devamanv
commented
1 year ago Updates and Progress so far
- Healthwatch and Healthwatch exporter have been installed as separate tiles in the Ops Manager instance running in GCP (as seen below in the screenshot). Although there are missing stemcells, due to invalid files available for download from VMWare Tanzu Network. But those can be easily fixed by importing the right ones.
- Configured URLs for Grafana dashboards to visualize metrics data. These are the dashboards we are concerned with and have some overlap with the Elastic dashboards, which will eventually have to be upgraded.
Next steps and suggestions:
- Healthwatch provides out-the-box [dashboards ]() and Health metrics similar to the currently supported v1.8, of which the most important ones are Super value metrics (SVMs) that the Prometheus instance in Healthwatch v2.2 generates in a similar format as the super metrics used in Pivotal Healthwatch v1.8 and earlier. A comparison between the two should reveal if all the metrics from the Pivotal Healthwatch 1.8 are coming through in addition to the new ones.
- Some metrics Healthwatch 2.2 will not be compatible with is 1 or 2 dashboards that we elastic have created. Those would need to be updated for full compatibility. Once we confirm that all the existing metrics are fine, we can reach out to the Kibana team to update the dashboards.
Current dashboards support data collected with Healthwatch 1.x (tested with 1.8), but they seem to lack support for Healthwatch 2.2.
Healthwatch 2.2 also exposes its metrics using the Prometheus exposition format (see docs), so maybe collection method for these metrics should be modified too. These metrics are collected now using the firehose, as any other metric collected from Cloud Foundry.
Add support for this version of Healthwatch.