Aggregator Diagnostic APIs

Below are APIs exposed by Kubecost Aggregator for troubleshooting without inspecting the PV directly.

/model/debug/orchestrator

Returns current state of the Orchestrator, which governs what state Aggregator is currently in.

/model/diagnostic/tableWindowCount

Used to determine the number of unique WindowStart/WindowEnd pairs exist in the table.

ParameterDefaultDescription

table

Required. The table containing container data to consider. Try container_1d.

/model/diagnostic/coreCount

Used to determine the peak number of monitored cores in the window.

ParameterDefaultDescription

window

2d

Duration of time over which to query and find peak number of cores. Accepts multiple different formats of time (see this Using the window parameter section for more info).

/model/diagnostic/containersPerDay

Used to determine the container rows per daily window, including max and min.

/model/diagnostic/nodesPerDay

Used to determine the node rows per daily window, including max and min.

/model/diagnostic/cloudCostsPerDay

Used to determine the Cloud Cost rows per daily window, including max and min.

/model/diagnostic/containerLabelStats

Used to determine data scale of labels on containers. Returns min, avg, and max label count for all containers.

/model/diagnostic/containerAnnotationStats

Used to determine data scale of annotations on containers. Returns min, avg, and max annotation count for all containers.

/model/diagnostic/containerWithoutMatchingNode

Used to determine if, in each window, there is container data without matching node data (matched on Provider ID).

ParameterDefaultDescription

containerTable

container_1d

The table containing container data to consider. Try also container_1d_reconciled.

nodeTable

node_1d

The table containing container data to consider. Try also node_1d_reconciled.

/model/diagnostic/containerDuplicateNoId

Used to determine if, in each window, there is duplicate container data according to (Cluster, Namespace, Controller Kind, Controller Name, Pod Name, Container Name).

ParameterDefaultDescription

containerTable

container_1d

The table containing container data to consider. Try also container_1d_reconciled.

/model/diagnostic/containerDuplicateWithId

Used to determine if, in each window, there is duplicate container data according to (Cluster, Namespace, Controller Kind, Controller Name, Pod Name, Container Name, Id).

ParameterDefaultDescription

containerTable

container_1d

The table containing container data to consider. Try also container_1d_reconciled.

/model/diagnostic/nodeDuplicateNoId

Used to determine if, in each window, there is duplicate node data according to (Cluster, Provider ID).

ParameterDefaultDescription

nodeTable

node_1d

The table containing container data to consider. Try also node_1d_reconciled.

Last updated