Skip to content

NODEBB: Nginx error performance & High CPU

Solved Performance
69 3 62.9k 2
  • @phenomlab

    number of user is better (408) but a lot of loose connexion. navigation is hard

    wrote on last edited by
    #12

    I have chaneg nginx conf with :

    worker_rlimit_nofile 70000;

    events {
    worker_connections 65535;
    multi_accept on;
    }

    CF is under attack mode

  • I have chaneg nginx conf with :

    worker_rlimit_nofile 70000;

    events {
    worker_connections 65535;
    multi_accept on;
    }

    CF is under attack mode

    wrote on last edited by phenomlab
    #13

    @DownPW I still have access to your Cloudflare tenant so will have a look shortly.

    EDIT: I am in now - personally, I would also enable this (and configure it)

    d85820ea-6643-49bd-98da-a8537e970f04-image.png

    b6a188a9-deba-4980-b4a1-99df7975160d-image.png

  • @DownPW I still have access to your Cloudflare tenant so will have a look shortly.

    EDIT: I am in now - personally, I would also enable this (and configure it)

    d85820ea-6643-49bd-98da-a8537e970f04-image.png

    b6a188a9-deba-4980-b4a1-99df7975160d-image.png

    wrote on last edited by DownPW
    #14

    @phenomlab I have already activate it and add a waf rules for russian country

    2e3108dc-8d68-4e48-91c7-e1c3bc00c229-image.png

    With this bots settings :
    e12409dd-df54-4def-b998-470786c3afa9-image.png

    and this settings for ddos protection :

    2e176374-f7d1-48e4-b38f-83360a0f182a-image.png

  • @phenomlab I have already activate it and add a waf rules for russian country

    2e3108dc-8d68-4e48-91c7-e1c3bc00c229-image.png

    With this bots settings :
    e12409dd-df54-4def-b998-470786c3afa9-image.png

    and this settings for ddos protection :

    2e176374-f7d1-48e4-b38f-83360a0f182a-image.png

    wrote on last edited by
    #15

    @DownPW said in NODEBB: Nginx error performance & High CPU:

    I have already activate it

    Are you sure? When I checked your tenant it wasn’t active - it’s from where I took the screenshot 😁

  • @DownPW said in NODEBB: Nginx error performance & High CPU:

    I have already activate it

    Are you sure? When I checked your tenant it wasn’t active - it’s from where I took the screenshot 😁

    wrote on last edited by
    #16

    @phenomlab

    yep I activate it after 😉

  • @phenomlab

    yep I activate it after 😉

    wrote on last edited by DownPW
    #17

    For your information @phenomlab ,

    • I have ban via iptables suspicious ip address find on /etc/nginx/accesss.log and virtualhost access.log like this : iptables -I INPUT -s IPADDRESS -j DROP
    • Activate bot option on CF
    • Create contry rules (Russie and China) on CF WAF
    • I left under attack mode option activated on CF
    • I have just change nginx.conf like this for test (If you have best value, I take it ! ) :
    worker_rlimit_nofile 70000; 
    
    events {
    
    	worker_connections 65535;
    	multi_accept on; 
    }
    
    http {
    
    	##
    	# Basic Settings
    	##
    
    	limit_req_zone $binary_remote_addr zone=flood:10m rate=100r/s; 
    	limit_req zone=flood burst=100 nodelay; 
    	limit_conn_zone $binary_remote_addr zone=ddos:10m; 
    	limit_conn ddos 100;
    

    100r/s iit’s already a lot !!

    and for vhost file :

    server {
    	.....
    
            location / {
    				
    				limit_req zone=flood; #Test 
    				limit_conn ddos 100; #Test 
    }
    

    –> If you have other ideas, I’m interested
    –> If you have better values ​​to use in what I modified, please let me know.

  • For your information @phenomlab ,

    • I have ban via iptables suspicious ip address find on /etc/nginx/accesss.log and virtualhost access.log like this : iptables -I INPUT -s IPADDRESS -j DROP
    • Activate bot option on CF
    • Create contry rules (Russie and China) on CF WAF
    • I left under attack mode option activated on CF
    • I have just change nginx.conf like this for test (If you have best value, I take it ! ) :
    worker_rlimit_nofile 70000; 
    
    events {
    
    	worker_connections 65535;
    	multi_accept on; 
    }
    
    http {
    
    	##
    	# Basic Settings
    	##
    
    	limit_req_zone $binary_remote_addr zone=flood:10m rate=100r/s; 
    	limit_req zone=flood burst=100 nodelay; 
    	limit_conn_zone $binary_remote_addr zone=ddos:10m; 
    	limit_conn ddos 100;
    

    100r/s iit’s already a lot !!

    and for vhost file :

    server {
    	.....
    
            location / {
    				
    				limit_req zone=flood; #Test 
    				limit_conn ddos 100; #Test 
    }
    

    –> If you have other ideas, I’m interested
    –> If you have better values ​​to use in what I modified, please let me know.

    wrote on last edited by phenomlab
    #18

    @DownPW my only preference would be to not set worker_connections so high

  • @DownPW my only preference would be to not set worker_connections so high

    wrote on last edited by
    #19

    @phenomlab

    Ok so what value do you advise?

