|
I am new to cp thats why i put the question to the wrong place
but thks ur answer is helpful. By the passage of time i will get used to CP.
 ![Rose | [Rose]](https://codeproject.global.ssl.fastly.net/script/Forums/Images/rose.gif)
|
|
|
|
|
That's fine - I hope you feel welcome here, and get a lot of use out of the site. Again, I was only directing for your benefit, if you ask in the right place, people will see and be able to help.
Christian Graus
Please read this if you don't understand the answer I've given you
"also I don't think "TranslateOneToTwoBillion OneHundredAndFortySevenMillion FourHundredAndEightyThreeThousand SixHundredAndFortySeven()" is a very good choice for a function name" - SpacixOne ( offering help to someone who really needed it ) ( spaces added for the benefit of people running at < 1280x1024 )
|
|
|
|
|
Hi Guys,
Just wanted to check out if anyone can help me out in one of the typical scenarios that I have…
The scenario is that of a typical Zero Down Time Updates (100% availability…) on a (ASP.Net 2.0 on IIS 6.0) production site with multiple tiers involved – F5 Balanced Web Farm, SQL cluster,( Biztalk 2004 Farm & Citrix Farm, etc ,which can be ignored for now )..
We have managed to find a strategy to deploy the updates - starting with SQL Cluster followed by Web Farm Tier., assuming the updates to the SQL cluster will be necessarily backwards compatible with the earlier version of the application.
The strategy to update the Web Tier is holding us up due to few issues as mentioned later below. The steps we are following is typically explained below :
1. Firstly, isolate any one node in the web farm, by setting up Node drain out feature in F5.
2. Once the node is isolated, deploy the new version of the application/site on the node.
3. Introduce the updated Node back into the farm so that it again starts receiving the requests from the load balancer.
4. Follow the same steps from 1-3 till all the nodes in the farm are updated
Now this leads us to the situation where :
1. the farm now has two versions of the application running side-by-side, (updated node has latest version of the application along with updates business components ,while the other nodes still have older version.. ),
2. there is possibility of the new node(new application) receiving a request which is essentially the post-back request from the user who was earlier serviced by the older version of the application & vice-versa.
3. assuming, at the new node, we figure out that postback request corresponds to the older version, we can think of forwarding the request back to any of the other nodes still having the earlier version of the application (if possible at all in a load balanced web farm? I assume this my cause performance hits on the node which was directed to..)
4. need to figure out a solution wherein after detecting that postback request was meant for an earlier version of the application, somehow trigger a postback again , thereby ensuring that Load balancer will this time, route it to some other node – WITH ALL THE VIEWSTATE & POSTBACK DATE (ALSO THE RAISED POSTBACK EVENTS) intact, so that , finally the request will be executed in the way intended without any loss of data & yet happening transparently to the user.
5. need to achieve this without any change to the existing code – Cross Page Post back, Server.Transfer & even the Response.Redirect don’t seem to help as the postback data & viewstate won’t be preserved. Besides even to Cross Postback, would imply posting back by only those controls supporting with PostBackURL property (i.e. implementing IButtonControl interface), whereas the existing code might have custom controls as well…
I would truly appreciate if someone can suggest a solution or even better alternatives.
Thanks,
Sandesh
|
|
|
|
|
Can you configure affinity groups in F5 and split your farm based on version?
How many nodes do you have? Is there any time during the day where your load is light enough that you can drain-stop onto a subset of machines (say half), update half, bring the "updated half" online, drain-stop the "old half", then update?
|
|
|
|
|
Hi Mark,
Thanks for your help...
I think there are few problems with the approach you suggested as added inline ..
Can you configure affinity groups in F5 and split your farm based on version?
Assuming the Affinity is turned on, if we split the Farm into two, this will lead to two
separate Virtual IPs (I mean one for load balancer serving the older version of the web farm
& another for the newer version…), which may not be desirable..
How many nodes do you have? Is there any time during the day where your load is light enough that you can drain-stop onto a subset of machines (say half), update half, bring the "updated half" online, drain-stop the "old half", then update?
Assuming the nodes is even number N,(which implies half the nodes i.e. N/2 are serving the older
version [say A], with the other half i.e. N/2 serving the newer version[say B]), when the
offline nodes,(N/2) for B are brought online, with corresponding nodes N/2 for A set to drain
out, these N/2 nodes for A will stop receiving any further new requests, but continue to process
the exiting requests. Now the problem with this is, the moment, drain out/stop has been
configured, & user has posted back a form, this postback request will be routed to the some
other node maybe from set B, despite the affinity being turned…- which still leaves our
initial question unanswered - one, where original request was server by a older version node,
but its postback is being routed to a newer version node…
Please let me know, if there is any indiscrepancy in my understanding …
Regards,
Sandesh
|
|
|
|
|
Hmmmm I thought you could configure session affinity on a load balancer, without having to use different IPs. By this I mean that a client will keep talking to the same node for the entire session, not just per request.
