AIS Tracing
Correlation-id tracer for Azure Cosmos DB
Follow a single value — a correlation id, an order reference — through every container of an Azure Cosmos DB account, and see where it got to. Samples your containers, works out which field links documents together, then draws each step on a timeline. Empty rows matter as much as full ones: they show where the data hasn't arrived.
Free for personal, educational and non-profit use. Commercial use — including a solo consultant on client work — requires a licence. Every build is produced by release CI and served from mayorana.ch.
The macOS build is signed with an Apple Developer ID and notarized, so it opens with a normal double-click.
This app is in beta. The core workflow is stable; some panels are still being refined.
Overview
Something did not arrive. You have a correlation id, an order reference, or a customer number, and a Cosmos DB account with more containers than anyone remembers the shape of. The question is simple — how far did this value get? — and answering it usually means writing a query per container.
AIS Tracing answers it in one step. It samples your containers, works out on its own which field links documents to each other, then follows your value across all of them and draws each step on a timeline.
The gaps are the point. A step with no document is not an empty result to scroll past — it is the answer, showing you exactly which hop the data never reached.
What it does
Automatic link-field detection
Samples the containers and infers which field ties documents of one flow together, so you do not have to describe your schema to it first.
Account-wide search
One value, every container in the account, in a single pass — rather than a query per container written by hand.
Timeline view
Each step the value touched, in the order it happened, so latency between hops is visible rather than inferred from timestamps you compare manually.
Gaps as first-class results
Containers where the value never appeared are drawn as empty lanes. Where the trail stops is usually the whole answer.
Requirements
- macOS on Apple Silicon, Windows, or Linux x86_64
- Azure CLI signed in (az login) — the app has no login of its own and uses your existing session
- An Azure Cosmos DB account you can read, including the separate Cosmos data-plane grant
- Containers that carry some correlating field across documents
Built with
Rust · Dioxus · Azure Cosmos DB
Licence
Source-available under the PolyForm Noncommercial License 1.0.0: free for personal use, learning, research and hobby projects, and for charities, schools, universities and government institutions. Companies using it for work buy a licence.
Source
The source is public and readable. Fork it and rebrand it — the name and logo are trademarks and are not covered by the licence.
View on GitHubQuestions
Do I have to tell it my schema?
No. It samples the containers and infers the linking field from the data, proposes a correlation key, and you paste a value to follow.
A scan failed with 403 — what does that mean?
Cosmos DB data access is a separate grant from normal Azure RBAC, so an account that can see the database still cannot read its contents. The app offers a button to give yourself the Built-in Data Reader role.
Does it write to Cosmos DB?
No. It is read-only against your account.
What if my data is in Log Analytics rather than Cosmos DB?
AIS Analytics is the same tracer pointed at a Log Analytics workspace.
Will it work on a large account?
It samples rather than scanning everything, so container count matters more than total document count. Very wide accounts take longer on the first pass.
Need this adapted to your stack?
These tools came out of real Azure integration work. If yours needs a variant — a different data source, your own conventions, an on-prem build — that is the kind of work we take on.
Get in touch