From c61093480c8b835d9a970835b48b7200c9afb071 Mon Sep 17 00:00:00 2001 From: Karol Wychowaniec Date: Mon, 5 Mar 2018 15:37:41 +0100 Subject: [PATCH] Add Troubleshooting sections to Heapster and Metrics Server addons documentation --- cluster/addons/cluster-monitoring/README.md | 13 +++++++++++++ cluster/addons/metrics-server/README.md | 20 ++++++++++++++++++++ 2 files changed, 33 insertions(+) create mode 100644 cluster/addons/metrics-server/README.md diff --git a/cluster/addons/cluster-monitoring/README.md b/cluster/addons/cluster-monitoring/README.md index ea04420469b..6642d9108de 100644 --- a/cluster/addons/cluster-monitoring/README.md +++ b/cluster/addons/cluster-monitoring/README.md @@ -5,4 +5,17 @@ Heapster collects signals from kubelets and the api server, processes them, and More details can be found in [Monitoring user guide](http://kubernetes.io/docs/user-guide/monitoring/). +## Troubleshooting + +Heapster supports up to 30 pods per cluster node. In clusters where there are more running pods, Heapster may be throttled or fail with OOM error. Starting with Kubernetes 1.9.2, Heapster resource requirements may be overwritten manually. [Learn more about Addon Resizer configuration](https://github.com/kubernetes/autoscaler/tree/master/addon-resizer#addon-resizer-configuration) + +### Important notices + +Decreasing resource requirements for cluster addons may cause system instability. The effects may include (but are not limited to): + - Metrics not being exported + - Horizontal Pod Autoscaler not working + - `kubectl top` not working + +Overwritten configuration persists through cluster updates, therefore may cause all effects above after a cluster update. + [![Analytics](https://kubernetes-site.appspot.com/UA-36037335-10/GitHub/cluster/addons/cluster-monitoring/README.md?pixel)]() diff --git a/cluster/addons/metrics-server/README.md b/cluster/addons/metrics-server/README.md new file mode 100644 index 00000000000..f5f063bb7a7 --- /dev/null +++ b/cluster/addons/metrics-server/README.md @@ -0,0 +1,20 @@ +# Metrics Server + +[Metrics Server](https://github.com/kubernetes-incubator/metrics-server) exposes +core Kubernetes metrics via metrics API. + +More details can be found in [Core metrics pipeline documentation](https://kubernetes.io/docs/tasks/debug-application-cluster/core-metrics-pipeline/). + +## Troubleshooting + +Metrics Server supports up to 30 pods per cluster node. In clusters where there are more running pods, Metrics Server may be throttled or fail with OOM error. Starting with Kubernetes 1.9.2, Metrics Server resource requirements may be overwritten manually. [Learn more about Addon Resizer configuration](https://github.com/kubernetes/autoscaler/tree/master/addon-resizer#addon-resizer-configuration) + +### Important notices + +Decreasing resource requirements for cluster addons may cause system instability. The effects may include (but are not limited to): + - Horizontal Pod Autoscaler not working + - `kubectl top` not working (starting with Kubernetes 1.10) + +Overwritten configuration persists through cluster updates, therefore may cause all effects above after a cluster update. + +[![Analytics](https://kubernetes-site.appspot.com/UA-36037335-10/GitHub/cluster/addons/cluster-monitoring/README.md?pixel)]()