Ever since the changes to let’s encrypt earlier this month my nextcloud instance has ground to a halt. To be honest a 56k modem would run faster. The server has always run fine before the authentication changes. Has anyone else had this issue could someone please advise?
It’s probably not any changes to Let’s Encrypt, since the software that runs on your server runs only rarely, and doesn’t consume a lot of resources. Is it possible you made some config changes recently?
If you run top, do you see any processes using a lot of CPU? What are the specific actions that became slower?
Nothing untoward with regards to CPU usage. Loading the login page takes an age. I’m using noIP enhance for dynamic DNS. It’s almost as if my connection is being throttled going to check the ratelimit
If restarting doesn’t help, one thing you can try is enabling OCSP Stapling on your web server, so that your browser does not need to connect to the Let’s Encrypt OCSP server independently. I’ve seen that cause similar symptoms before.
Something like:
# OCSP Stapling, only in httpd 2.3.3 and later
SSLUseStapling on
SSLStaplingResponderTimeout 5
SSLStaplingReturnResponderErrors off
SSLStaplingCache shmcb:/var/run/ocsp(128000)
The other thing is lowering either your computer’s MTU to ~1300 and see if that helps as well.
Well… rebooted the router after checking the QoS settings. Reset them to standard and changed the DNS away from my pi-hole ad-blocker and the pages are loading superfast. So will change the DNS back to pi-hole to see if it slows things down. If it doesn’t slow things down then I can only assume someone (i.e. me!) must have changed the QoS to the point where it wasn’t playing nicely with nextcloud.