  • @phenomlab

    Ok so what value do you advise?

    wrote on last edited by phenomlab
    #20

    @DownPW you should base it on the output of ulimit - see below

    https://linuxhint.com/what-are-worker-connections-nginx/#:~:text=The worker_connections are the maximum,to accommodate a higher value

    With that high value you run the risk of overwhelming your server.

  • @DownPW you should base it on the output of ulimit - see below

    https://linuxhint.com/what-are-worker-connections-nginx/#:~:text=The worker_connections are the maximum,to accommodate a higher value

    With that high value you run the risk of overwhelming your server.

    wrote on last edited by DownPW
    #21

    @phenomlab

    Thanks mark 😉

    My ulimit is 1024, so I will set it to 1024

  • @phenomlab

    Thanks mark 😉

    My ulimit is 1024, so I will set it to 1024

    wrote on last edited by
    #22

    @DownPW And the worker_processes value ? I expect this to be between 1 and 4 ?

  • @DownPW And the worker_processes value ? I expect this to be between 1 and 4 ?

    wrote on last edited by DownPW
    #23

    @phenomlab

    worker_processes auto;
    
  • @phenomlab

    worker_processes auto;
    
    wrote on last edited by
    #24

    @DownPW ok. You should refer to that some article I previously provided. You can probably set this to a static value.

  • @DownPW ok. You should refer to that some article I previously provided. You can probably set this to a static value.

    wrote on last edited by DownPW
    #25

    @phenomlab

    Ok I will see it for better worker_processes value

    I add a rate limite request and limit_conn_zone on http block and vhost block :

    – nginx.conf:

    http {
    
    	#Requete maximun par ip 
    	limit_req_zone $binary_remote_addr zone=flood:10m rate=100r/s; 
    	#Connexions maximum par ip 
    	limit_conn_zone $binary_remote_addr zone=ddos:1m;
    
    -- vhost.conf :
            location / {
    				
    				limit_req zone=flood burst=100 nodelay; 
    				limit_conn ddos 10;
    

    –> I have test other value for rate and burst but they cause problem access to the forum. If you have better, I take it

    I add today a proxy_read_timeout on vhost.conf (60 by default)

    proxy_read_timeout 180;
    

    I have deactivate underattack mode on CF and change for high Level

    I have add other rules on CF waf :

    9059db6c-f4f3-4415-bb47-ab88b2b41861-image.png

  • @phenomlab

    Ok I will see it for better worker_processes value

    I add a rate limite request and limit_conn_zone on http block and vhost block :

    – nginx.conf:

    http {
    
    	#Requete maximun par ip 
    	limit_req_zone $binary_remote_addr zone=flood:10m rate=100r/s; 
    	#Connexions maximum par ip 
    	limit_conn_zone $binary_remote_addr zone=ddos:1m;
    
    -- vhost.conf :
            location / {
    				
    				limit_req zone=flood burst=100 nodelay; 
    				limit_conn ddos 10;
    

    –> I have test other value for rate and burst but they cause problem access to the forum. If you have better, I take it

    I add today a proxy_read_timeout on vhost.conf (60 by default)

    proxy_read_timeout 180;
    

    I have deactivate underattack mode on CF and change for high Level

    I have add other rules on CF waf :

    9059db6c-f4f3-4415-bb47-ab88b2b41861-image.png

    wrote on last edited by
    #26

    @DownPW what settings do you have in advanced (in settings) for rate limit etc?

  • @DownPW what settings do you have in advanced (in settings) for rate limit etc?

    wrote on last edited by
    #27

    @phenomlab said in NODEBB: Nginx error performance & High CPU:

    @DownPW what settings do you have in advanced (in settings) for rate limit etc?

    In cloudflare ?

  • wrote on last edited by DownPW
    #28

    @phenomlab

    I wanted to test awstats on virtualmin with root account and it hasn’t updated since August 2022.

    I wanted to regenerate the files but I have a problem of rights.

    df12b039-076d-4e64-9a8d-ad6580ed57dc-image.png
    4030eec2-29ce-4dcf-83f6-adc60f87be5e-image.png

    What do you think ? and how t ore-generate a rapport correctly

    I would like to use it to better manage the @ips that connect to the server

  • @phenomlab said in NODEBB: Nginx error performance & High CPU:

    @DownPW what settings do you have in advanced (in settings) for rate limit etc?

    In cloudflare ?

    wrote on last edited by
    #29

    @DownPW no, sorry - in NodeBB ACP

  • @phenomlab

    I wanted to test awstats on virtualmin with root account and it hasn’t updated since August 2022.

