Hi @bradwilcox,
I reported the resource usages in the similar topic at causing high resource consumption?.
Please try the followings:
1. Disabling DNS reverse lookup
DNS lookup spends server resources relatively high. So if you enable “Block by country” on “Front-end target settings” section, replace HOST to *. Please refer to the document “UA string and Qualification” about HOST.
2. Disabling “Record validation statistics” on “Record settings” section.
3. Disabling “Record validation logs” or “Only when blocked” on “Record settings” section.
I think that No.1 is a suspect of this issue. So I should provide an option to disable/enable DNS lookup in the next release.
I’d appreciate if you provide me some feedback about the above results.
Thank you for heads up!
P.S. Related to the number of instances, Apache also makes multiple HTTP daemon to listen the requests and run PHP. I think it’s quite normal.
Thank you for the reply.
I think I already had DNS reverse lookup disabled: it seems to be disabled by default in IIS8.
I would like to initially keep both record setting options on for now, because it is helping me watch who is trying to hack my site.
Could you explain what you wrote about the HOST and * ? I am not quite understanding you. Currently in that setting field all I have is the default:
Google:HOST,bot:HOST,slurp:HOST
spider:HOST,archive:HOST,*:FEED
*:HOST=embed.ly,Twitterbot:US,Facebot:US
If you want to make changes in that filed perhaps you would edit the above so I can copy paste it?
By the way, I don’t want to give twitter or facebook or anything other than google access to crawl the site.
-
This reply was modified 9 years, 1 month ago by
dbarraza.
Hi @bradwilcox,
> it seems to be disabled by default in IIS8.
OK. Just to be sure, please search “66.249.66.1” (googlebot.com). If the result of “host” indicate “66.249.66.1”, that’s OK. But if it indicates something like “crawl-66-249-66-1.googlebot.com”, then the fallback function in this plugin works. It depends on how DNS lookup is disabled in your server.
Regarding HOST and *, this plugin never try to get PTR record for DNS reverse lookup if “UA string and qualification” does not include “HOST“. For example, Google:* just checks if user agent string includes Google and never verify its host.
Please copy and paste the followings:
Google:*,bot:*,slurp:*
spider:*,archive:*,*:FEED
embed.ly:*,Twitterbot:US,Facebot:US
And the last thing I should notice is “Validation timing” in “Validation rule settings“. Please try “mu-plugins” (ip-geo-block-mu.php). You can refer to the document “Validation timing“.
Thanks.
Thank you for the help. The server’s CPU has stabilized finally today at about 2% but even better it’s smooth now. What I pasted seemed to have helped even though I think I had other issues outside of your plugin that I also solved. LOVE your plugin. What a relief to block a bunch of foreign traffic that was all illegitimate anyway.