controllermanager

README

Controller Manager Stats

Version history

• June 2019, v1.14.3

Resources

Each controller emits a set of metrics, there is no source file to reference but a set of source files that are gathered into a single metrics resource.

Metrics insight

http_request_duration_microseconds The HTTP request latencies in microseconds. Summary - handler

http_requests_total Total number of HTTP requests made. Counter - code - handler - method

http_request_size_bytes The HTTP request sizes in bytes. Summary - handler

http_response_size_bytes The HTTP response sizes in bytes. Summary - handler

rest_client_requests_total Number of HTTP requests, partitioned by status code, method, and host. Counter - code - host - method

workqueue_longest_running_processor_seconds. Gauge - name:

workqueue_unfinished_work_seconds: How many seconds of work has done that is in progress and hasn't been observed by work_duration. Large values indicate stuck threads. One can deduce the number of stuck threads by observing the rate at which this increases. Gauge - name:

process_cpu_seconds_total: Total user and system CPU time spent in seconds. process_open_fds Number of open file descriptors. process_resident_memory_bytes Resident memory size in bytes. process_start_time_seconds Start time of the process since unix epoch in seconds. process_virtual_memory_bytes Virtual memory size in bytes

node_collector_evictions_number Number of Node evictions that happened since current instance of NodeController started. - zone

node_collector_unhealthy_nodes_in_zone Gauge measuring number of not Ready Nodes per zones. - zone

node_collector_zone_health measuring percentage of healthy nodes per zone. - zone

node_collector_zone_size: measuring number of registered Nodes per zones. - zone

leader_election_master_status

Setup environment for manual tests

WIP: controller manager will usually run at every master node, but that might not be the case. It could be executed as a host process or an in-cluster pod.

• If host process (for example, systemd), metricbeat should be running at that same node gathering data from the controller.
• If executing as a pod:
  • A metricbeat instance can be also executed using the same affinity and deployment object (deployment, daemonset, ...) as the controller manager.