This means that in the second scenario the drain-stop would take a long time, as clients ended their session one by one on group A. Meanwhile, any new clients would get affinitied (affinitated, affiliated? is that even a word?) to the B group.
I have a feeling you are right though, and the common drain-stop would override any affinity settings, and would push clients affiliated with an A node immediately over to a B node. Maybe not though, it might be worth checking the manual.
|
|
|
|
|
hi
How we can get web browser height and width information on the same page.If anybody knows plz reply me..Thanks
Rajendran.AL
|
|
|
|
|
hai
I have an asp application which retrieve data from Database and write in text file using Server.CreateObject("Scripting.FileSystemObject").It Works good.
before downloading(writing into a file), i have made some validation [retrieves data and checks data in a frontend[asp]] and recordset contains minimum 10000 records.My problem is after validation,I have tried to download a file, it shows an error 80070008 at line xx.
At line xx,
set fso=Server.CreateObject("Scripting.FileSystemObject").
I have checked all objects [connection,recordset] is closed or not.it was closed.
But without validation, it works well.
if anyone known about this,help me
Thanks in advance.
Known is a drop , Unknown is an Ocean
|
|
|
|
|
Hi all!
I need code JavaScript : about 5 second will change images and link .Please help me
Hello
|
|
|
|
|
I guess you need to use AJAX. I assume you're not using ASP.NET, so what ARE you using ?
Christian Graus
Please read this if you don't understand the answer I've given you
"also I don't think "TranslateOneToTwoBillion OneHundredAndFortySevenMillion FourHundredAndEightyThreeThousand SixHundredAndFortySeven()" is a very good choice for a function name" - SpacixOne ( offering help to someone who really needed it ) ( spaces added for the benefit of people running at < 1280x1024 )
|
|
|
|
|
Hi all, i need code for an email search engine that i am creating.Its is supposed to collect the email addresses, transfer them via ftp and store them on MySQL database, it should not disconnect the ip address used in the search. anyone with any ideas as to how or where i can get it from. 
|
|
|
|
|
mukera wrote: i need code for an email search engine that i am creating
note the contradiction ?
Christian Graus
Please read this if you don't understand the answer I've given you
"also I don't think "TranslateOneToTwoBillion OneHundredAndFortySevenMillion FourHundredAndEightyThreeThousand SixHundredAndFortySeven()" is a very good choice for a function name" - SpacixOne ( offering help to someone who really needed it ) ( spaces added for the benefit of people running at < 1280x1024 )
|
|
|
|
|
|
Hi all
I have created a very nice nested AJAX accordion menu, and am trying to iron out the last creases. As we know, when a users clicks on an accordion pane, this expands the pane and shows its content area. Now this event is hidden within the AJAX .net toolkit itself. What I require to do is, if the pane is already open, that the user cannot close it by clicking the same pane, but must instead click a different pane thus opening another pane, and in turn closing the open pane. Therefore i need to know how to catch the click event on an AJAX Accordion pane before it fires. I am aware I can hack this by after the event fire setting the "Selectedindex" of the pane back, but this is visible to user as the pane will close then open again. Any ideas?
|
|
|
|
|
By live, I mean where it can be accessed on the internet.
I am an old Windows Application developer. When I started my current job as a windows application developer, I inherited an existing website. Over the past five years I have worked on my this website. I have no problem building web applications for this website.
My problem is, the site had already been set up before I got here. My predissesors left no notes or instructions on what they did. I would like to know how it’s done. How does it go from the web server out to where it is accessed on the internet? Is it something I need to by a book for?
I would like to move my website from our old server to the new one. I would also like to upgrade it from .net 2002 to .net 2005, but those are other topics.
Robert
|
|
|
|
|
Depending on the configuration you need to do the following:
Assuming you are going to use Windows 2003 or Windows 2008 as your webserver. Be sure to read the details up in the manual of IIS 6 for windows 2003 or IIS 7 for windows 2008. This is just a basic outline of the steps involved.
Before you start you should do two things:
1. Have a static ip-address with a DNS entry attached under which the website can be accessed.
2. Have your windows server configured so that IIS is installed.
With all that setup perform the following steps to install the website for internet use.
1. Set up the windows user rights on the directory containing the site content so that the Network service user has read access.
2. Configure a new website on IIS with the dns name under which it will be reachable as the hostheader (Check docs for hostheaders in IIS for more info on how to set this up).
