Monitoring in Neon

Overview

To find out what's going on with your Neon projects and databases, Neon offers several ways to visualize your usage and track your metrics:

• Project dashboard — Find widgets that display the most recent usage metrics across your project
• Autoscaling graphs — a visualization of recent vCPU and RAM usage to help understand your sizing needs.
• Branch-specific metrics — View key metrics for all branches in your project on the Branches page.
• API metrics — Use the Neon API to gather a variety of usage metrics for your project.
• neon_utils — A custom Postgres extension that helps you monitor how autoscaling allocates vCPU in response to workload.
• Other Postgres options — Other extensions and tools like pg_stat_statements and PgHero can provide details about SQL execution counts, totals, average execution time, and other insights into how your database is being used.

Dashboard

When you open a project in the Neon Console, the Dashboard gives you a snapshot view of the most recent usage metrics for that project:

• Branches widget — Shows how much storage your main branch has currently used. It also lists a subset of other branches.
• Usage since... widget — Shows a snapshot of your current consumption metrics since your last billing date (the last day of the previous month).
• Project settings widget — Shows current project settings like your compute size default and history retention window.
• Operations widget — Lists a subset of the most recent Neon operations like create_branch, start_compute, and suspend_compute. If there are any critical issues or events, you should see these show up here.

In any of these widgets, you can click the link in the top right corner to get more detail, a wider view, and take various actions.

Autoscaling graphs

For Pro users, when you edit your compute from the Branches page, there is a graph showing your recent vCPU and RAM usage. This can help visualize whether your current minimum and maximum sizes are suitable for your current level of demand.

See Monitoring Autoscaling for details.

Branch-specific metrics

The Branches page provides key usage metrics for all your branches, like active hours, compute hours, storage space used, and whether the branch includes a read-write compute, all organized into a table view that lets you scan your list of branches. You can also use search if your list of branches is quite long.

API metrics

Using the Neon API, you can collect a variety of usage-based consumption metrics like data_storage_bytes_hour and compute_time_seconds.

Use this GET request to get details from an individual project.

curl --request GET \
     --url https://console.neon.tech/api/v2/projects/[project_ID] \
     --header 'Accept: application/json' \
     --header "Authorization: Bearer $NEON_API_KEY" | jq

For more details, see Retrieving details about a project.

The neon_utils extension

The neon_utils extension provides a num_cpus() function you can use to monitor how Neon's Autoscaling feature allocates vCPU in response to workload. The function returns the current number of allocated vCPUs.

For full details, see neon_utils.

Other Postgres options

Neon recommends a few tools from the Postgres community that can help you understand what is happening within your database:

• pg_stat_activity
• pg_stat_statements
• PgHero

pg_stat_activity

This system view built into Postgres provides real-time information on current database connections and queries being executed. A helpful for determining who is accessing your database.

Try this query out in the Neon Console:

SELECT * FROM pg_stat_activity;

You should see a range of information about what's currently happening in the selected database. For details, see pg_stat_activity.

pg_stat_statements

A Postgres extension that tracks information such as execution counts, total and average execution times, latency, and so on, which can help you analyze and optimize SQL query performance.

Here are some typical queries that you might find helpful:

• Monitor slow queries
• Find the most frequently executed queries
• Find the most time-consuming queries

For more information, see our pg_stat_statements documentation page.

PgHero

PgHero is an open-source performance tool for Postgres that can help you find and fix data issues, using a dashboard interface.

A quick look at the interface gives you an idea of what you’ll find in PgHero.

Among other things, you can use PgHero to:

• Identify long-running queries
• Identify tables that require vacuuming
• Identify duplicate or missing indexes
• View connections by database and user
• Explain, analyze, and visualize queries

How to install PgHero

PgHero supports installation with Docker, Linux, and Rails. Here, we’ll show how to install PgHero with Docker and connect it to a Neon database.

Before you begin:

• Ensure that you have the pg_stat_statements extension installed. PgHero uses it for query stats. See above.
• Ensure that you have Docker installed. See Install Docker Engine for instructions.

PgHero is available on DockerHub. To install it, run:

docker pull ankane/pghero

Next, grab your Neon database connection string from the Connection Details widget in the Neon Dashboard.

Finally, run this command, replacing $NEON_DB with your Neon database connection string.

docker run -ti -e DATABASE_URL='$NEON_DB' -p 8080:8080 ankane/pghero

Then visit http://localhost:8080 in your browser to open the PgHero Dashboard.

Feedback and future improvements

At Neon, we understand that observability and monitoring are critical for running successful applications.

Check out our roadmap to see what's coming next. And if you've got feature requests or feedback about what you'd like to see in Neon observability features, let us know via the Feedback form in the Neon Console or in our feedback channel on Discord.