Skip to content

[BLOG] How to monitor Cortex #6736

Open
@oberkem

Description

@oberkem

Describe the blog post

There is a lot of inquiries in Cortex Slack channel about debugging Cortex's microservices. Maintainers usually ask the same questions about how many instances are running, size of the instances, resource utilization as well as RED metrics. It would be great to write a blog post and provide dashboards as templates as code (grafana/perses etc.)

Expected Title

Monitoring Cortex

Authors Name

Orcun Berkem if no takers

Authors Email

oberkem

Target Draft Date

No response

Blog Post Category

technical

Target Publication Date

No response

Additional Info

No response

Metadata

Metadata

Assignees

No one assigned

    Labels

    type/observabilityTo help know what is going on inside Cortex

    Type

    No type

    Projects

    Status

    No status

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions