Search results

That is a pretty generic PHP error message. Looks like it timed out.

Yeah. it seems to me like you'd be better off disabling QC. Just put: query_cache_size=0 in your my.cnf file that you found with Peter's help. Some more details: How to Fully Disable Query Cache in MySQL - Percona Database Performance Blog

You'll want to define custom directives in the "Additional NGINX directive" section in Plesk for the domain

We're on CentOS 7, no errors on 18.0.38 Update #3, besides some UX/UI changes that we had to update. No dropbox extension.

Seems like a bug. @Rasp wanna get a bug report in?

Yep. Saw this yesterday. Looks like a bug.

PHPMailer can use sendmail, but also an external SMTP relay so you can connect to localhost. We do this for our WP installations so email works. What you mentioned is an option, but carries the downsides you mentioned, which are obviously less than ideal. You can manually pass/rewrite the...

So, there isn't really a simple way for a chroot to be truly secure AND to have a sendmail binary. If you are up for it, there's something like this: Lisa Rushworth Home Page Otherwise, some implementation use a sort of proxy to mediate communication. The best option is just have the PHP...

No. What does the actual per thread breakdown look like? Is this a VPS, where the host may be throttling? Do you have a proc limit/PHP worker bottleneck?

Project is also mostly "dead" - based on their forums: Is this project dead? - Free Modsecurity rules - Comodo Web Application Firewall

Yes, can confirm this has been our experience too. Not the caching issue specifically, but Comodo is just not worth any of the benefits (selective rule classes, pricing) it brings. Atomicorps Advanced has served us well.

With all due respect that's not really a Plesk issue. If you don't want to/don't know how to transfer the configurations over to Nginx configs, but also want to use Nginx as a proxy, you'll have to make a compromise and choose one, I don't see a purpose in allowing the proposed scheme...

What problems with NGINX to Apache do you have exactly?

You should be looking at what is causing the difference. Correlation != causation. If it's text compression, turn on Brotli/GZIP for NGINX. NGINX does not "slows down the Apache strongly towards the outside." You cannot turn of NGINX for just one domain. NGINX is bound to 80/443 on your IPs...

Run a plesk repair and see if it's a permissions issue.

Open a support ticket with Plesk if you need urgent support. In the future, you obviously shouldn't randomly disable Apache modules that you aren't sure of. That's basically begging for trouble. First, I'd see if apache is working. Can you connect directly to the Apache port? (Usually 7081)...

NGINX still lacks HTTP/3: Our Roadmap for QUIC and HTTP/3 Support in NGINX - NGINX Aoache, well, doesn't care, We probably won't be seeing stable HTTP/3 support in Plesk any time soon. If you require QUIC/HTTP3, consider using Cloudflare to proxy your content.

100% agree.

Testing assets, ie via wpt vs running a concurrency benchmark are totally different things and both have their validity. That shouldn't be the case. FPM is FastCGI pooled, which has the benefit of not needing to spin up as many threads and persistent cache. How are they configured? Ie, if...

What does the resource usage breakdown look like?