Product changelog
What changed in Owlpane, newest first. Dates are when a change was merged; it can reach the hosted service a few days later.
- Collectors: each collector shows the configuration it is actually running, with secrets redacted.
- Services with the same name in different namespaces (
service.namespace) now stay separate. - Kubernetes: network flow edges show packet counts.
- New Collectors page for OpenTelemetry collectors that report over OpAMP, with remote configuration and staged rollouts to a canary group first. Collectors silent for 30 days are removed from the list.
- Add hosts and VMs, Windows included, with one install command from Setup.
- Traces and logs show the OpenTelemetry scope and resource, structured log bodies and the log event name.
- Metrics explorer: exponential histograms and summaries.
- Profiler: a flame graph of stored profile stacks, narrowed to one pod if you want, and a span’s own CPU profile when samples are linked to it.
- Kubernetes network: capture filters, how each probe captures (tier and kernel), and round-trip time and retransmits on flow edges.
- New starter dashboard for runtime security findings from Falco and Tetragon, when the agent’s security bridge is on.
- Kubernetes cluster data is visible only to members granted access to that cluster.
- Fixed: a profile without a service name no longer shows up under every service.
- Alerts: a PromQL anomaly rule type, and templates for postmortems.
- Workflows can run steps when an alert fires or resolves or an incident opens, including a step that proposes muting.
- Postmortem editor with ruled-out causes, action items you can tick off, and a shift handoff card.
- PostgreSQL: table health, unused indexes, deadlocks and slow query plans.
- Uptime: manage private locations and run a check from one.
- AI SRE: see what it remembers, a risk inbox, an accuracy report, two-approver actions and shadow mode.
- Amazon CloudWatch metrics can be sent through CloudWatch Metric Streams and Amazon Data Firehose.