3. Point the home directory of the new website to the directory where to content of the website is placed on the harddrive
4. Configure anonymous access for the website so that users can view the pages on the webserver.
5. Configure additional security options like windows logon if you are using any authentication and authorization mechanisms.
Now you're all set to serve up the website.
Did I mention that this is only a global description of what should be done? Well, it is  and there might be some variations, so again: Do check the manual of IIS for the exact details.
WM.
What about weapons of mass-construction?
"What? Its an Apple MacBook Pro. They are sexy!" - Paul Watson
My blog
|
|
|
|
|
Thanks WM.
My last attempt brought the website down. After I was done having a heart attack, I got it up and running again.
I'll give this a try.
|
|
|
|
|
hey im trying to make an app that will log in to facebook ,
im using the following method to establish connection and post my credentials:
public static void start_post(string strPage, string strBuffer)
{
//Our postvars
byte[] buffer = Encoding.ASCII.GetBytes(strBuffer);
//Initialisation
HttpWebRequest WebReq = (HttpWebRequest)WebRequest.Create(strPage);
//Our method is post, otherwise the buffer (postvars) would be useless
WebReq.Method = "POST";
//We use form contentType, for the postvars.
WebReq.ContentType = "application/x-www-form-urlencoded";
//The length of the buffer (postvars) is used as contentlength.
WebReq.ContentLength = buffer.Length;
//We open a stream for writing the postvars
WebReq.Referer = "http://www.facebook.com/index.php?";
Stream PostData = WebReq.GetRequestStream();
//Now we write, and afterwards, we close. Closing is always important!
PostData.Write(buffer, 0, buffer.Length);
PostData.Close();
//Get the response handle, we have no true response yet!
HttpWebResponse WebResp = (HttpWebResponse)WebReq.GetResponse();
//Let's show some information about the response
Console.WriteLine(WebResp.StatusCode);
Console.WriteLine(WebResp.Server);
//Now, we read the response (the string), and output it.
Stream Answer = WebResp.GetResponseStream();
StreamReader _Answer = new StreamReader(Answer);
reply = _Answer.ReadToEnd();
Console.WriteLine(reply);
}
i get the following respond from facebook:
"sorry, we're not cool enough to support your browser. please keep it real with the following browsers: ...."
any idea how to work around this facebook check , as im sending same data as my IE agent i dont know what else will work , asking for your expert help.
thank you.
Net
|
|
|
|
|
|
i posted that question thought i'll have more luck here ,
there are no work arrounds mentioned on google ,
only entioned that facebook sucks for this :P ,
btw its not a problem its some protection against none cool browsers
im looking for a programatic wrok arround
Net
|
|
|
|
|
Hi there.
Not sure if this is correct messageboard, so sorry if it's isn't.
I have recently started using INtrexx Extreme at work to create portals and applications. While designing these there is a button to view your work in a browser.
When i do this, i get a browser which just says 'Service Unavailable'.
If I go into IIS and browse the website, it brings the same thing up.
I found a thread online that say's if you go into the properties of the 'My Websites' folder and check the box 'Run WWW service in IIS 5.0 isolation mode'. When I do this it solves the 'Service Unavailable' problem (although a cross appears next to the 'My Websites' folder) but when i try to view the default site again it just brings up the page cannot be displayed message.
My internet connection is working, just incase anyone thought of that.
Any help would be greatly appreciated.
|
|
|
|
|
Have you tried the vendors website? You are more likely to find your ansewr there.
Bob
Ashfield Consultants Ltd
|
|
|
|
|
Hi all,
I developed a sample web site which prints current system date time. After publishing the web site I created a virtual directory name and mapped the published folder.
When i try to access the website from browser as http://localhost/test/default.aspx[^] ,it returning me a error as Your are not authorised to view this page.
This error (HTTP 403 Forbidden) means that Internet Explorer was able to connect to the website, but it does not have permission to view the webpage.
But when i do the same process in another system it works fine.
I modified the Directory security setting by enabling the Anonymous access.... still i am getting the same error.
I check my user account details it has rights to access the IIS.
Please suggest me to over come this problem
Thanks in advance
Know is Drop, Unknown is Ocean
|
|
|
|
|
Check the following:
1. Is anonymous access enabled for the virtual directory (Check the directory security settings in IIS for this)
2. Check that the network service account has access to the physical directory (ASPNET account under Windows XP)
3. Check the security settings in your web.config, is there a line in <authorization> saying <deny users="?"/>
This should help overcome your HTTP 403 error.
WM.
What about weapons of mass-construction?
"What? Its an Apple MacBook Pro. They are sexy!" - Paul Watson
My blog
|
|
|
|
|
Can any One tell how can we implement the return url setting in the web config so that we secure our web pages.
tell me
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
