getting 403 forbidden (guru mediation)

July 26, 2017, 2:02 pm

≫ Next: Test results vs. Chrome DevTools

≪ Previous: Setting up a Webpagetest Relay Server

$

0

0

I'm consistently getting

Quote:Error 403 Forbidden

Forbidden
Guru Mediation:

Details: cache-iad2122-IAD 1501102044 1042052109

Varnish cache server

when trying to access any page on webpagetest.org on my Ubuntu 16.10 LTS laptop when using Firefox in private mode. Anything I can do to fix?

thx,
frank

---

Test results vs. Chrome DevTools

July 26, 2017, 11:08 am

≫ Next: Need help to improve the Page load time

≪ Previous: getting 403 forbidden (guru mediation)

$

0

0

I ran a test on a webpage using my local Chrome browser with F12 developer tools turned on. Page load was 3.07 seconds. Using the instance of Chrome out of Dulles, I got almost double the load time: 6.3 seconds. Is there a difference in what these two tools are measuring?

Need help to improve the Page load time

July 27, 2017, 5:01 pm

≫ Next: New create-update.ps1 not checked in?

≪ Previous: Test results vs. Chrome DevTools

$

0

0

Here is my page result

http://www.webpagetest.org/result/170727_H1_1SDQ/

Can anyone help me how to reduce the page load time?

https://beassistance.com

New create-update.ps1 not checked in?

July 28, 2017, 12:39 pm

≫ Next: public WPT instance and blocking requests

≪ Previous: Need help to improve the Page load time

$

0

0

I was wondering if anyone else uses this script to create the wptupdate.zip file for updating agents.

When I run it, it doesn't seem to pull in all the files that are in the public update file: http://www.webpagetest.org/work/update/wptupdate.zip

I suppose I can put in my own files and regenerate the hashes for them, but it would be great to get the updated script if there are any other changes in it.

public WPT instance and blocking requests

July 28, 2017, 1:37 pm

≫ Next: Help needed for webprog24

≪ Previous: New create-update.ps1 not checked in?

$

0

0

Hello,

I was trying to test a page using the public WPT instance . I wanted to block out several beacon related domains and thus listed them, space delimited, in the block tab. The test completed but the blocked domains specified were still fetched for beacons by WPT and showed up in the waterfall. Our private WPT instance does not have this issue and blocks out the beacon domains correctly.

Can someone comment on their having successfully tested blocked domains using the public WPT instance?

Thanks

Help needed for webprog24

July 28, 2017, 2:49 pm

≫ Next: Unable to click image link using webpagescript

≪ Previous: public WPT instance and blocking requests

$

0

0

Hello,

I run the website https://www.webprog24.com and I am beginning to bang the head against the wall because the page speed is low. I did really a lot:

• Varnish
• Memchached
• APC Cache
• Cloudflare
• Compression of JS and CSS
• whatever
  

The problem is the website is dramatically slow. It just does not work the way I want. I have a few small things I can do but in general, I really did a lot to increase the page speed of this website. Can anybody give me tips on how to improve the page speed there?

Unable to click image link using webpagescript

August 1, 2017, 4:34 am

≫ Next: How to compare two test results

≪ Previous: Help needed for webprog24

$

0

0

hi,

For one for our project we need to click the image link so that it will navigate to the next page. I used the below code but it is not working. Could any one suggest how to click image link using webpagescript.

logData 1
navigate https://yyy.xxx.com/store/
logData 0
setValue id=keyword mobile
logData 1
clickAndWait id=searchSubmitButton
logData 1
clickAndWait id=https://yyy.xx.com/stibo/smallweb/std.lang.all/89/66/4678966.jpg

How to compare two test results

August 1, 2017, 7:35 am

≫ Next: Gaps in Waterfall

≪ Previous: Unable to click image link using webpagescript

$

0

0

Hi,

I have executed two tests with 9 transactions each. When I try to compare these tests , I am able to see only first transctions in the compare mode.. Is it possible to see/compare all the transactions on the compare view. Kindly suggest

---

Gaps in Waterfall

August 1, 2017, 9:46 am

≫ Next: Got Access Denied for Downloading

≪ Previous: How to compare two test results

$

0

0

I searched through and found a few other threads on this topic, but unfortunately nothing really cleared it up for me.

