Requests Time out on App platform

Posted November 7, 2020 1.4k views
LaravelDigitalOcean App Platform

We have Laravel application running on the app platform with a VueJs front-end that gets it’s data through Axios.
Frequently the api calls to the back-end take a very long time or time out.

The server log shows various errors and warnings when it occurs:

WARNING: [pool www] child 293, script '/workspace/public/index.php' (request: "GET /index.php") executing too slow (3.452512 sec), logging

ERROR: failed to open /proc/293/mem: No such file or directory (2)

WARNING: [pool www] child 293, script '/workspace/public/index.php' (request: "GET /index.php") execution timed out (30.478413 sec), terminating

WARNING: [pool www] child 293 exited on signal 15 (SIGTERM) after 532.548333 seconds from start

[proxy_fcgi:error] [pid 174:tid 139809301464832] [client] AH01067: Failed to read FastCGI header

[proxy_fcgi:error] [pid 174:tid 139809301464832] (104)Connection reset by peer: [client] AH01075: Error dispatching request to : ,

We tried to change our database from Postgress to Mysql to see if it changed anything, but that’s was not the case. Before we ran the application at AWS and these problems weren’t there.
The Insights do not show any high load on the components.

Would like to know what we can do to find the cause of these problems.

These answers are provided by our Community. If you find them useful, show some love by clicking the heart. If you run into issues leave a comment, or add your own answer to help others.

Submit an Answer
3 answers

I have the same problem. Did you ever find a solution? Thanks, Jon

  • No, unfortunately it is still not resolved.

    • Too bad. I have the same problem and DO support is trying to say it is something about the code in my app but it is literally a blank brand new Symfony application with an empty controller action and a vanilla twig template. When I deploy the exact same thing to Heroku I do not have the problem.

Any update on this? We’ve encountered same problem :/

Same Problem Here. We are facing slow requests, some responding in 500ms, 2ms others +30s, we did several tests with NGINX, APACHE, Regionsm CloudFlare ON/OFF and Redis, at all stages the slowness persisted and the problem “[pool www] child xxx exited on signal 15 (SIGTERM)” occurs. We did a test with the Laravel Telescope and it reports as if the application responded in 200ms…

So far we can’t understand the reason for the problem, has anyone managed to find a solution? We are reversing and disabling various parts of the code and yet the problem remains.