ACE : Stickyness problem with http cookies

Hi,
I am facing a serious problem with stickyness in a e-commerce configuration.
Here is the setup :
An ACE load balance user requests on two Apache servers
cookie-insert is used to stick a user on one Apache server
The home page is accessed via http on port 80
On the Home page, there is a link to allowing the user to login
The login process uses SSL
During the login, backend SSL is required between the ACE and the selected Apache server
The login is a POST request to the Apache server
After a successful login, the home page is reloaded on port 80 and the name of the user should appear on the top of the page
The ACE configuration :
Two sticky groups are configured : one for HTTP acess and another for HTTPS access
Two server farms are defined, both using the same real servers, but with different ports (80 and 441)
     sticky http-cookie STICKED-TO ECOM_STICKY_TEST_HTTP
       cookie insert browser-expire
       timeout 240
       replicate sticky
       serverfarm ECOM_FARM_TEST_HTTP
          sticky http-cookie STICKED-TO ECOM_STICKY_TEST_HTTPS
       cookie insert browser-expire
       timeout 240
       replicate sticky
       serverfarm ECOM_FARM_TEST_HTTPS
     serverfarm host ECOM_FARM_TEST_HTTP
       description *** e-Commerce Test Server Farm ***
       probe ECOM_PROBE_TEST
       rserver HQCHECOM01 80
        inservice
       rserver HQCHECOM02 80
        inservice
         serverfarm host ECOM_FARM_TEST_HTTPS
      description *** e-Commerce Test Server Farm ***
      probe ECOM_PROBE_TEST
      rserver HQCHECOM01 443
       inservice
      rserver HQCHECOM02 443
       inservice
The problem :
Let analyse the sequence of events and the value of the http cookie for each of them :
When the the home page is originally loaded, the ACE selects SERVER-1
The ACE inserts the cookie "A" in the server responses
The user is sticked to SERVER-1
Then, the user tries to login and an SSL session is established with the ACE
The user sends a POST request containing the cookie "A"
A backend SSL session is established with SERVER-1
The POST request is forwarded to SERVER-1
SERVER-1 responds with a 200 OK and the ACE generates another cookie "B" as it belongs to the sticky group ECOM_STICKY_TEST_HTTPS
The client browser reloads the page on port 80 and provides the cookie "B" (the last received) !!
The ACE sees the cookie "B" but does not find it in its database for the sticky group ECOM_STICKY_TEST_HTTP
The ACE perform another load balancing decision and selects SERVER-2 ! (instead of SERVER-1)
The page is reloaded, but the name of the user does not appear on it
The question :
As it is not possible to have only one sticky group in this configuration what would be the solution to make sure that the same server is selected for http and https ?
Thank you for any hints,
Yves

