Solutions
See where a digital banking request crosses services.
A retail bank's app shows errors when customers move money between accounts. The team brings service health, request traces and logs together for the investigation.
The situation
A retail bank's app lets customers move money between their own accounts. One evening, an alert fires on the transfer service: the share of requests returning errors has risen for a ten-minute window, and then fallen again.
In banking, a short burst of errors still matters. The team needs to know which transfers were affected and whether the cause is still there.
What the team sees first
Service health shows the transfer service with errors and the account-balance service healthy, so the problem is not simply that data is missing. The alert's window tells the engineer exactly which minutes to inspect.
How the investigation goes
Traces of failing transfers all show the same step: a call from the transfer service to a ledger service that returns an error span. Traces for transfers outside the window pass through the same call without trouble.
Logs from the ledger service for those ten minutes show a short period in which it rejected requests while a routine job held a lock on a table. When the job finished, the rejections stopped, and the errors ended with them.
The team can say what happened, when it began and ended, and why it is unlikely to recur at the same hour unless the job and the traffic collide again.
What gets shared
The evidence and the response notes stay together: the window, the failing call, the log lines that explain the lock, and the decision to move the job to a quieter hour. When someone asks later what happened that evening, the answer is in one place.
What this shows
A digital banking request touches many services. Putting health, traces and logs in front of the team for the same window turns a brief alert into an explanation and a fix.
These stories are illustrative examples. They describe invented scenarios in an industry, not a real customer, a real incident or any company's systems.
Start with your own services.
A trial workspace is free for 14 days. Send your telemetry and follow your own requests.