WebPagetest Forums

Full Version: Cookies and multiple connections
You're currently viewing a stripped down version of our content. View the full version with proper formatting.
Hi,

I made a lot of test but I receive different results and don't undarstand why.

With www
1. http://www.webpagetest.org/result/100720_1KMX/
2. http://www.webpagetest.org/result/100720_1KMH/
3. http://www.webpagetest.org/result/100720_1KMQ/

Withouth www

1. http://www.webpagetest.org/result/100720_1KMV/
2. http://www.webpagetest.org/result/100720_1KMJ/
3. http://www.webpagetest.org/result/100720_1KMK/

1. The first question is why for tests with www. some or all files are cookiless and for tests withouth www all files have cookies ????Huh
2. The second question is why somethimes in the Connection View files are served only from investx.net and other thimes are served from http://www.investx.net also?? And this it seemes that don't depend if the test is made with or withouth www. HuhHuh

As I can see very big differances aren't between tests, but I also want to know what is the best solution??

1. Withoth cookies ( as I can see from comparations it seems that test with cookies are sometimes faster )..
2. With files served only from http://www.investx.net or also from investx.net. (neither here the tests aren't clear ..in some cases when files are served only from investx.net it seems it is faster ...)
The cookie difference is because the cookies are set by your base request without an explicit domain so when you go to http://www.investix.net the cookies are set to the http://www.investix.net domain and any requests to investix.net (without the www) will not include the cookies. When you browse to investix.net (without the www) the cookies are set to the investix.net domain (which includes any sub-domains) so all requests regardless of the domain will include the cookies.

As for the requests coming from both http://www.investix.net and investix.net I would bet that there is a configuration somewhere in your joomla setup where you specify the site domain and it is configured as investix.net. Any resources that just use a relative path will be served from the same domain you browsed to and any that use an absolute path will use the configured site domain and come from investix.net.

For the variability, it doesn't have anything to do with the cookies. I ran 10 tests all for the same domain (with www) and from the same browser with the tests executing on the same machine back-to-back and the first byte times for the base page varied from 0.7 sec to 1.2 sec. I expect that when the tests are run at longer intervals (and nothing is cached in the server's memory) the variations will be even wider. You need to look into your hosting and the back-end performance of the site if you want to get that under control.

-Pat
(07-20-2010 10:43 PM)pmeenan Wrote: [ -> ]The cookie difference is because the cookies are set by your base request without an explicit domain so when you go to http://www.investix.net the cookies are set to the http://www.investix.net domain and any requests to investix.net (without the www) will not include the cookies. When you browse to investix.net (without the www) the cookies are set to the investix.net domain (which includes any sub-domains) so all requests regardless of the domain will include the cookies.

As for the requests coming from both http://www.investix.net and investix.net I would bet that there is a configuration somewhere in your joomla setup where you specify the site domain and it is configured as investix.net. Any resources that just use a relative path will be served from the same domain you browsed to and any that use an absolute path will use the configured site domain and come from investix.net.

For the variability, it doesn't have anything to do with the cookies. I ran 10 tests all for the same domain (with www) and from the same browser with the tests executing on the same machine back-to-back and the first byte times for the base page varied from 0.7 sec to 1.2 sec. I expect that when the tests are run at longer intervals (and nothing is cached in the server's memory) the variations will be even wider. You need to look into your hosting and the back-end performance of the site if you want to get that under control.

-Pat

The problem with the cookie difference I understand it and I agree with your explanation. Now the only question is if it is recommended that all the visitors of my site to acces it through http://www.investx.net ?? If yes , how can I made that my site to be accesed only through http://www.investx.net ???

As for the first byte variability when you say to the back-end performance of the site you mean that I must to check the extensions,applications,modules,plugins,database etc. or what?

And please tell me how to look into my hosting and what to look for?? I have a shared host with access to the php.ini and htaccess.
You should be able to put in a redirect from investx.net to http://www.investx.net (you'd want to do that for SEO anyway so the two don't compete with each other). Some hosting providers have it as part of their panel configuration, otherwise you can do it with a .htaccess rule.

For the back-end variability (and performance) it could be anything from server/memory utilization to the database performance. I don't have a lot of experience tuning a joomla install but this site looks like it has good docs: http://www.joomlaperformance.com/compone.../Itemid,1/

-Pat
Reference URL's