What It Takes to Run Monitoring and Observability Webinar

Most content comparing monitoring and observability stays at the definitional level: three pillars, metrics versus traces, known and unknown unknowns. This session skips that and looks at the operational side instead.

Kris Buytaert, Practice Lead Observability at Percona, and Blerim Sheqa, COO at Icinga, talk through what it actually takes to run each approach. Not which one is better, but what each one demands: how long it takes to get the first useful data, what can and cannot be covered, where the configuration lives and who owns it, and how much of a team’s time goes into keeping it working a year later.

Recorded live on October 27, 2026.

In this session, we cover:

• Why observability won’t fix broken monitoring, and why it’s usually the next step in maturity, not a replacement
• Getting data flowing: agent rollout, check definitions and templates on one side, eBPF-based agents that map your environment within minutes on the other
• Deliberate checks versus broad collection: deciding up front what matters, or deferring that decision to query time
• Coverage limits: switches, routers, storage appliances, PDUs and UPS, and applications nobody can instrument anymore
• Ownership: central monitoring teams versus application teams, and why “you build it, you run it” matters more than the tool
• Keeping it running: why the old Graphite and log stacks became too expensive to operate, and how long you actually need to keep the data
• What changes at scale: firewall rules, security reviews, tenants and access control, and the point where automation stops being optional
• Why enterprises keep running both, and where the boundary between them usually sits
• Interactive Q&A: top-down vs. bottom-up, noisy neighbours, data retention, and getting teams interested in their own alerts

This session gives you the operational picture rather than the marketing one.

0:00 Welcome & format

1:47 Introducing Kris Buytaert

3:01 "Observability will not fix your broken monitoring"

5:29 Who asks for observability?

6:35 Monitoring vs. observability from an ops perspective

10:21 Debugging: from Graphite stacks to modern tools

13:21 When to fix the basics first

15:21 Day one with Icinga: modelling the desired state

18:32 Day one with an observability platform

22:49 Collect everything vs. deliberate checks

28:02 Coroot and eBPF

32:34 Who owns the configuration?

37:54 Observability as a marketing buzzword

39:20 Operating at scale: the organizational problems

41:52 What gets easier at scale? Automation

44:17 Hybrid infrastructure, black boxes and SNMP

47:04 Q&A: Top-down vs. bottom-up

48:58 Q&A: Noisy neighbours and shared resources

50:15 Q&A: How long should you keep data?

52:24 Q&A: Getting teams interested in alerts

54:49 Key takeaways

56:41 Wrap-up & next webinar