Help Center/Admin Operations/Read Database Health
Admin Operations•Last updated: 2026-10-20•3 min read

Read Database Health

Database Health runs read-only probes on the live Postgres: size, connections, cache hit ratio, the 15 largest tables with their index counts and any statement running longer than 5 seconds.

Have questions or need step-by-step guidance on this topic?

Ask AI Copilot

Step-by-Step Procedure

1

Open the page

Open "System Operations" and click "Database Health" (address /admin/system/db-health).

2

Read the tiles

"Database size", "Connections" (total of the maximum, with active and idle-in-transaction counts), "Cache hit ratio" (it says "no reads yet" when nothing has been read), "Tables / indexes", "Long-running queries" and "Probe latency".

3

Read the table

Rows of kind "Table" give the estimated row count, total size including indexes, number of indexes and dead rows with the last autovacuum. Rows of kind "Long-running query" show the process, how long it has run and the statement with its literal values replaced by question marks.

4

Filter

Use the Status filter: OK shows tables, SLOW shows long-running queries.

Expected Operational Outcome

You can tell whether the database is connected, how big it is, which tables are largest and whether anything is running unusually long.

Troubleshooting & Failure Modes

Symptom: The page says the database did not answer
Likely Cause: The probes timed out or the connection failed.
Resolution: Press "Refresh"; if it persists check "Services" and the Neon console.

Was this guide helpful?

Help us refine our platform documentation