I'm seeing small (and sometimes large) gaps in my test results. Often the CPU Utilization is maxed out during these gaps, so it seems like the site is trying to do something.

Here is an example I just ran: https://www.webpagetest.org/result/17080...view_step1

I've also attached some other interesting screenshots that also illustrate this.

I've seen recommendations for using the tcpdump and netlog, but when looking at those files, I really can't understand them.

I am using HTTP2 Server Push as well as a Service Worker on this WordPress website.

When I view/time the website myself, it loads as fast as a quarter of a second on repeat view, but WPT is reporting 4.5+ seconds for repeat view. I'm just trying to understand this a little bit more to see if I can find what to focus on next to optimize more.

I'm happy to provide more data if needed.

  ad263ee2-2f57-11e7-8cb4-d0ef64f5883c.png (Size: 470.78 KB / Downloads: 8)

  b0b81b20-2f57-11e7-823f-b522b345d7dd.png (Size: 496.67 KB / Downloads: 8)

  154add14-2fe6-11e7-9d3c-e609b0409eed.png (Size: 691.25 KB / Downloads: 8)

Got Access Denied for Downloading

August 3, 2017, 4:05 am

≫ Next: CDN not showing

≪ Previous: Gaps in Waterfall

$

0

0

I'm writing a web-extension on Firefox Nightly to automatically download some files in the background using downloads API[1].
It works well except when Firefox is launched by WebPagetest, complaining it doesn't have write permission to user's Downloads folder even thought I've set the full access to Everyone.

Does anyone know if WebPagetest is doing any protection like this? I didn't find any code doing this, just want to make sure I'm right, thanks.

[1] https://developer.mozilla.org/en-US/Add-...s/download

CDN not showing

August 4, 2017, 9:49 am

≫ Next: Need to understand Cache static content and first byte time

≪ Previous: Got Access Denied for Downloading

$

0

0

I installed Key CDN on https://cigarleisure.com and refreshed the page and could see assets being delivered from the subdomain (pull zone) created there. Yet those urls show as the originals on the test and the page still shows an x for "Effective Use of CDN".

Need to understand Cache static content and first byte time

August 4, 2017, 11:54 am

≫ Next: My First Bytes is too long!

≪ Previous: CDN not showing

$

0

0

Hi guys!
Looks like a great helpful forum here! My results are here https://www.webpagetest.org/result/170804_QN_1KEQ/ I am trying to understand what I need to change to get my D and F to A or B

Any input really appreciated!


Kevin

My First Bytes is too long!

August 6, 2017, 10:59 pm

≫ Next: No China test location

≪ Previous: Need to understand Cache static content and first byte time

$

0

0

Hi ,

I have a page which not well behave and recently go worst! The first bytes received is after 15seconds!
https://www.webpagetest.org/result/17080...view_step1

Any hint/clue about this case? Im using AWS EC2 m4.4xlarge

No China test location

August 8, 2017, 5:36 am

≫ Next: iOS testing with public instance

≪ Previous: My First Bytes is too long!

$

0

0

Hi,
A few months ago there was a test location in Beijing (China). But this location is currently not available and there is no other in China, except Hong Hong that is outside the GFW.
I know that it is really complicated to maintain this test server, but it is very useful as it is the only way to test in the restricted context of China.
Can somebody give some feedback on a possible (near ?!?) come back of the Beijing server ?
Thanks,
Eric

iOS testing with public instance

August 10, 2017, 3:32 am

≫ Next: Install/configure Docker images

≪ Previous: No China test location

$

0

0

Hello,
I am trying to test a website with public instance.
I am running a test location for the public instance of WebPagetest.
What value for location I should use for iOS devices?

below is my script:
values = {
'fvonly' : 1,
'f' : "xml",
'runs' :1,
'requests':1,
'k':"xxx",
'location':"Dulles:iOS"
}

---

Install/configure Docker images

August 9, 2017, 11:51 am

≫ Next: Chrome on LInux - Slow fullyLoaded time

≪ Previous: iOS testing with public instance

$

0

0

Hello! I am attempting to run the WPT server and agent docker images, and having some trouble getting things connected. First (and most importantly), is there any documentation around working with the official images?

