNS appropriately at Cloudflare, I could notice that I end up on the default page for "empty" websites (the "welcome to Server... blabla" page).
So the server's default website is the only one getting this issue...
I've been spending two days on it, trying nearly every other Apache Reverse Proxy configuration I could find (yes, even the weird one where you go through DNS settings that you can find somewhere in these forums).
Anyway, some help would be greatly appreciated

Ok... So, while I did not find a solution, I did find a workaround.
Get ready people...
First, create www.yourdomain.com on port 8080 in the Server App.
Then, create www.yourdomain.com on port 80 in the Server App, and choose to make a redirection from this site to www.yourdomain.com:8080.
After, go to the 0000_any_8080_www.yourdomain.com.conf file, and open it with textedit.
Replace everything in it with
<VirtualHost *:8080>
     ServerName www.domain.com
     ProxyPreserveHost on
     ProxyPass / http://localhost:2368/
</VirtualHost>
Save it (don't forget replace the original 0000_any_8080_www.yourdomain.com.conf with it if needed and to keep the.conf extension)
And there you go !
For SSL, do the same thing, except you need to replace the 80 port by the 443 port, and the 8080 port by the 8443. And don't forget the certificates, for both of them.

Usage of command node in smartforms

Hi
my doubt is can I branch to multiple pages using command node in smart forms.if so how to do it.

Hi,
Please check with the following threads
[Command node|Re: Smartforms: Option of Printing Specific Pages Only;
[Document on command node|http://www.sdn.sap.com/irj/scn/index;jsessionid=(J2EE3417200)ID0250336250DB00400831958585035375End?rid=/library/uuid/a01b2c4d-711f-2b10-3eb2-abe3464bb370&overridelayout=true]
Hope this helps you.
Regards,
Rajani

Newbie usage of Proxy Server 3.6 SP5 : files not cached at all

Hello,
I'm trying to test the proxy server to see if it is caching files coming from our portal. It seems that nothing is cached but at the same time the browser is able to cache files.
The urls to cache have mostly the following form:
http://myportal/portal?cmd=image&object=toto.gif
In the access log, the url seems to return a 304 http code and is not cached.
So my question is : is the proxy server supposed to cache files that can be cached by the browser or is his behaviour different from the cache management of a browser.
Thanks for your help

Hi B Noir,
I recently had to do a System Restore on my Windows 7 64 bit machine and then I got the dreaded "Safari has stopped working" windows whenever I try to launch it. To be more specific, Safari, doesn't even launch. I get that message. Then when I click "Close the program", I get the "Safari has stopped working" message again. After I close that, then it stops.
I've tried reinstalling the latest version of Safari twice now, and it still isn't working.
I also tried creating another Windows account, and Safari does not work on the new account either, unfortunately.
Finally, I also tried your "So-much-for-subtlety" instructions and that unfortunately did not work.
Any ideas? If it helps, here's the problem signature from when I get the "Safari has stopped working window":
Problem signature:
Problem Event Name:          APPCRASH
Application Name:          Safari.exe
Application Version:          5.34.52.7
Application Timestamp:          4eb16d27
Fault Module Name:          KERNELBASE.dll
Fault Module Version:          6.1.7601.17651
Fault Module Timestamp:          4e211319
Exception Code:          80000003
Exception Offset:          0001280c
OS Version:          6.1.7601.2.1.0.768.3
Locale ID:          1033
Additional Information 1:          00b8
Additional Information 2:          00b85080f77ffe6e49f095c1fe3be289
Additional Information 3:          b2f2
Additional Information 4:          b2f2b584c57e4d0d4f1a9442f91680f3
Read our privacy statement online:
http://go.microsoft.com/fwlink/?linkid=104288&clcid=0x0409
If the online privacy statement is not available, please read our privacy statement offline:
C:\Windows\system32\en-US\erofflps.txt
Thank you.

How Useful, Effective and Stable is the Node Function

Please note my current MacPro below.
I'm seriously thinking about purchasing a used ("Refurbished") MacPro similar to the one below to help share the sequencing workload. Of course, there are a number of ways to setup the the second MacPro computer to share the workload. One of them is to use the Node Function.
The manual states that utilizing the Node function will help with the CPU workload of software synths. If I understand the manual correctly, all that needs to be done is to initialize the node function after connected the second Apple computer to the first via the Ethernet cable. I have a number of instrumental libraries that uses Native-Instrument's Kontakt 2 sampler software (GPO, JABB, Kirk Hunter "Diamond" orchestral libraries, Chris Hein Guitars, etc., etc.) I like ALL of these libraries but they do definitely put a drain on the computer's CPU resources.
So. . . Here are my questions.
How effective is using the Node function to help "share the workload"?
Putting into consideration that the second MacPro computer will be similar to the one noted below, how many more orchestral instruments (using the Kontakt 2 software) can I expect to load within a project using the Node function?
It looks like the purchased second "refurbished" MacPro will have the Leopard operating system loaded on it. (I plan on purchasing it from Apple.) I am reluctant to "upgrade" my current computer from Tiger to Leopard. What problems/issues, if any, might one expect if the primary computer runs with the Tiger operating system and the secondary computer runs with the Leopard operating system???
Please share your trouble or success story regarding your usage of the Node function.
Thank you!
Ted

