No search results were found in Documentation!
I just find some error messages from nginx and mod_pagespeed.
The error messages seems related to the theme but I am not sure about that.
These errors do not affect the output, I post this topic just for reference only.
2014/12/09 12:35:23 [alert] 3054#0: [ngx_pagespeed 1.8.31.4-4009] http://www.enterpr1se.info/:16777215:HtmlElement Parents of StartElement style do not match:
Actual: <head></head> 5…173
Expected: <div class=”td-module-comments”> (not yet closed) 1168…0
2014/12/09 12:42:14 [alert] 3054#0: [ngx_pagespeed 1.8.31.4-4009] http://www.enterpr1se.info/:16777215:HtmlElement Parents of EndElement style do not match:
Actual: <head></head> 5…173
Expected: <div class=”td_module_wrap td_module_6″ itemscope itemtype=”http://schema.org/Review”> (not yet closed) 1047…0
2014/12/09 13:13:25 [alert] 3054#0: [ngx_pagespeed 1.8.31.4-4009] http://www.enterpr1se.info/:16777215:HtmlElement Parents of EndElement style do not match:
Actual: <head></head> 5…173
Expected: <div class=”td-post-date”> (not yet closed) 1180…0
2014/12/03 00:00:40 [alert] 10334#0: [ngx_pagespeed 1.8.31.4-4009] http://www.enterpr1se.info/2014/11/interstellar-review/:16777215:HtmlElement Parents of StartElement style do not match:
Actual: <head></head> 5…779
Expected: <div class=td-next-prev-wrap> (not yet closed) 1448…0
@gomerblog . The problem isnt newspaper theme. You need to set a better config in your server. My site hits more than 150K visits per day on a little VPS whit only 4 cores and 2GB of ram. But im running on Nginx + Memcache and some configs to improve stability and speed.
Conclusion, you need to hire an SysAdmin
i found this error when i install the theme
“413 Request Entity Too Large
nginx/1.6.2”
fix it please 🙁
Hi,
You can make a backup search online for a proper nginx wordpress configuration and test it. Using the nginx cache, it may be possible that the request will be cached even via ajax.
You can also follow this topic: https://forum.tagdiv.com/topic/pagespeed/page/6/#post-15416
You can also get someone to optimize nginx settings for you.
Hope this helps!
Thanks
-
This reply was modified 11 years by
Lucian.
hi guys,
I really want to know the answer to this.
i noticed in documentation this statement:
make sure that you have APC installed or another ‘op cache’ for php. This will reduce your server response time.We have receive some reports that the accelerator has some problems in some specific configurations with our theme and Nginx.
I want to ask my server to enable but afraid something will crush.
can you let us know proper settings?
cheers!
One big post to rule them all…that would be awesome! (and very helpful)
I feel your pain. Had to do a server migration recently, as well. Looks like Centmin Mod Nginx SSL + SPDY will be my next big project.
For your site did you
1) setup the gzip/deflate for your server to serve compressed content / mandatory
2) did you add expiration data in your htaccess (Apache) file or similar for NGINX / mandatory
3) did you optimize your images for best size/speed
4) did you use functions.php to remove ?=123 version queries and set better image optimization default for WordPress; then perhaps regenerate thumbnails
5) remove external elements as needed, such as using social media counters, jetpack, or other systems which call on outside websites for data, slowing page down?
6) did you optimize your database using wp-optimize
Many of these items are covered in detail in the Page Speed sticky thread.
Items 1 and 2 are good for at least 20 points alone.
You might try this tool recommended for WordPress based websites:
http://sitecheck.sucuri.net/
Website: hm86.ru/
Status: No Malware Detected by External Scan. Additional Actions Recommended!
Web Trust: Not Currently Blacklisted (10 Blacklists Checked)
Web Server Details
Scan for: http://hm86.ru/
Hostname: hm86.ru
IP address: 69.10.55.48
System Details:
Running on: nginx
Powered by: PHP/5.2.17
Cached from: Backend
Web application details:
Application: WordPress 3.9.2 - http://www.wordpress.org
Running cPanel 11.44.1.18: hm86.ru:2082
Web application version:
WordPress version: WordPress 3.9.2
WordPress Version 3.9.x based on: http://hm86.ru//wp-admin/js/common.js
WordPress theme: http://hm86.ru/wp-content/themes/Newspaper/
List of scripts included
http://hm86.ru/wp-includes/js/jquery/jquery.js?ver=1.11.0
http://hm86.ru/wp-includes/js/jquery/jquery-migrate.min.js?ver=1.2.1
http://hm86.ru/wp-content/plugins/bbpress/templates/default/js/editor.js?ver=2.5.4-5380
http://hm86.ru/wp-content/plugins/contact-form-7/includes/js/jquery.form.min.js?ver=3.51.0-2014.06.20
http://hm86.ru/wp-content/plugins/contact-form-7/includes/js/scripts.js?ver=3.9.1
http://hm86.ru/wp-content/themes/Newspaper/js/external.js?ver=3.5
http://hm86.ru/wp-content/themes/Newspaper/js/site.js?ver=3.5
http://hm86.ru/wp-content/themes/Newspaper/wpbakery/js_composer/assets/js/js_composer_front.js?ver=3.6.8
http://html5shim.googlecode.com/svn/trunk/html5.js
And frankly if your hosting company isn’t using a virus scanner it is likely a good idea to look at another host!
