Closed stuartcusackie closed 2 years ago
I'm going to close this as it was stable for 10 days many months ago. Maybe just an issue with one of that packages that has been updated. Feel free to comment so we can reopen if this is still an ongoing thing!
My staging server has a recent Statamic / Blade Directives build and it keeps crashing and producing 502 bad gateways. The server is still running with plenty of available resources but nginx stops responding. A simple nginx restart fixes it. UPDATE: It has been stable for 10 days now.
A few points:
Statamic\Tags\Glide::generate()
Things I've tried:
PHP7.4-fpm.log Some requests are taking really long. This one took 36 seconds but there are others that are taking 100s+. Maybe the server was overloaded at this point. I enabled PHP slow logs on requests that take over 30 seconds to see if I could get any more detail (below).
PHP slow.log Below is matching slow log entry for the above (same process id). A problem with Guzzle's CurlHandler??
laravel.log I can see lots of error blocks like below. I have no idea why production.ERROR appears so many times. I can't be 100% sure that these errors are directly connected to the timeouts but it's a strange coicidence at least.
I've caught these exceptions in Laravel's Exception hander and logged the request details. There was nothing unsual in there, just bot requests for pages and files that don't exist.
Statamic version: 3.1.14 PHP version: 7.4 Existing Laravel app
Will update if I learn any more. Cheers.