Instance status page

[Twixes]

Is your feature request related to a problem?

We currently have metrics.posthog.com, which offers crucial performance information, but is a solution separate from PostHog itself. Otherwise we don't offer any metrics, which makes detecting and explaining PostHog instance problems more time-consuming for users and also for the core team.

Describe the solution you'd like

PostHog could use an instance status-type page (accessible only to organization admins on Self-Hosted or is_staff users on Cloud [or is_superuser? this distinction is actually useless for us]). For example, we could easily show database statistics like numbers of rows or table sizes, Redis health, event ingestion load – essentially data that comes up useful when users ask about problems.

Describe alternatives you've considered

Tell users to investigate instance internals themselves.

Additional context

Thought of this when a user needed to know why their PostHog Docker container was taking up so much space. I've told them to query for table sizes:

SELECT relname as "table", pg_size_pretty(pg_total_relation_size(relid)) AS "size" FROM pg_catalog.pg_statio_user_tables ORDER BY pg_total_relation_size(relid) DESC;

but it'd be much simpler if this information was available in the UI.

Thank you for your feature request – we love each and every one!

[timgl]

The list of things we should include

• Event table size (both in MB and Approximate size)
• Element table size (both in MB and Approximate size)
• Redis queue size
• Possibly memory size of workers?
• Number of events in last day/week

[Twixes]

Closed by #1875.