Make the production deploy foolproof against the shared nginx container's read-only bind mounts on the Digital Ocean droplet. The previous maintenance flow tried to docker-cp/docker-exec into /var/www/html and /etc/nginx/conf.d, both of which are mounted :ro on prod, and /var/www/html happens to point at the WordPress html dir — so writes either failed silently or risked scribbling into another site's tree. Maintenance assets and the engagement flag are now written directly to host paths (/docker/nginx/maintenance and /docker/nginx/conf.d/maintenance.flag) that nginx already sees through its existing bind mounts, so the script no longer depends on a writable container layer, survives container rebuilds, and works regardless of read_only settings. A pre-flight check verifies the maintenance bind mount is actually present on the nginx container and fails fast with a clear "run the one-time setup" message if it isn't, instead of silently serving stale content. The nginx config now serves maintenance.html and /m/ from a dedicated /var/www/maintenance root rather than sharing the WordPress html dir.
On the front end, hero images on Pack Walks, 1:1 Walks and Puppy Visits were rendering at whatever aspect ratio their source files happened to have, so one page felt tall, another wide, another oversized. They are now locked to a 4:3 frame with object-fit: cover, matching the About Us section images, which were given the same treatment. The About Us body grid was also alternating between 0.7fr/1.3fr and 1.3fr/0.7fr columns depending on whether a section was reversed, which made the copy width jump between sections; both layouts are now an even 50/50 split, with the existing order swap still handling the image-left vs image-right alternation.
The reveal-on-scroll action used to require 18% of an element to intersect before fading it in, with an additional -8% bottom margin, which meant the section directly below a service-page hero stayed invisible on initial load until the user scrolled — making the page look blank below the hero on navigation. The action now does a synchronous bounding-rect check on mount and reveals anything already in the viewport immediately, falling back to the IntersectionObserver for everything below the fold.
The "Explore our services" block on About Us was a bespoke icon-tile grid that did not match the homepage's "What we do" cards; it now reuses the shared ServicesSection component (with the heading exposed as a prop), so both pages produce identical card layout, descriptions, "from $" prices, and Learn more CTAs. The footer Explore column was missing the About Us link — added between Our Pricing and Contact Us so it propagates through the homepage content sync into PostgreSQL on the next deploy.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
Replaces the earlier auto-fallback-on-upstream-error approach with an
explicit flag-file toggle controlled by the deploy script. The flag
is touched before stopping the app and removed on successful finish
(or via trap if the deploy aborts), so a failed deploy doesn't strand
the site in maintenance.
- nginx/goodwalk.co.nz.svelte.conf.example: error_page 503 routes to
/maintenance.html (internal); /m/ serves static maintenance assets;
the / and /api/submit blocks return 503 when /etc/nginx/conf.d/
maintenance.flag exists.
- nginx/maintenance.html: brand-styled "Be right back" page — full
Goodwalk green background, white card with yellow accent, real
Goodwalk logo, contact details fallback, auto-reload after 60s.
- nginx/logo.png: maintenance-time logo (served from /m/logo.png).
- nginx/nginx.conf: reverted the earlier auto-fallback edits; this
file is not deployed (the prod conf is goodwalk.co.nz.svelte.conf
.example).
- scripts/deploy-remote.sh: copies maintenance.html + logo into the
nginx container, reloads nginx so the new conf is live, touches
the flag, then runs the rebuild, then clears the flag. Adds a
trap-based clear_maintenance_flag fallback. Also adds a defensive
env-file merger that appends new keys from deploy.env.template
without clobbering live values, with a timestamped .env backup.
Plus a small a11y polish unrelated to maintenance:
- ServicesSection: "Learn more" links now include screen-reader-only
"about <Service>" context.
- base.css: adds .visually-hidden utility class.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
When the SvelteKit upstream is unreachable (container restart, deploy)
nginx now serves a static, brand-styled "Be right back" page instead
of the default 502/503. Auto-reloads after 60s so visitors don't sit
on it once the app is back.
- nginx/maintenance.html: self-contained, no external assets, inline
paw SVG, brand colours, contact details fallback
- nginx/nginx.conf: proxy_intercept_errors + error_page 502/503/504
on both location blocks; 2s proxy_connect_timeout so nginx fails
over fast instead of holding the connection for 60s
Deploy note: the html file needs to live at /var/www/html/maintenance.html
inside the nginx container (already mounted from /docker/wordpress/goodwalk.co.nz/html).
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
admin
ponzischeme89