Hi Gilles,
I followed your recommendation to configure static cookie entries in each sticky group, but I still experience the problem of sessions getting re-load balanced to the second server when returning from HTTPS to HTTP :
It seems that the ACE ignores the static entries !
To make my question clear, I repeat hereafter the setup and the encountered problem :
Here is the setup :
An ACE load balance user requests on two Apache servers
cookie-insert is used to stick a user on one Apache server
The home page is accessed via http on port 80
On the Home page, there is a link to allowing the user to login
The login process uses SSL
During the login, backend SSL is required between the ACE and the selected Apache server
The login is a POST request to the Apache server
After a successful login, the home page is reloaded on port 80 and the name of the user should appear on the top of the page
The ACE configuration :
Two sticky groups are configured : one for HTTP acess and another for HTTPS access
Two server farms are defined, both using the same real servers, but with different ports (80 and 443)
In the ECOM_STICKY_TEST_HTTP stick group the two following cookies are automatically generated :
R105816849   for the server HQCHECOM01
R105852786   for the server HQCHECOM02
In the ECOM_STICKY_TEST_HTTPS stick group the two following cookies are automatically generated :
R355972695   for the server HQCHECOM01
R357158616   for the server HQCHECOM02
I statically configured in the each sticky group the cookies used by the other sticky group, to allow stickiness when the browser switches from HTTP to HTTPS and vice versa :
sticky http-cookie STICKED-TO ECOM_STICKY_TEST_HTTP
cookie insert browser-expire
timeout 240
replicate sticky
serverfarm ECOM_FARM_TEST_HTTP backup WEB_REDIRECT_001
56 static cookie-value "R355972695" rserver HQCHECOM01
64 static cookie-value "R357158616" rserver HQCHECOM02
sticky http-cookie STICKED-TO ECOM_STICKY_TEST_HTTPS
cookie insert browser-expire
timeout 240
replicate sticky
serverfarm ECOM_FARM_TEST_HTTPS backup WEB_REDIRECT_001
72 static cookie-value "R105816849" rserver HQCHECOM01
80 static cookie-value "R105852786" rserver HQCHECOM02
serverfarm host ECOM_FARM_TEST_HTTP
description *** e-Commerce Test Server Farm ***
probe ECOM_PROBE_TEST
rserver HQCHECOM01 80
   inservice
rserver HQCHECOM02 80
   inservice
serverfarm host ECOM_FARM_TEST_HTTPS
description *** e-Commerce Test Server Farm ***
probe ECOM_PROBE_TEST
rserver HQCHECOM01 443
   inservice
rserver HQCHECOM02 443
   inservice
The problem :
Let analyse the sequence of events and the value of the http cookie for each of them :
When the the home page is originally loaded, the ACE selects SERVER-1
The ACE inserts the cookie "A" in the server responses
The user is sticked to SERVER-1
Then, the user tries to login and an SSL session is established with the ACE
The user sends a POST request containing the cookie "A"
A backend SSL session is established with SERVER-1
The POST request is forwarded to SERVER-1
SERVER-1 responds with a 200 OK and the ACE generates another cookie "B" as it belongs to the sticky group ECOM_STICKY_TEST_HTTPS
The client browser reloads the page on port 80 and provides the cookie "B" (the last received)
The ACE sees the cookie "B" and should use the static cookie entry to select the SERVER-1
But instead, the ACE perform another load balancing decision and selects SERVER-2 !
The page is reloaded, but the name of the user does not appear on it
LiveHTTP Trace on Firefox :
GET /ecom/medias/sys_master/8800775602206/Home-page-main-banners-video.jpg HTTP/1.1
Host: ecom.test.toto.com
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.8) Gecko/20100202 (CK-IBM) Firefox/3.5.8
Accept: image/png,image/*;q=0.8,*/*;q=0.5
Accept-Language: en-us,en;q=0.5
Accept-Encoding: gzip,deflate
Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
Keep-Alive: 300
Connection: keep-alive
Referer: http://ecom.test.toto.com/uk/en/home
Cookie: STICKED-TO=R105816849;
HTTP/1.1 200 OK
Set-Cookie: STICKED-TO=R105816849; path=/
Date: Mon, 18 Oct 2010 15:31:37 GMT
Server: Apache/2.2.13 (Red Hat)
Connection: close
Transfer-Encoding: chunked
Content-Type: image/jpeg
Here we switch on HTTPS :
https://ecom.test.toto.com/uk/en/j_spring_security_check
POST /uk/en/j_spring_security_check HTTP/1.1
Host: ecom.test.toto.com
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.8) Gecko/20100202 (CK-IBM) Firefox/3.5.8
Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
Accept-Language: en-us,en;q=0.5
Accept-Encoding: gzip,deflate
Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
Keep-Alive: 300
Connection: keep-alive
Referer: http://ecom.test.toto.com/uk/en/home
Cookie: STICKED-TO=R105816849; JSESSIONID=089DCF987DC03CAE0F516298EB886DAB.node1;
Content-Type: application/x-www-form-urlencoded
Content-Length: 75
spring-security-redirect=&j_username=yves144%40yahoo.com&j_password=junon01
Here we see cookie for the same server but for the HTTPS sticky group :
HTTP/1.1 302 Moved Temporarily
Set-Cookie: STICKED-TO=R355972695; path=/
Set-Cookie: _hybris.tenantID_=""; Expires=Thu, 01-Jan-1970 00:00:10 GMT; Path=/; HttpOnly
Date: Mon, 18 Oct 2010 15:31:39 GMT
Server: Apache/2.2.13 (Red Hat)
Location: http://ecom.test.toto.com/uk/en/home
Content-Length: 0
Connection: close
Content-Type: text/plain; charset=UTF-8
Here we switch back to HTTP :
http://ecom.test.toto.com/uk/en/home
GET /uk/en/home HTTP/1.1
Host: ecom.test.toto.com
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.8) Gecko/20100202 (CK-IBM) Firefox/3.5.8
Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
Accept-Language: en-us,en;q=0.5
Accept-Encoding: gzip,deflate
Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
Keep-Alive: 300
Connection: keep-alive
Referer: http://ecom.test.toto.com/uk/en/home
Cookie: STICKED-TO=R355972695; JSESSIONID=089DCF987DC03CAE0F516298EB886DAB.node1;
Here we see that the second server has been wrongly selected !
HTTP/1.1 200 OK
Set-Cookie: STICKED-TO=R105852786; path=/
Set-Cookie: _hybris.tenantID_=""; Expires=Thu, 01-Jan-1970 00:00:10 GMT; Path=/; HttpOnly
Set-Cookie: JSESSIONID=5A0F6EB8FBF63D5D0590FECEC62A302E.node2; Path=/; HttpOnly
Date: Mon, 18 Oct 2010 15:31:40 GMT
Server: Apache/2.2.13 (Red Hat)
Pragma: no-cache
Expires: Thu, 01 Jan 1970 00:00:00 GMT
Cache-Control: no-cache, no-store
Content-Language: en-GB
Connection: close
Transfer-Encoding: chunked
Content-Type: text/html;charset=UTF-8
http://ecom.test.toto.com/ecom/medias/sys_master/8796174057502/uk.gif
GET /ecom/medias/sys_master/8796174057502/uk.gif HTTP/1.1
Host: ecom.test.toto.com
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.1.8) Gecko/20100202 (CK-IBM) Firefox/3.5.8
Accept: image/png,image/*;q=0.8,*/*;q=0.5
Accept-Language: en-us,en;q=0.5
Accept-Encoding: gzip,deflate
Accept-Charset: ISO-8859-1,utf-8;q=0.7,*;q=0.7
Keep-Alive: 300
Connection: keep-alive
Referer: http://ecom.test.toto.com/uk/en/home
Cookie: STICKED-TO=R105852786; JSESSIONID=5A0F6EB8FBF63D5D0590FECEC62A302E.node2;
HTTP/1.1 200 OK
Set-Cookie: STICKED-TO=R105852786; path=/
Date: Mon, 18 Oct 2010 15:31:40 GMT
Server: Apache/2.2.13 (Red Hat)
Content-Length: 382
Connection: close
Content-Type: image/gif
Hypothesis :
It seems that the static entries are not considered by the ACE...

