Search results

I have come to the conclusion that I am not doing things quite right to avoid database / disk errors, when either restarting OLS, restarting lsphp processes, installing server updates... I think one of the things that trip me up are WP cron jobs that keep running from crontab files every so many...

Looking at my server logs, I have not experienced the "No request delivery notification" message since 11/14. I unfortunately do not know what I changed at that point that may have made a difference. My shell history does not go that far back. I don't see any server restart between the 12th and...

Is the nginx server running on the same machine or a different machine? I see the proxy address is defined as: address 172.16.0.188:80 but it looks like your listener is also on port 80. I would also advise again calling the proxy extapp and the virtual host both 'nginx'. That's sure to create...

You may want to look at 'Use Client IP in Header' on the Server Configuration > General tab. This may help: https://help.cachewall.com/article/71-real-client-ips-litespeed

It came back after I stopped lsws, killed all lsphp processes, and restarted lsws. Will probably go away again at some point. At least that has been my experience. Works after restart only for a while.

Well, what do you know, today http/3 is gone again. One thing I did was enabling high debug server log and then turning it off. Another one was adding this to the headers: and this to the .htaccess file: # Pick webp images when available <IfModule mod_rewrite.c> RewriteEngine on RewriteCond...

Update: Running lsup.sh -e 1.7.7 today to get the updated development release (for a separate caching issue) seems to have re-enabled http/3 & QUIC.

Thank you, this helps. How does this parameter interact with 'Connection Keep-Alive Timeout'? What's the default setting for 'Connection Keep-Alive Timeout' and what's a good setting for it? I think Cyberpanel sets it to 1 second. Is this setting moot in http/2 and http/3?

This just started happening again soon after I disabled Cloudflare proxying to try to see if HTTP/3 & QUIC were working with OLS 1.7.7 (which they are not). David (who replied to my bug report email thread) says it's a usage issue, not a server bug. I don't really understand how this could a...

Okay, will do. PS: If I was still using CF, I think the 'server' header would say Cloudflare instead of LiteSpeed.

I disabled the plugin for now and cleared the cache. No change as far as I can see but the cf-edge-cache header is now gone.

Is that a problem? This is added by the Cloudflare Wordpress plugin: // add header unconditionally so we can detect plugin is activated if (!is_user_logged_in() ) { header( 'cf-edge-cache: cache,platform=wordpress' ); } else { header( 'cf-edge-cache: no-cache'...

Okay, Cloudflare disabled, here is my url: https://swinginatthesavoy.com/ Just tested with Chrome and https://http3check.net/?host=swinginatthesavoy.com, no luck. My server and vhost QUIC settings are attached.

Nevermind, totally forgot I had re-activated Cloudflare for now because I could not get http/3 to work reliably anyway. So in my case, Cloudflare is now the problem. Apologies for the mislead. Let me disable it and rerun.

My understanding is that the difference is at least partly due to not having lsphp already running for the first one. Is there a way to keep it running? PS: I need a bit more time to generate traces for the other websites that use caching.

Here is a site that does not use any sort of caching. One request shows 1.12 sec of waiting time, the other one 186 ms.

When I run my sites through GTMetrix, I sometimes see a short server TTFB and other times it's much longer, several seconds. I do not understand why this is the case. It's as if there was some really long server startup time. This usually happens when no accesses have been done for a bit. Any...

@edan, it might be more efficient for you to share your /usr/local/lsws/conf/httpd_config.conf and /usr/local/lsws/conf/vhosts/proxy-vhost/vhost.conf files here or with LiteSpeed tech support.

I think the rewrite rule is not correct: The documentation says: which in your case would be 'nginx'. The documentation is a bit confusing because it starts with 'apache' at the top and ends up with 'vhost-proxy1' instead of something like 'webserver-extapp' or 'apache'. I mean why...

I've been experiencing the same issue. Still not working for me with 1.7.7.