I'm able to communicate between the MacPro and the Dell workstation using Music Lab's MidiOverLan 3 software. It looks like I'll be using a similar setup for that second MacPro.
Yes, a solution like that is what I would recommend in your situation, and pipe the audio back in via an audio interface (you can go the networked audio route if you wish, but I haven't found it a particular stable or workable solution so far - it's a bit random - going via digital or even analog on a second audio interface to bring in the remote audio is more solid, although obviously your limited in the number of channels to independently bring over (you'll probably need to submix at least).
Bee Jay, thank you very much for the reply and the muchly needed "head's up"!
You're welcome!

Webservice proxy generation

I am using JDeveloper 10.1.3 to generate a proxy for an existing webservice.
After running the wizard "WebService proxy" (from Gallery), I find that not all of expected classes are being generated. (conclusion drawn after comparing classes generated against webservice provider samples)
Is there any additional step to be performed to generate all classes?
thanks,

Select the Proxy node in the Application Navigator and View the Structure Pane, which will display all the related files for the proxy node..
Regards,
Sunil..

Create OLE DB synthetic that has failover ability if proxy is down much like resource pools

As the title indicates, I would like to create a monitor that is fault tolerant and queries an OLE DB datasource. Today you can target multiple agents to perform a synthetic transaction using the OOB templates, however I only want a single agent
performing the query at any given time. If the primary proxy/agent is down, I would like the synthetic transaction to failover to the other proxy node(s). This behavior is like network or unix/linux devices which use resource pools.

You must create unhosted class or it can be unhosted singleton class (group) and target your monitor on this class.
Vladimir Zelenov | http://systemcenter4all.wordpress.com
I assume I would then create a rollup monitor so that I don't trigger an alert should one proxy be down or experience a technical issue.

Securing proxy servers

We are looking three ways of securing proxy servers for authorised hosts :
Essentially you add an <authorized-hosts> section to the <tcp-acceptor> element of the
proxy scheme and either specify the hosts or a range or specify a filter which evaluates a host against a cache that holds authorised hosts.
Does this cover the normal acceptable methods of securing proxy servers or are there any more recent/better solutions that we should consider?

Do we have any documentation/linkgs security touchpoint diagrams pertaining to security proxy server(s) based on Authorized Hosts? Also - If we have a LoadBalancer in between the client & the proxy node - how does "authorized hosts" works - which ip should we provide in the "authorized hosts" ip address/range?
Thanks
Vishy

Multiple TCP*Extend Proxy

Hi,
I am developing a C++ Client for Coherence and using the TCP*Extend to connect to the Cache.
I am able to run multiples nodes on same machine.
My goal is to add more machines to this setup (is different machine called cluster or node ?).
I would like to have a setup as explained below:
NamedCache: AQRCache
Cache Type: Near Cache (Local and distributed/remote)
Host_A:
Tcp*extend A
Node 1
Node 2
C++clientA running on Host_A talking to TCP*Extend_A.
Host_B:
Tcp*extend_B
Node 3
Node 4
C++clientB running on Host_B talking to TCP*Extend_B.
Host_C:
Tcp*extend_C
Node 5
Node 6
C++clientC running on Host_C talking to TCP*Extend_C.
Other C++ Clients:
C++clientD running on Host_D talking to TCP*Extend_A.
C++clientE running on Host_E talking to TCP*Extend_A.
C++clientF running on Host_F talking to TCP*Extend_A.
Questions:
1. To Add more machines, if I running one more instance on different machine with the same configuration file, will it work?
2. Is it possible to run seperate TCP*Extend Proxy on each Host and it will be part of the same cluster?
3. Or should all the C++ clients should talk to only one TCP*Extend Proxy?
Thanks,
NS

1. To Add more machines, if I running one more instance on different machine with the same configuration file, will it work?Yes, you can add more machines/nodes using the same configuration file.
2. Is it possible to run seperate TCP*Extend Proxy on each Host and it will be part of the same cluster?Yes, you can run multiple proxy nodes in a single cluster. Just about every production cluster has multiple proxy nodes.
3. Or should all the C++ clients should talk to only one TCP*Extend Proxy?No, there is no reason to force all C++ clients to connect to a single proxy. In fact you might consider configuring each client with the entire list of proxy servers. Each client will pick a server at random to connect to, thus ensuring that a single proxy isn't overloaded.
Also take a look at this document: http://coherence.oracle.com/display/COH35UG/Best+Practices+for+Coherence+Extend
Thanks,
Patrick

Is usage of Proxy nodes expensive ?

Similar Messages

Maybe you are looking for