Similar Messages

Gmail error message "We've detected a problem with your cookie settings." Tried everything...

THIS IS THE PAGE:
Google
We've detected a problem with your cookie settings.
Enable cookies
Make sure your cookies are enabled. To enable cookies, follow these browser-specific instructions.
Clear cache and cookies
If you have cookies enabled but are still having trouble, clear your browser's cache and cookies.
Adjust your privacy settings
If clearing your cache and cookies doesn't resolve the problem, try adjusting your browser's privacy settings. If your settings are on high, manually add www.google.com to your list of allowed sites. Learn more
©2013 Google - Google Home - Terms of Service - Privacy Policy - Help
i have done everything listed here so many times and nothing is fixing this. Made sure cookies were enabled, cleared cache and cookies, made sure Gmail was approved in security settings, and then as a last ditch effort, I disabled all add-ons for Firefox and still no Gmail... I was so desperate that I even re-installed ff-19 and still cannot get into my email account. What is going on here? Any help or small nudge in the correct direction would be much appreciated!

I used the same thing used in the first link 950400...
Firefox > Options > Options > Privacy > remove individual cookies > typed google > remove all cookies > typed gmail > remove all cookies
restarted Firefox and now I can log into gmail...

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 advance

It 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.

ACE Stickyness problem

I am trying to configure stickyness on an ACE appliance. I can't seem to get it to work. I have tried a http cookie and a IP Netmask and can't get it to work. When I do a show stat sticky or a show sticky database I get nothing. Attached is the config of my ace.

you need to assign sticky resources to your context before you can start using it.
Use the following command to see if you have allocated sticky resources
switch/Admin# show np 1 me-stats "-slb -v" | i Stick
Num Active Sticky Entry: 1 0
Num Active Reverse Sticky Entry: 0 0
Free Sticky Entry Count: 944765 0
switch/Admin#
Gilles.