First of all tagDiv Social Counter Version 2.2 cause massive loading time. When is activated the main page have big latency. Here is my test with the plugin on:
Image: http://i.imgur.com/gCXBKhf.png
Pingdom speed test: http://tools.pingdom.com/fpt/#!/cC7mI2/http://edmsoul.com/
And here are the results with the plugin off:
Image: http://i.imgur.com/v2Dfhhd.png
Pingdom speed test: http://tools.pingdom.com/fpt/#!/bAtGG0/http://edmsoul.com/
Also on the main page the navigation have disappeared after the update. I use nginx on Ubuntu vps server.
Hi,
if you are using the nginx cache, it may be possible that the request will be cached even via ajax. I would recommend that you use w3 total cache with a lower caching time like 1 hour.
Thanks,
Radu O.
Hi tagDiv,
My site running with cache system by nginx, when i add new post then old posts with ‘SIMILAR ARTICLES’ not show these new post, i think, i will convert ‘SIMILAR ARTICLES’ to ajax load ‘SIMILAR ARTICLES’, how to modify code?
Best regards,
Yup. Web server is different from a local PC since you don’t control the PHP settings, security settings, or the order things load, or external async elements.
You can’t presume anything that works on localhost will work on web; and fact that it doesn’t would likely cause one to reconsider use of “brilliant” in this context.
My experience has been – test on the web with the actual server. Some plugins are not tested against say an nginx environment vs an apache environment, or with timing of how things load in production environment on server, even though one would assume if it works on localhost or virtual machine it should work on the web, not always so.
I’ve had a lot of great plugins which break on version xxx of WordPress, and can never be used ever again. Also note that the theme has been optimized around some of the pending changes for WP 4.0, and not all plugins have done that yet.
But you asked me to look at this, so just giving you my thoughts. If it doesn’t work on the server, you need to figure out what is loading before it that breaks it.
-
This reply was modified 11 years by
simchris.
we’re using nginx in front of application server now. stable with upt op 15k pi / hour 60k pi / day
We use last modified on our sites, but your case may be different; we’re running Apache vs Nginx which is a tad different.
I guess your problem is if you’re seeing a page saying last modified 1970 vs “yesterday” that could simply be its way of saying “not updated – ignore me for recrawl” – which is likely the case you’re seeing.
It could be the cookies your site is sending on page load, where the cookie has the 1970 expiration, and the page has yesterday.
This page might be more helpful in checking your headers, dunno:
http://www.microsoft.com/search/Tools/default.aspx
check with your hosting provider … if the clock is off or something. I may be missing what your issue is.
You can check your headers for pages with various online tools like http://web-sniffer.net/
HTTP Response Header
Name
Value
Delim
Status: HTTP/1.1 200 OK
Date: Wed, 13 Aug 2014 20:49:28 GMT
Content-Type: text/html; charset=utf-8
Transfer-Encoding: chunked
Connection: close
Set-Cookie: __cfduid=d2376a9f44b520530e88bf3049c2796811407962968768; expires=Mon, 23-Dec-2019 23:50:00 GMT; path=/; domain=.cherrykoks.de; HttpOnly
Expires: Wed, 13 Aug 2014 21:49:28 GMT
Pragma: public
Cache-Control: max-age=3600, public
X-Powered-By: W3 Total Cache/0.9.4
Last-Modified: Thu, 01 Jan 1970 00:00:00 GMT
Cache-Control: private, must-revalidate
Vary: Accept-Encoding
Server: cloudflare-nginx
CF-RAY: 1597c78ace8e064a-FRA
Content-Encoding: gzip
Note if having issues with your setup it’s best to check with your CDN provider to debug any issues between hosted content on server and on the CDN and how they sync.