    I wanted to regenerate the files but I have a problem of rights.

    df12b039-076d-4e64-9a8d-ad6580ed57dc-image.png
    4030eec2-29ce-4dcf-83f6-adc60f87be5e-image.png

    What do you think ? and how t ore-generate a rapport correctly

    I would like to use it to better manage the @ips that connect to the server

    wrote on last edited by
    #30

    @DownPW from recollection, awstats is accessible via the web front end, so (for example) https://mydomain.com/awstats

    However, as you are pushing everything into a reverse proxy you’ll need to add a custom route in the nginx.conf file you are using on the website so this can be rendered outside of NodeBB.

  • @DownPW no, sorry - in NodeBB ACP

    wrote on last edited by
    #31

    @phenomlab said in NODEBB: Nginx error performance & High CPU:

    @DownPW no, sorry - in NodeBB ACP

    Here is the value

    28e6dd12-e727-4a6a-a06f-b7c7dd3ed431-image.png


Did this solution help you?
Did you find the suggested solution useful? Support 💗 Sudonix with a coffee ☕
If your organisation needs deeper expertise around infrastructure, security, or technology leadership, learn more about Phenomlab Ltd. Many of the deeper technical guides behind Sudonix are published there.

Related Topics
  • Planned sunset of NTFY plugin

    Pinned Announcements push nodebb ntfy
    7
    1
    8 Votes
    7 Posts
    4k Views
    phenomlab
    I’ve noticed that I’m the only one subscribed to the push notifications on this site. If you were using NTFY previously, and have noticed that you’ve not had any alerts for a while, it’s because this feature has been disabled. You’ll now need to use the push notification to replace NTFY as mentioned in the first post.
  • NodeBB socket with CloudFlare

    Solved Performance socket cloudflare nodebb
    24
    3 Votes
    24 Posts
    14k Views
    DownPW
    Solved. Tuto here and here
  • SEO and Nodebb

    Performance nodebb seo
    2
    2 Votes
    2 Posts
    2k Views
    phenomlab
    @Panda It’s the best it’s ever been to be honest. I’ve used a myriad of systems in the past - most notably, WordPress, and then Flarum (which for SEO, was absolutely dire - they never even had SEO out of the box, and relied on a third party extension to do it), and NodeBB easily fares the best - see below example https://www.google.com/search?q=site%3Asudonix.org&oq=site%3Asudonix.org&aqs=chrome..69i57j69i60j69i58j69i60l2.9039j0j3&sourceid=chrome&ie=UTF-8#ip=1 However, this was not without significant effort on my part once I’d migrated from COM to ORG - see below posts https://community.nodebb.org/topic/17286/google-crawl-error-after-site-migration/17?_=1688461250365 And also https://support.google.com/webmasters/thread/221027803?hl=en&msgid=221464164 It was painful to say the least - as it turns out, there was an issue in NodeBB core that prevented spiders from getting to content, which as far as I understand, is now fixed. SEO in itself is a dark art - a black box that nobody really fully understands, and it’s essentially going to boil down to one thing - “content”. Google’s algorithm for indexing has also changed dramatically over the years. They only now crawl content that has value, so if it believes that your site has nothing to offer, it will simply skip it.
  • NodeBB v3 Chat Very Slow

    Moved Performance nodebb v3 nodebb chat
    47
    11 Votes
    47 Posts
    21k Views
    phenomlab
    @DownPW Seems fine.
  • NodeBB v3.0.0-rc.1

    Performance nodebb v3.0.0
    1
    1 Votes
    1 Posts
    701 Views
    No one has replied
  • Optimum config for NodeBB under NGINX

    Performance nginx performance
    4
    3 Votes
    4 Posts
    2k Views
    phenomlab
    @crazycells hi - no security reason, or anything specific in this case. However, the nginx.conf I posted was from my Dev environment which uses this port as a way of not interfering with production. And yes, I use clustering on this site with three instances.
  • 0 Votes
    5 Posts
    2k Views
    phenomlab
    @qwinter this particular site uses the code I wrote if you want to see it in action. It’s a information and intelligence gatherer I designed for collecting various information security articles from around the globe and consolidating them in one place. Essentially, each “post” is in fact generated by the script, and the NodeBB API. https://hostrisk.com/
  • NodeBB 1.19.3

    Solved Performance nodebb 1.19.3 performance
    33
    4 Votes
    33 Posts
    16k Views
    DownPW
    @phenomlab I find the problem Mark The error message indicated this path : http://localhost:4567/assets/plugins/nodebb-plugin-emoji/emoji/styles.css?v=6983dobg16u I change the path url on config.json [image: 47bacc80-f141-41e4-a261-3f8d650cc6f6-image.png] And all it’s good Weird, I didn’t have to change that path before 1.19.3 But this does not prevent the problem from a clean install with Emoji Plugin EDIT: After test, that resolv the problem installation for 1.18.x but not for 1.19.x (I have other error message when I run ./nodebb Setup For resume: NodeJS 16_x with 1.18.x is ok