Here's the specific problem I'm currently trying to solve:

I have the server and agent containers running locally. The agent is polling the server, but the locations dropdown does not show the configured location. /install shows 0 agents connected for that location.

However, if I use the ?hidden=1 param I'm able to kick off a test, and the /install page then shows 1 agent connected. The test runs for as much as an hour and finishes with the "Test Data Missing" error.

I suspect there's some configuration mismatch between locations.ini on the server and wptdriver.ini on the agent, but I'm unable to find wptdriver.ini in the agent container.

Any ideas?

Chrome on LInux - Slow fullyLoaded time

August 11, 2017, 1:20 pm

≫ Next: Error connecting to dev tools interface

≪ Previous: Install/configure Docker images

$

0

0

I'm experimenting with moving my WPT agents from windows to Linux but have encountered a potential blocker. Some pages are now showing fully loaded times of 30+ seconds when they were consistently ~ 3 seconds. The waterfalls show no network activity during the delay. The result is that test execution times have skyrocketed and agent throughput has tanked. Is this an issue with Chrome on Linux? Is there a way I can tell WPT to not wait so long? There is an API parameter to stop the test at document complete, but this is too soon.

Error connecting to dev tools interface

August 14, 2017, 12:06 pm

≫ Next: page load results showing slower than local

≪ Previous: Chrome on LInux - Slow fullyLoaded time

$

0

0

Hello!

I'm new. I've been trying to run some tests (mobile, samsung 5 out of Dulles) and I keep getting Test Error: Error connecting to dev tools interface.

Is there something I'm missing

page load results showing slower than local

August 16, 2017, 8:41 am

≫ Next: startRender time more than docComplete for IE 11

≪ Previous: Error connecting to dev tools interface

$

0

0

Hi,

I have been running some page test results for a client and I have used a lot of different tools.

The results I get back from WPT are excellent but I am seeing a much slower response from WPT than I am from in Chrome and from a server running pagespeed.io.

I know that Page Load is generally not a RUM measurement, however it is something that the client cares about.

In Chrome and PageSpeedIO the load time is around 7 seconds with a TTFB of 209ms. However in WPT I am seeing generally 12-25 seconds Load with a first byte of 353ms.

I have changed the location from Dulles to London and Manchester and played around with the bandwidth settings to match that of Chrome & PS.io (30000 down 15000 up) but I still get similar results even with max repeat's using cache?.

Is this standard for WPT as I'm having trouble selling WPT as a reliable tool particularly when the web results are not even comparable. I understand you could argue that they are more indicative of an average user, but I wouldn't expect to see such a large discrepancy between the load times.

Has anyone else seen this?

startRender time more than docComplete for IE 11

August 17, 2017, 12:55 am

≫ Next: Ideal AWS templates for Webpagetest private instance

≪ Previous: page load results showing slower than local

$

0

0

I ran a test on a website using the wpt API, for IE11 and Chrome. For IE11, I am getting start render higher than doc complete time. I am getting such result constantly for the locations NewYork, Dulles, San Jose and Virginia.

Any suggestion on, why I am seeing render time higher than doc complete is welcome. I am getting doc complete time less while running manually as well as with API on IE. I get right data when executing a test for Chrome and Firefox.

  wpt_startrender_more.JPG (Size: 56.78 KB / Downloads: 1)

  wpt_startrender_more_in_repeatView.JPG (Size: 51.63 KB / Downloads: 1)