No cpu metrics from running pods on stackdriver
7/13/2016
Hi im trying to setup stackdriver to monitor my containers but the cpu metrics dont seem to work, im working with the following versions
Master Version 1.2.5
Node Version 1.2.4
heapster-v1.0.2-594732231-sil32this is a group a create for the databases (it also happens for the wildfly pod and modcluster), i have a couple of other questions,
- is it posible to monitor postgres or i have to install the agent on the docker image
- can i monitor the images on kubernetes, or the disks on Google cloud?
-- PaulMB
google-kubernetes-engine
kubernetes
stackdriver
Similar Questions
1 Answer
7/13/2016
Do your containers have CPU limits specified on them? The CPU Usage graph on that page is supposed to show utilization, which is defined as cores used / cores reserved. If a container hasn't specified a maximum number of cores, then it won't have a utilization either, as mentioned in the description of the CPU utilization metric.
-- Alex Robinson
Source: StackOverflow