You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Nothing wrong in the docs per-se but i went down a rabbit hole where nginx was failing to run, which meant that all my projects were dumping me into the Devilbox intranet login page, after having installed a MacOS system update and restarted my host machine - despite the fact that everything else appeared to be working inside my php/memcd/mysql containers.
Eventually i spotted the following in my docker logs after all the vhost logs:
devilbox-httpd-1 | 2023/05/31 15:23:46 [emerg] 1406#1406: could not build server_names_hash, you should increase server_names_hash_bucket_size: 64
devilbox-httpd-1 | nginx: [emerg] could not build server_names_hash, you should increase server_names_hash_bucket_size: 64
devilbox-httpd-1 | watcherd: [2023-05-31 15:23:46] [ERR] TRIGGER failed: nginx -s stop
This led me to realise that the error wasn't caused by the host update i'd just installed (i suspected something docker network related), rather that i hit some sort of default nginx limit due to the number of projects in my data/www directory, that only surfaced after restarting docker and devilbox since adding a new project.
What is missing in the documentation?
I'd suggest a section on the troubleshooting page that points users to search the docker logs for any issues with their devilbox-httpd container, for example in my case needing to increase nginx's server_names_hash_bucket_size setting.
Are you willing to provide a PR to address this?
Yes
The text was updated successfully, but these errors were encountered:
Hi Vikas, i've got 52 project directories there. The latest additions were blesta_virtualizor/ and provision-provider-software-licenses-enhance/. My resolution was adding a .conf file under cfg/nginx-stable/ which set server_names_hash_bucket_size 128;.
Yep - this issue isn't to try and get support, I was suggesting enhancement to the troubleshooting page of the docs. If you don't think that's within scope for that page, no worries.
What is wrong in the documentation?
Nothing wrong in the docs per-se but i went down a rabbit hole where nginx was failing to run, which meant that all my projects were dumping me into the Devilbox intranet login page, after having installed a MacOS system update and restarted my host machine - despite the fact that everything else appeared to be working inside my php/memcd/mysql containers.
Eventually i spotted the following in my docker logs after all the vhost logs:
This led me to realise that the error wasn't caused by the host update i'd just installed (i suspected something docker network related), rather that i hit some sort of default nginx limit due to the number of projects in my data/www directory, that only surfaced after restarting docker and devilbox since adding a new project.
What is missing in the documentation?
I'd suggest a section on the troubleshooting page that points users to search the docker logs for any issues with their
devilbox-httpd
container, for example in my case needing to increase nginx'sserver_names_hash_bucket_size
setting.Are you willing to provide a PR to address this?
Yes
The text was updated successfully, but these errors were encountered: