Different engineering teams like backend and SREs operate on different sides of the system, requiring collaboration during incidents.
The initial step in incident response is to establish a common understanding and transparency across teams before applying quick fixes.
After resolving an incident, teams must collaborate on prevention by documenting best practices for patterns like service retries.
The discovery phase relies on making systems observable by collecting telemetry data like logs, metrics, and traces.
OpenTelemetry provides a vendor-neutral standard for instrumenting applications, preventing vendor lock-in for observability backends.
The open-source Autometrics library uses decorators to automatically generate key metrics like latency, errors, and request rate from functions.
A live demo shows how Autometrics provides live metrics in the IDE and helps define SLOs that can be visualized in Grafana.
A recap of the three key phases for collaborative incident handling: resolving, preventing, and discovering issues together.
The speaker answers audience questions about managing tool complexity and the role of community contributions in open-source projects.
Hubert Burda Media
München, Germany
ROSEN Technology and Research Center GmbH
Lingen (Ems), Germany
29:55Malin Litwinski
21:39Johannes Hansen
59:00Ixchel Ruiz
27:27Christian Seifert
26:18Arto Liukkonen
37:33Tobias Dunn-Krahn
27:39Jemiah Sius
44:31Martin Beránek
Jobs that call for the skills explored in this talk.
Dennemeyer Group
Hesperange, Luxembourg
Smartclip
Berlin, Germany
Digistrat consulting
Paris, France
Synsel Techniek
Rotterdam, Netherlands
Nebius
Amsterdam, Netherlands
Infomaniak | The Ethical Cloud
Geneva, Switzerland
leadtech
