Hi HN,<p>In my recent jobs I've been working mostly on graphQL projects, and despite the technology being more mature now, I wasn't happy with the monitoring in place. So I decided to create my own platform, with a focus on simplicity (as little config as possible to get things in place) and target only GraphQL (generic monitoring platforms that try to also support GraphQL tend to be incomplete or hard to use due to difference between GraphQL and other API paradigms).<p>At the moment the platform allows to see the latest requests, see operations based on usage, error rate and latency, show the errors, traces (with a breakdown by resolvers) and create alerts based on metrics (with notifications by email, slack or telegram).<p>How it works:<p>- Create an account, copy the API key
- Add a plugin to your graphql server (currently only Apollo Server is supported, but the goal is to support as many languages/libs as possible, please let me know in the comments if you're interested but are using another framework!)
- The data are ingested by the platform (internally the stack uses a mix of Clickhouse, Postgres and Redis)
- You can immediately observe the most common operations, traces, errors and create alerts