Search results

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+...

Enabling the 'JS Combine' options somehow made this issue go away. A bit puzzling still.

As seen in the waterfall below, the 561f5.js?caea7 files takes 833ms to transmit whereas (jpeg) files twice the size take less than 100ms. Wondering why this one file takes so much longer.

I am wondering how OLS manages the 'Connection: Keep-Alive' header. I had added it to my custom headers and it was causing issues with HTTP/2, so I removed it. But I still see it reappear by itself: Is that something OLS could insert by itself based on settings?

I think I fixed it. The /usr/local/lsws/admin/tmp/ directory was missing read permissions for group/other. Once I fixed that, $_SERVER['LSWS_ADMIN_SOCK'] got its proper value. I'm not sure how the permission problem happened, could have been my doing. Does OLS have a "fix permissions" tool?

Most likely unrelated to my problem, but a bit surprised to see a case-sensitive string comparison for 'uds://' when the documentation says to use 'UDS': public static function getCommandSocket($cmd) { $ADMSOCK = $_SERVER['LSWS_ADMIN_SOCK']; if (...

If I print $_SERVER['LSWS_ADMIN_SOCK'] in the related code, it comes as empty.

Not sure if it's related, but I sometime have two openlitespeed sets of processes:

Maybe this is due to this error I see in the server error.log: I have not figured out what's causing this.

Well somehow that does not work for me. I just made a change to the headers, did the graceful restart, run killall lsphp, reloaded the non-cached checkout page I was seeing the header issue on, and the issue was still there. I tried killall lsphp another time, reload, and still the same. Only...

Doing a graceful restart does not kill/restart old openlitespeed/lsphp processes as far as I can see. This is what I am using:

I submitted a bug report at bug@litespeedtech.com with a log showing that it was sometime picking the private cache and other times the public cache. Waiting to hear if they have an idea what it could be...

I would appreciate help to figure out why HTTP/3 works if I restart the server but then disappears after some time (most likely when new lsphp processes get created).

Thanks for the advice, that makes sense.

That was it. That's what I get for doing too much copy/paste. Thank you for the quick reply. Very much appreciated. I see the header is prohibited in HTTP/2. Is it useful for HTTP/1.1? And if so, is there a way to define protocol-specific headers in OLS?

Oh thanks, will try that.

I am getting the following error on two of my websites after adding the following custom headers: Chrome and Firefox do not seem to complain on desktop, but Safari returns an error on desktop/mobile and Chrome returns an error on mobile. I am open to the idea that I wrote something wrong in...

And now it's gone again.

And now it's back. This is puzzling. I should retract my statement about not changing anything on the server config. I deleted a directory that another vhost was pointing to (because I didn't need it). Is it possible that created an error that killed HTTP/3? Seems unlikely but I guess I don't...

Well I have not changed anything on the server config and I now see HTTP/3 is gone if I test on Geekflare or HTTP/3 Check.