Hi! @manuel
The mod_pagespeed system runs at the server level as an nginx or apache module (we use apache as main system, with nginx used for some other things on same box, but not for WP sites). So, it runs above the domains, but re-writes content on the fly and then caches stuff and works with the server-based caching.
Benefits include automatic blank space removal, comment removal (e.g., “SEO powered by Yoast SEO” and other stuff you don’t want to “advertise” to the world, etc.) rewrite of image dimensions for anything resized to “fit” and cached version of the new size image, optimization of images, conversion of gif to png where necessary, DNS pre-fetch, etc., etc. — but you can read about it on the Google site.
We’re used the self-hosted module (aka “GPS” https://developers.google.com/speed/pagespeed/module ), but they also have a service to host content (aka “PSS” https://developers.google.com/speed/pagespeed/service ), which is a mix of CDN and the pagespeed modules together and you can test the before/after speed on their site to see if it benefits you (remember you have to do test couple of times for it to cache stuff). Test: http://www.webpagetest.org/compare
@christophersimmons:
what you understand by “graphics loading from the network portal vs localized”? As far as I know where isn’t any CDN in Germany thus we decided to server pictures from nginx ram with having in mind that this should be faster or at least comparable to cloudfront.
Fonts and Google Analytics are loaded from Google, I think this is a preset from the theme.
jQuery and Co are loaded from public CDN. I think this is faster than serving them myself since they already should be precached.
In addition to this the only thing loaded from external servers are ads and content tests. The content tests are programmed by myself and run completely async after the on page load event. For the ads we are working on a hack to load them async, however this isn’t stable yet. The only left over part is the iam.js script. which is running sync. This is the only party I’m not allowed to change, since it is used to track the germany publisher ranking. Really sucks that they only supply a synchronous script, but it should be reasonable.
I’m not that experienced in page speed optimizations yet, thus it would be interesting for me to know which part exactly makes the page load so horribly slow? And how can I figure this out? I think up to 40 sek on mobile devices is just not acceptable for clients.
How can I activate the mod_pagespeed setting to insert image dimensions and what is it doing?
And by the way thx for your awesome expertise!
As of 12:13pm PDT 7/29/14 the themeforest servers are working just fine. I just downloaded the “all files” zip for the theme, so you should have no problem unless it’s a local issue on your network/PC.
It is urgent please give me the plugins only somewhere else because the themeforest’s servers are not working.
Hi Radu, i checked out the php error log and found out that it was both of the td_speed_booster and td_social_counter plugins. This was happening because i transferred the website with latest version of theme but the plugins were outdated and not working correctly. But there’s a problem. I have to migrated the website and the themeforest’s servers are not working and i am unable to download the full theme files. The server hangs after merely reaching to 70%.
Please give me both of the plugins.zip file over dropbox please because it is necessary.
Hi,
please provide your nginx error log file also try to enable wp_debug. The html of the theme is not complete http://screencast.com/t/d6zD483YAWf .
I’ve tested and run the theme on nginx.
Radu O.
Please, it’s urgent because i’ve to transfer my website really fast. Please give me solution to this problem.
The URL for the server is http://104.131.133.193/
Hello, I’ve just migrated my website from an apache based web server to nginx based web server. Previously, everything was working correctly but now, i’ve replaced my theme with the latest version from themeforest and since then, the theme is not working properly. Like the footer is missing from the homepage and the search is not working and many other problems. My website is not live on the new server but i’ve installed the theme and migrated the data. You can check the website here – http://http://104.131.133.193/ Please suggest anything because i’ve to migrate my website as soon as possible.
If you’re having trouble optimizing via plugin, I would likely contact your hosting provider and optimize your dbase from phpmyadmin or similar outside of wordpress (e.g., command line, or from server side).
But yes… on a 190MB dbase, WP-Optimize clears out on average about 27.5MB of crap, depending on when you do an optimization.
I think the pagespeed insights reference to browswer caching is having
1) expires headers from the server (or via htaccess on Apache, or config file on nginx), which specify a time to live, such as “cache fonts for one week, don’t reload them every time a page loads” … that is the purpose of “browser caching” — same with CSS, images like a logo, javascript, etc. — my settings are in the sticky for pagespeed.
2) remember to have gzip/deflate running on server side, you don’t need plugin for that as the server can compress everything on request
I’m not currently using W3 cache, and frankly I’d leave that off until you get everything else sorted as otherwise you end up caching incomplete pages from the other errors or overloads. Once EVERYTHING else is solid, THEN you would turn on the cache as the LAST thing in the chain.
Hope that helped a little (been there!). 🙂
-
This reply was modified 12 years by
simchris.