𝟯 𝗣𝗼𝘀𝘁-𝗗𝗲𝗽𝗹𝗼𝘆 𝗖𝗵𝗲𝗰𝗸𝘀 𝗜 𝗥𝘂𝗻 𝗔𝗳𝘁𝗲𝗿 𝗘𝘃𝗲𝗿𝘆 𝗖𝗹𝗼𝘂𝗱𝗳𝗹𝗮𝗿𝗲 𝗣𝗮𝗴𝗲𝘀 𝗕𝘂𝗶𝗹𝗱
I spent two weeks fixing bugs that only appeared in production.
One _redirects rule blocked my sitemap. A race condition between image uploads and Cloudflare deployment caused another issue.
Now, I run three specific checks after every deploy. These are not full tests. They solve the exact problems I face with my Astro 5 SSG sites.
Check 1: Sitemap Availability
I verify sitemap-index.xml returns a 200 status on all domains.
I also check sitemap-0.xml. This file contains the actual URLs. I check that the URL count stays above a certain number. For one site, if the count drops below 1,000, I know my data pipeline failed.
I use curl to check this. I do not follow redirects. This helps me catch broken redirect rules that look fine in a browser but block crawlers.
Check 2: IndexNow Submission
After the sitemap check, I run a script to submit URLs to IndexNow. This notifies Bing, Yandex, Naver, and Seznam about new content.
If IndexNow returns a 403 error, it means my key verification file failed to deploy. Catching this immediately prevents delays in search engine indexing.
I run this manually after a deploy. I do this to ensure I only submit URLs that are live and stable.
Check 3: Lighthouse Trends
I run a Lighthouse check on a schedule, not after every deploy.
I monitor three metrics:
- Performance (I look for scores below 80)
- CLS (I look for scores above 0.1)
- Accessibility scores
Since my sites use static HTML and CSS, these scores should stay steady. If they drop, a change in my Tailwind config or a component likely broke the layout.
I do not use these scores to block deploys. I use them to monitor trends.
Why these three?
I do not use uptime monitors or end-to-end user tests. My sites are static deployments on a CDN. The database is only queried at build time.
These three checks cover the only real risks I face with this setup.
Source: https://dev.to/morinaga/three-post-deploy-checks-i-run-after-every-cloudflare-pages-build-70b