On two random days last week, the daily CDN usage for our client's site jumped up from approx 2gb per day to 20gb per day, and fell back down again, pushing the near repository near the limit for the month. This is not typical behaviour, and we have worked hard to optimise the deployment processes to limit the number of API calls to prevent the other limit being reached.
Please can you advise why the usage on purcell.prismic.io was so high on those two days? There are no videos hosted within Prismic, and the Google Analytics do not show increased traffic levels on those two days.
Hey team. Here are some things to take into account when seeing API usage spiking:
Too Many API Hits:
Client-Side Overload: If your site's making a ton of requests from the client side, it could be driving up those API queries. Consider shifting things to the server side or using static site generation to cut down on the requests.
Content Complexity: Are your pages packed with lots of nested content? That could mean lots of requests for just one page. Simplify things and maybe use GraphQuery to fetch nested content more efficiently.
CDN Bandwidth Issues:
Larges Files: Check if you're serving up huge videos straight from Prismic. That's a bandwidth guzzler. Optimize your images and use Prismic's image tools for smoother delivery.
API Impact: Is your CDN usage getting hit hard by Content API requests? Maybe you're fetching more than you need. Streamline your queries and think about using GraphQuery to be more selective.
Try these out, they can help you get spikes like these under control
We have made recently significant changes to limit API usage by only building the site twice a day.
I am looking for the specific reasons for the extremely high usage on those two days so we can mitigate those events. Please can you outline exactly what caused the usage to increase ten-fold on April 30th and May 1st?
