search

F5Networks / k8s-bigip-ctlr

Repository for F5 Container Ingress Services for Kubernetes & OpenShift.
Apache License 2.0
364 stars 195 forks source link

Expose metrics of the time which CIS finish configuration of BIGIP to Prometheus #2256

Closed kkfinkkfin closed 9 months ago

kkfinkkfin commented 2 years ago

Title

Expose metrics of the time which CIS finish configuration of BIGIP to Prometheus

Description

We want to know the time difference when CIS monitors the change of resource object to the BIGIP generation configuration, and expect to be able to pull this metrics data through prometheus.

Actual Problem

When the scale of cluster resource data is relatively large, CIS sometimes takes a long time to update the BIGIP.

Solution Proposed

It is expected to be able to pull in precise configuration times via prometheus for alerting.

Alternatives

It is possible to calculate the update time of every VS config and every POD related objects in the program and output each updated cost time. Then we can use PromSQL to calculate the average time, output the top 10 maximum time consumption, etc

Additional context

trinaths commented 2 years ago

Created [CONTCNTR-3177] for internal tracking.

olvandeng commented 2 years ago

PR: https://github.com/F5Networks/k8s-bigip-ctlr/pull/2265/files is to address the the requirement, talking more for details, and working with HYD team to review the code. Thanks.

trinaths commented 2 years ago

@olvandeng please link this github ID and fill the details for the PR summary.

trinaths commented 9 months ago

Closing as no further discussion.