There are quite a few community end users as well as monitoring vendors who are interested in a visualization and dashboarding solution that is under the Apache-2.0 license, owned by the Linux Foundation, and governed by a mix of community end users and vendors. The CoreDash Project is designed as a place where we can work together on this solution. Please note that CoreDash is not designed to be a fork of Grafana, but will instead take a more componetized approach as a collection of subprojects that users can select from to meet their needs.
CoreDash is the current working name / code name for this effort, but we expect to rename it after we've made a bit more progress and better defined what we plan to do.
Please join the coredash-discuss@googlegroups.com mailing list where we have discussions about the project.
We meet on the 1st and 3rd Thursday of every month at 7am PT | 10am ET | 15:00 UK for 1 hour. Agendas, meeting notes, and joining details can be found in our working session doc.
Participation in this community is subject to our Code of Conduct.
If you would like to contribute to this repo, please read our Contributing Guide.
Because this is an umbrella project, you should consider contributing to one of our subprojects, which will have different instructions for contributing depending on the project.
If you have a project (or want to start one) that would be a good fit for this effort, please reach out on the mailing list to discuss contributing a new subproject to this Linux Foundation effort.
NOTE: This scope section is still evolving and should be expected to change as we get more people and projects involved in this effort.
If we look at the Observability and Analysis segment of the CNCF Landscape, the three projects in metrics based monitoring (Prometheus, Cortex and Thanos) are all Prometheus ingest format and PromQL based solutions. The other graduated or incubating projects are Fluentd and Jaeger. Fluentd is a logging solution, which has no standard ingest protocol or query language yet. Jaeger has a custom ingestion protocol and custom querying / visualization / UX, and the community is standardizing the ingesting format with OpenTelemetry.
To give this new project the best chance of being successful, reducing the initial scope to provide faster value for end users is ideal from our perspective. Therefore, having a visualization project that is focused on Prometheus and PromQL seems to be a logical starting point. Most open source and vendored solutions in the CNCF Monitoring landscape support both Prometheus ingestion format as well as PromQL.
Perhaps over time, the project can look to add other industry standard monitoring protocol support like Graphite or expand into visualization for logging and distributed traces, but to start, the focus will be on deriving value for one use case (Prometheus/PromQL) and building up from there versus attempting to replicate all of the various visualization capabilities of Grafana.
The project will also improve the state of the art for finding and choosing Prometheus metrics to visualize, and for constructing PromQL queries.
NOTE: The goals section is still evolving and should be expected to change as we get more people and projects involved in this effort.
We aim to make CoreDash easy to use by not requiring highly specialized skills to be able to benefit from these tools and dashboards. Anyone with basic web front end skills (Javascript, etc.) should be able to create dashboard widgets and screens and connect them to a variety of data sources without needing to write complex code or be a sys admin / DevOps expert. We hope that by making these technologies more inclusive to people with a wide variety of skills, we can open the field to a larger set of well established users who create and customize dashboard screens.
Goals are broken down into two sections: 1) Dashboards and 2) Ease of querying.
Dashboards: Core visualization goals which are focused on visualization of metric data based on PromQL and allowing users to create savable and shareable dashboards from them:
Ease of Querying: Goal focused on helping beginners and first time users of PromQL to craft their queries. If PromQL is going to be the long term industry standard open source query language, many more end users will have to become familiar with it over time:
This project is Licensed under Apache 2.0.