Episode description
Released on the same day as Kubernetes, cadvisor is a container monitoring daemon that collects metrics and serves them to monitoring tools. It’s built into the Kubelet, and underpins many components in Kubernetes, such as eviction and autoscaling. David Ashpole of Google Cloud is TL of Kubernetes SIG Instrumentation, and the maintainer of cadvisor; he joins Adam and Craig this week to explain where instrumentation fits in the stack, and what you should do as a Kubernetes maintainer vs. a cluster administrator.
Do you have something cool to share? Some questions? Let us know:
- web: kubernetespodcast.com
- mail: kubernetespodcast@google.com
- twitter: @kubernetespod
- In Craig’s neighbourhood:
- Masks
- National Ice Cream Day
- GKE Ingress features:
- Exposing services on GKE
- OpenShift 4.5
- Spring Cloud Data Flow for Kubernetes from VMware; part of the Spring Runtime package
- k8spin.cloud is closing and making their code open source
- Custom Pod Autoscaler (and docs) by Jamie Thompson
- Envoy 1.15 round-up from Tetrate; release notes from the team
- Fluent Bit 1.5 summary at the CNCF
- k3d v3.0 and new web site
- Best practices for creating a highly available GKE cluster
- Recommended alerts for AKS
- Ingress support added to AWS App Mesh
- Platform9 adds new apps to their Managed Kubernetes Service
- CVE-2020-8557: Node disk DOS by writing to container /etc/hosts
- CVE-2020-8559: Privilege escalation from compromised node to cluster
- Threat Alert: Attacker Building Malicious Images Directly on Your Host from Aqua Security
- Certified Kubernetes Security Specialist (CKS) coming in November
- Sign up for a free pass to Virtual KubeCon EU keynotes
- Diving Into Istio 1.6 Certificate Rotation by Christian Posta
- SIG Instrumentation
- inodes
- cgroups
- cadvisor
- Monitoring metrics with Prometheus
- Victor Marmol and Vish Kannan
- Episode 22, with Dawn Chen
- CRI
- Resource metrics pipeline
- Heapster
- Metrics Server
- kube-state-metrics
- Managing Your Costs on Kubernetes by Karl Stoney from Autotrader
- Episode 52, with Russell Warman and Karl Stoney
- Metrics Stability Framework
- Structured logging
- Distributed tracing in Kubernetes
- Node out of memory eviction
- Pod priority
- David Ashpole on Twitter