Hello everyone, recently we have changed the theme of our wordpress website and we have installed Newspaper.
We have the website hosted on a LAP server and the database that runs on a dedicated one.
Both are Linux-Debian based server and are in the same LAN.
Since the change of the theme, we have noticed a very significant increase of the traffic volume from the database server and a little increase of the traffic to the DB.
The increase of the traffic from the DB is about five times as much (from 5 to 20/25 Mbps)
At the same time we detected an increase (about 50%, from 2 to 3 Mbps) of the traffic to the Internet and a little decrease of the traffic from Internet.
The question is: “is that normal?”.
It’s quite obvious that the increase of the traffic is related to the new theme (contents didn’t change).
[IMG]http://i60.tinypic.com/4r2d06.png[/IMG]
[IMG]http://i58.tinypic.com/ny7hwn.png[/IMG]
Thank you in advance
Depends on old theme vs new theme ….. obviously.
If you look at a snapshot of your old theme post page layout, count how many queries are being done for each section of the page. Do the same with the new theme.
For example on one of our sites, we changed the sidebar module to NOT show thumbnails with each item on the “latest items” list. We chose not to use megamenu. Etc.
Themes are not apples to apples unless identical layout and number of queries per page.
Visual Composer does add some overhead, sadly. But caching and mod_pagespeed, etc. help with that.
Not sure if that helped in any way, just some food for thought 🙂
Hi Chris,
we are experiencing troubles. The site goes on timeout often, we really don’t know if it’s some apache settings, server limitated resource, or a theme problem. Do you know someone who can help us with the server and apache settings to work perfectly?
I contacted my host, the site works for 2 days, but then it goes on timeout again, with some user that can access, and other that cannot access to the site.
Serious issues 🙁
F
You might try one of the plugin profilers to see which plugin is using the most resources. This is what I’ve often done to debug a “runaway” plugin. I had a related posts plugin which was causing major headaches once upon a time; a bad version of Yoast SEO was causing issues; people trying to hack something caused overloads.
I also recommend using things like Limit Login Attempts to stop bots from running guessing games on your login 10,000 times in a row which can kill the server right there. I also disable pingbacks completely here for same reason.
Hi Chris! Thank you for all your suggestion, the server seems to be better now, unfortunately we have issue on speed of the site again:
https://developers.google.com/speed/pagespeed/insights/?url=http%3A%2F%2Fwww.rnbjunk.com%2F
http://www.webpagetest.org/result/150129_7M_MXV/
http://gtmetrix.com/reports/www.rnbjunk.com/XdWKMwAw
We have activated TagDiv Speed, Speed Booster Pack, W3 Total Cache (Page CDN Browser CLOUDFLARE).
Well, all those ad networks, and the stuff like variables on the cdn, etc., are stuff “off the site” which are difficult to optimize. Generally once you have more than 3 ad networks and affiliate systems running, it will kill your speed. Trade off on income vs speed, and you know which one usually will win there 😉 … often you can try to address some of those issues with the ad networks.
The weirdo thing with the http + https issue with Google ads, I’ve not yet been able to sort out, as we’re seeing that too for the first time as of December/January. Maybe Jan. 1 ? It’s possible Google is doing some kind of redirect for legacy ad units in their network to force https loading of http content coming off client servers (client http://image.jpg –> https://google.image.jpg) but guessing on that.
80 mobile and 90 desktop are the “ideal” … but arguably you can test your score with AdSense “off” to get a more realistic score since we’ve seen that Google doesn’t penalize your score in real world use for their own sites not being optimized for their own testing system (!).
Too much fun!
