Inconsistent server response times (sometimes 5+ seconds)

[gilles]

I am getting very inconsistent server response times. Wondering why that could be.

For instance, I see the "waiting" part in GTMetrix for the main URL vary from 100ms to 5s, and even 25s at times.
Similar variations with the WooCommerce get_refreshed_fragments Ajax POST request from 470ms to 5+ seconds.

I'm running OLS v1.7.5 (with Cyberpanel) on Ubuntu 20.04. I only have 3 WP sites and 2 nodejs apps running with low traffic right now. The server has 2 vCPUs and 4GB of RAM. CPU utilization rarely goes above 60%. Memory is usually around 50%.

Would welcome any advice to ensure I get consistent results.

 

[gilles]

Here are two snapshots of the same requests.

Better case:


Not good case:

 

[David]

For the server speed, if you have any concern, you need to test the static file.
For the static content speed, if you think it is too low, you need to tune the php script or setup using the cache.
You said the POST is not fast, this depends on the post data, can you give me your testing URL?

 

[gilles]

Here is another screenshot showing a slow static asset (WNH-Logo-menu-bar.png.webp):



This does not happen often but I am surprised it happens at all.

Here is the first POST request information:

URL: https://wednesdaynighthop.com/?wc-ajax=get_refreshed_fragments​

Here is the second POST request information:

URL: https://wednesdaynighthop.com/wp-admin/admin-ajax.php​

Data:​

action=vc_get_vc_grid_data&vc_action=vc_get_vc_grid_data&tag=vc_basic_grid&data%5Bvisible_pages%5D=5
&data%5Bpage_id%5D=106&data%5Bstyle%5D=pagination&data%5Baction%5D=vc_get_vc_grid_data&data%5Bshortcode_id
%5D=1602622776101-b0fa640c-cc69-7&data%5Bitems_per_page%5D=3&data%5Bauto_play%5D=false&data%5Bgap%5D
=5&data%5Bspeed%5D=-1000&data%5Bloop%5D=&data%5Banimation_in%5D=none&data%5Banimation_out%5D=&data%5Barrows_design
%5D=vc_arrow-icon-arrow_05_left&data%5Barrows_color%5D=black&data%5Barrows_position%5D=outside&data%5Bpaging_design
%5D=pagination_square&data%5Bpaging_color%5D=sky&data%5Btag%5D=vc_basic_grid&vc_post_id=106&_vcnonce
=b11a759406

For comparison, I was consistently getting around 1 second when hosted at Cloudways.

 

[gilles]

I am looking into my redis object cache and it's possible there is an issue there. I switched another site to LSMCD and I am getting much better ajax request times.

The site above is also showing intermittent Redis read connection error.

 

[gilles]

I resolved part of the Redis error issue. I had two redis servers running with different (conflicting) configs, one running through systemctl and the other one through Docker. But I still saw read error after that.

I bumped up the server from 2 vCPUs/4GB to 4 vCPUs/8GB and things look better. Waiting to see if I see more errors in the log.