If using Handlebars in a custom vhosts config is supposed to work, consider this issue a bug. If it's not, consider this issue a feature request.
Some of my custom vhosts configs use Handlebars syntax, as supported by https://github.com/bitnami/render-template and seen in https://github.com/lando/nginx/blob/main/config/nginx.conf.tpl. These configs worked before upgrading to lando 3.21, but not afterwards. It looks like the sed is processing the vhosts config instead of render-template, even though render-template is available. Relevant lines from container logs:
2024-06-06 11:43:54 landonginx 15:43:54.78 INFO ==> copying /tmp/server.conf.lando to /opt/bitnami/nginx/conf/nginx.conf
2024-06-06 11:43:54 landonginx 15:43:54.78 INFO ==> copying /tmp/fastcgi_params.lando to /opt/bitnami/nginx/conf/fastcgi_params
2024-06-06 11:43:54 landonginx 15:43:54.79 INFO ==> Using sed to place correct webroot in vhost template
2024-06-06 11:43:54 landonginx 15:43:54.79 INFO ==> Rendered template /tmp/vhosts.lando to /opt/bitnami/nginx/conf/vhosts/lando.conf
2024-06-06 11:43:54 landonginx 15:43:54.80 DEBUG ==> # WordPress single site rules. # Designed to be included in any server {} block. # LANDOWORDPRESSNGINXCONF # Upstream to abstract backend connection(s) for php upstream php { server fpm:9000; } server { listen 80 default_server; listen 443 ssl; server_name localhost; ssl_certificate /certs/cert.crt; ssl_certificate_key /certs/cert.key; ssl_verify_client off; ssl_session_cache shared:SSL:1m; ssl_session_timeout 5m; ssl_ciphers HIGH:!aNULL:!MD5; ssl_prefer_server_ciphers on; port_in_redirect off; client_max_body_size 100M; ## Your only path reference. root "/app/."; ## This should be in your http block and if it is, it's not needed here. index index.php; {{#if MEDIA_REDIRECT}} # Media redirect location ^~ /wp-content/uploads { expires max; log_not_found off; try_files $uri @prod_site; } location @prod_site { rewrite ^/(.*)$ "https://{{MEDIA_REDIRECT}}/$1" redirect; } {{/if}} location = /favicon.ico { log_not_found off; access_log off; } location = /robots.txt { allow all; log_not_found off; access_log off; } location / { # This is cool because no php is touched for static content. # include the "?$args" part so non-default permalinks doesn't break when using query string try_files $uri $uri/ /index.php?$args; } location ~ \.php$ { #NOTE: You should have "cgi.fix_pathinfo = 0;" in php.ini fastcgi_param SCRIPT_FILENAME $document_root$fastcgi_script_name; fastcgi_param PATH_TRANSLATED $document_root$fastcgi_script_name; include fastcgi_params; fastcgi_intercept_errors on; fastcgi_pass php; fastcgi_buffers 16 16k; fastcgi_buffer_size 32k; } location ~* \.(js|css|png|jpg|jpeg|gif|ico)$ { expires max; log_not_found off; } }
All is well if start the app without custom config, run render-template "/path/to/custom-vhosts.conf.tpl" > /opt/bitnami/nginx/conf/vhosts/lando.conf inside the container, and restart nginx.
If using Handlebars in a custom vhosts config is supposed to work, consider this issue a bug. If it's not, consider this issue a feature request.
Some of my custom vhosts configs use Handlebars syntax, as supported by https://github.com/bitnami/render-template and seen in https://github.com/lando/nginx/blob/main/config/nginx.conf.tpl. These configs worked before upgrading to lando 3.21, but not afterwards. It looks like the
sedis processing the vhosts config instead ofrender-template, even thoughrender-templateis available. Relevant lines from container logs:All is well if start the app without custom config, run
render-template "/path/to/custom-vhosts.conf.tpl" > /opt/bitnami/nginx/conf/vhosts/lando.confinside the container, and restart nginx.