Problem with https

I used Firefox for a long time with no problems. I never played with the internal settings (from about:config). One day, with version 21, I found out that every time I do a image search in Google, I could see only 15 to 20 pics correctly and all the others appeared as grey boxes. I deleted cookies and cache but nothing changed. I tried to find a solution and in a Mozilla’s forum was said to reset Firefox from “Troubleshooting Information” menu. I did it and the problem disappeared. Then, another, worse problem, appeared. All sites were tried to get opened with https in front and in all cases Firefox was telling me that the connection is untrusted and can’t confirm that my connection is secure. I was selecting “I understand the risks” and the site was appeared but it was altered, it hadn’t the look that it should have. I did a google search to find a solution. Nothing really helpful came up. In your forum I found a setting, “browser.urlbar.autoFill” that should be turned to false. I did it and the problem was fixed is some pages. Now I can open google and some other pages with no problem but some others, like yahoo mail, that require secure connection (https) cannot get accessed. I tried to uninstall Firefox using Revo Uninstaller, to be sure that all data will be deleted. I reinstalled Firefox but the problem still exists. I uninstalled Firefox again and downloaded and installed version 22 beta 3 of Firefox with the same luck. So, what’s wrong? What should I do?

Check out why the site is untrusted (click "Technical Details to expand that section) and if this is caused by a missing intermediate certificate then see if you can install this intermediate certificate from another source.
You can retrieve the certificate and check details like who issued certificates and expiration dates of certificates.
*Click the link at the bottom of the error page: "I Understand the Risks"
Let Firefox retrieve the certificate: "Add Exception" -> "Get Certificate".
*Click the "View..." button and inspect the certificate and check who is the issuer.
You can see more Details like intermediate certificates that are used in the Details pane.
If "I Understand the Risks" is missing then this page may be opened in an (i)frame and in that case try the right-click context menu and use "This Frame: Open Frame in New Tab".
Some firewalls monitor secure (https) connections and send their own certificate instead of the website's certificate.
*ESET setup -> advanced setup -> extend web and email tree -> SSL
*SSL protocol: Do not scan SSL protocol

Problem with https internet url for Portal

Hello ,
We are running EP 7 SP 13.
We had a problem with the j2ee engine going down due to a db problem .It is ok now and and j2ee is up and running fine and the portal is accessible from the intranet.
But there is a problem with the https url that we use for internet access. It gives "page cannot be displayed error".
There is a IISproxy in place for internet access.Https certificate has not expired. Couldnt find much in the portal logs and trace or am I looking in the wrong place?
Any help would be most appreciated.
Rgds

Hi Vineeth
Check the IIS settings, , they might not have started properly.
EP5: Enabling SSL and renewing the J2EE certificate
SAP Note Number: 685306
Can you check the contents of Keystore in Visual admin,
Regards
Kaleem
Edited by: Kaleem on Aug 19, 2008 2:11 PM

InDesign problem With Https cURL Seems Very Slow at first time to handshaking to server

I am trying to connect the server using cURL from Indesign, but when i am trying to call to server the indesign takes so much time to open and initialize the socket it takes 25 seconds at first time only, but On all subsequent calls it is fast,
*but when i am tryid that same programe with an c++ empty project it is very fast in 2 second we get a result back,
*So the Problem With Indesign Socket initialization using https at first time because i have tryid same Curl On Simple c++ program and it is fast at all calls
Specification:
I am Currently using
     Indesign cs7 version 9.0
     visual studio 2010
Can Anyone tell me why It happens,
my code snippet is as follows please try to find if anyone knows better.
         curl_global_init(CURL_GLOBAL_ALL);
          curl = curl_easy_init();
         struct data config;
      config.trace_ascii = 1; /* enable ascii tracing */
      struct curl_slist *chunk = NULL;
         curl_slist_append( chunk, "Content-Type: application/json");
      curl = curl_easy_init();
      struct AppMemoryStruct chunk1;
         chunk1.memory = (char *)malloc(1); /* will be grown as needed by the realloc above */
         chunk1.size = 0;    /* no data at this point */
       if(curl) {
             curl_easy_setopt(curl, CURLOPT_URL, url.GetPlatformString().c_str());
             curl_easy_setopt(curl,CURLOPT_SSLCERTTYPE,"PEM");
             curl_easy_setopt(curl, CURLOPT_SSLCERT, "C:\\test\\omg.aps.net.pem");
             curl_easy_setopt(curl, CURLOPT_SSL_VERIFYPEER, FALSE);
          curl_easy_setopt(curl, CURLOPT_SSL_VERIFYHOST, 2);
             curl_easy_setopt(curl,CURLOPT_KEYPASSWD,"");
             char error[1024];
             curl_easy_setopt ( curl, CURLOPT_ERRORBUFFER, error );
       /* send all data to this function */
             curl_easy_setopt(curl, CURLOPT_WRITEFUNCTION, &AppWriteMemoryCallback);
      /* we pass our 'chunk' struct to the callback function */
             curl_easy_setopt(curl, CURLOPT_WRITEDATA, (void *)&chunk1);
      /* some servers don't like requests that are made without a user-agent
          field, so we provide one */
             //curl_easy_setopt(curl, CURLOPT_USERAGENT, "libcurl-agent/1.0");
             //curl_easy_setopt(curl, CURLOPT_HTTPHEADER, chunk);
             //curl_easy_setopt(curl, CURLOPT_HTTPGET,1);
     //heres the performance get down its around 25 seconds when i am calling The server Using Curl Object Having
     //https url
          res = curl_easy_perform(curl);
    // till here
             if(CURLE_OK != res)
                 errorCode = kFalse;
                 if(chunk1.memory)
                     free(chunk1.memory);
                  curl_easy_cleanup(curl);
               return errorCode;
             errorCode = kTrue;
             if(chunk1.memory)
                 returndata.SetCString(chunk1.memory);
                 free(chunk1.memory);
           /* always cleanup */
         curl_easy_cleanup(curl);curl_global_init(CURL_GLOBAL_ALL);
      please try to find a key issue why the performance is very bad using https.
Thanks,

I was tempted to move this over to the InDesign SDK forum where there are coders, but there isn't a lot of traffic over there, so I'm going to leave this open here with the advice that you probably won't get an answer from anyone here becasue we are, for the most part, not folks who work with code beyond scripting (and very few scripters hang out here, either).
You should cross-post over there yourself and hope someone who understnds the qusetion sees it in one place or the other.

Sneak Preview ABAP 2004s: Problem with HTTP-Server

Hi..
I have just installed the sneak preview.
Everything works well.
But there is a problem with the webdynpro
for abap. When i try to start the example
i become a message: "Windows can not find
://:/sap/bc/webdynpro/sap/wdt_quiz ....... "
This error is from the HTTP options.
In transaction SMMS over extert functiuns->http->display
the parameter http_status_code is -1.
Can anyone help me what i can do?

when you go into transaction SMICM->go to ->services you should atleast see one entry for HTTP
it will be something like below.
No. Protocol Service Name/Port Host Name Keep Alive Acti External Bind
1 HTTP 1081 xxxx.domain.com 30
i dont think that for this reason , you have to go for a reinstall, i am sure some config is missing.
Regards
Raja

Problems with http header "Content-Location"

Does anyone know how to override the
"Content-Location" http header. We are having issues with search engines and this header being returned from Apache/oc4j. In a nutshell, I have a site that uses the
Struts framework, where the actual urls submitted would be for example
(http://mysite/home.do), where ".do" is just a servlet mapping. When we have
tried to follow the one link that has been spidered, it actually contains
the full path that appears in the "Content-Location" header (i.e.
http://mysite/web-inf/jsps/bogus.jsp) which in this case can't even be
accessed. The feedback we get from third-party utilities that try to spider
the site is that it is stopping because it has already indexed "bogus.jsp",
which in reality will always appear since it is a template, where the actual
urls will be different as is above.
Because the "Content-Location" header is being returned to any
client hitting the site, search engine spiders stop indexing at the first page because the value in "Content-Location" is the same.
Solutions tried:
mod_headers in Apache - have tried "Header unset"
HttpServletResponse.setHeader()
Any help would be appreciated

Hi there,
i'm having a similar problem to this when trying to run some web page speed optimisation software...
i think the issue also causes problems with the Opera browser (although this may have been fixed in the latest version).
anybody any ideas how to stop the header being sent in the response?
many thanks,
Andy

Problem with http - serverlet interaction?

Hi. I'm having a bit of problem with a HttpURLConnection.
First of all, this isn't something I'm especially knowledgeable in, so..
Basically, I'm requesting a page that carries a token used for validating a form submission to prevent multiple posts. However, when I try to request said page, I get an internal server error. The same happens if I try to type the URL into my browser; the page only loads properly if I go to it via a link. However, checking the source of the page with the link shows nothing special about it, which has me a little confused. I tried making a new page and linking to the page I'm trying to load, and it loaded fine, so it seems that it will load as long as it was accessed by clicking a link.
I'm assuming this problem has to do with the serverlet needing to generate a token for the session, but since I'm requesting it from a HttpURLConnection, it's not able to get the token. I've been trying to think of a way around it, but I've got nothing.
Can anybody give me a hand here? Thanks.

You are looking for the mouseEnabled property.
So, for instance, where image is the instance name of the image on the border:
image.mouseEnabled = false;

Problems with Http-servlet : UTF-8

I am having some difficulty with UTF-8 encoded
chracaters in a Java servlet.
My servlet accepts an XML file conteining a question and returns an HTM page. The XML has cyrillic characters encoded as utf-8.. The rendering
servelt copes with this fine, and the HTML produced
displays OK in the browser (the response type on the
Java servelet has to be set to "text/html;
charset=UTF-8" for this to work).
I have to send cyrillic characters back in the
response to the question in a text field in the HTML form.The browser is
correctly sending back the byte stream (which I am
printing here as hex): d0b3d0bed180d0bed0b4 (this is a
cyrillic word correctly coded as utf-8).
However, on collecting the response (using
request.getParameterValues(fieldname)) the servlet
returns the byte stream: d0b3d0bed13fd0bed0b4.
A mistake in the fifth byte.
Can anyone help with this problem? Is there a known problem with the JAVA UTF-8 converter?
Regards
Graham

I now know the answer to this problem thanks to Bruno Van Haetsdaele .
Before calling request.getParameterValues(fieldname));
one should call request.setCharacterEncoding("UTF-8");
Hope that helps somebody else!

Problem with https WSDL locations

I am trying to import WSDL with https in the PartnerLink and get errors saying can't recognize the WSDL. WSDL locations with http work fine.
Both the WSDLs used to work fine before the network upgrade.
I had to tweak the obsetenv.bat for the proxy server setting as detailed in the TechNote #3 to get the http WSDL to work. Is there something else I should do for https ?
Thanks

yes you need to configre the jvm for jdev/and later bpel with the keystore and certificate ..
if you want to have it easier - use your browser, download the wsdl and save it locally .. into the project ..
for the runtime, you will still need to add the certificate into the keystore of the jvm used..
hth clemens

Problem with javascript cookies and frames in Safari

I have a web page with 2 frames in it. One frame holds a pulldown menu for quick navigation, which changes the content in the other frame and also writes a cookie that makes that choice the default when the user comes back to the web page. My problem is that when the page that has the navigation is in frames it will not change the cookie when the navigation choice is changed. It will only change the cookie if the navigation page is displayed on its own web page (not in frames). The javascript works fine in both FireFox and Google Chrome. It's only Safari that has this problem. Is this a bug in Safari, or do I need to code this differently? I also tried putting the javascript function in the main index.html page that sets up the frames, but that didn't help either.

HI Patrick,
Try posting in the Apple Technoligies forums here.
http://discussions.apple.com/forum.jspa?forumID=735
Carolyn

Problems with Arrowpoint cookies for clients behind a Proxy

I have in a WebSite clients being load balanced using Arrowpoint cookies to a virtual Server. The CSS load balance between three Apache real servers.
I have some clients that are behind some kind of Proxy Cache and I have seen with a sniffer that the proxies causing the problem Re-use proxy to our server connections for different requests for multiple clients.
Then, as I understand the CSS make the forwarding decission based on the cookie of the first request for the first client behind the proxy after establishing the HTTP connection, but when there is a request from other client using this same connection (that must be forwarded to other real server) the request is forwarded to the original web server and fails because we need sticky connections.
I thought that this wasn't correct but I have read some documents that say that this is called a Proxy role as a "connection cache". Then my question is if there is any workaround for this problem.
Thanks

I believe your problem is that the proxy open a few persistent connections with the CSS and loadbalance your client's request over them.
Once the CSS has associated a connection with a service, it does not look into the request anymore.
The solution is to disable persistence on the CSS with the command 'no persistent' and 'persistence reset'.
Find more info at :
http://www.cisco.com/en/US/products/hw/contnetw/ps789/products_tech_note09186a0080093e06.shtml#crp
Gilles.

Weird problem with http connection

hi,
i'm trying to implement a midlet that connects the mobile to an http server. i've downloaded several midlets and also tried the method given in the MIDP2 api documentation, but i doesn't work, it can't connect.
i tried it on several phones and some of them freeze during the Connector.open method, but the Nokia 6230i tells that i have to register to GPRS services before i can connect this way. :(
how is it that i can't connect with my midlet but i can do it through the wap function of the phone ? how can i bypass this problem ?
thanks.

i went to the parameters menu and changed "GPRS connection" to "permanent" but it didn't change anything. there doesn't to be any other paramter that could help with this. :(

ACE : Stickyness problem with http cookies

Similar Messages

Maybe you are looking for