While updating from version 2.0.* to 2.3.* to take advantage of the valet use, we hit a bug where sites were no longer being served.
After running composer global update to get version 2.3.0 of valet, our sites stopped loading. We got “Site can’t be reached error” in Chrome. Checking brew services list everything looked ok. Eventually we spotted that Nginx was no longer running. After checking /usr/local/var/log/nginx/error.log we could see the following error:
2019/04/19 11:34:33 [emerg] 53582#0: open() “/Users/rnicol/.valet/Log/nginx-error.log” failed (2: No such file or directory)
On closer inspection Valet has moved it’s config and log directory from ~/.valet/ to ~/.config/valet/ So to fix nginx we changed all reference of this directory in /usr/local/etc/nginx/valet/valet.conf and ran valet restart, which solved the problem .
UPDATE: Following the above after restarting the Mac the sites again stopped loading. This was because dnsmasq could not start, as it also had a reference to wrong directory. To resolve this we changed /usr/local/etc/dnsmasq.conf and updated the conf-file line as follows: