Changelog

New updates and product improvements

Supabase Edge Runtime version 1.54 is compatible with Deno 1.43.

Supabase's hosted platform was upgraded to use this release when serving Edge Functions starting today.

If you're using Supabase CLI for local development latest stable release 1.176.10, it adds compatibility for Deno 1.43.

How do I find which version of Edge Runtime I'm running?

Supabase CLI (local)

When you run supabase functions serve, it should show the current version of Edge Runtime used (and its Deno compatibility)


_10
> supabase functions serve
_10
_10
Setting up Edge Functions runtime...
_10
Serving functions on http://127.0.0.1:54321/functions/v1/<function-name>
_10
Using supabase-edge-runtime-1.54.2 (compatible with Deno v1.43.0)

Hosted Platform

You can check the served_by field in log events to see which Edge Runtime version was used to serve your function .

We try our best to maintain backward compatibility in these upgrades. If you're experiencing any issues, please feel free to make a support request

Supabase underwent Consolidation Month™ to focus on initiatives that improve the stability, scalability, and security of our products. We also have exciting product announcements that we can’t wait to share. Let’s dive in!

Consolidation Month™

We kicked off Consolidation Month (no it’s not actually trademarked) during the month of May. During this time, every product team within Supabase addressed outstanding performance and stability issues of existing features. Here’s a small subset of initiatives and product announcements as part of Consolidation Month:

Auth Launches @supabase/ssr for Better SSR Framework Support

The newly released @supabase/ssr package improves cookie management, developer experience, and handling of edge cases in various SSR and CSR contexts. We’ve added extensive testing to prevent issues that users experienced with the @supabase/auth-helpers package.

Announcement

pgvector v0.7.0 Release Features Significant Performance Improvements

pgvector v0.7.0 introduced float16 vectors that further improve HNSW build times by 30% while reducing shared memory and disk space by 50% when both index and underlying table use 16-bit float. The latest version also adds sparse and bit vectors as well as L1, Hamming, and Jaccard distance functions.

Announcement

Edge Functions Improves Memory Handling

The Edge Functions team has significantly reduced the error rate for functions encountering memory issues by implementing better safeguards. This has greatly minimized errors with the 502 status code. Additionally, status codes and limits are now documented separately.

Status Codes | Limits

Dashboard Supports Bigger Workloads as Projects Grow

The Supabase Dashboard is now better equipped to handle your projects, regardless of their size. We have implemented sensible defaults for the amount of data rendered and returned in the Table and SQL Editors to prevent browser performance issues while maintaining a snappy user experience.

Announcement

Realtime Standardizes Error Codes

Realtime now emits standardized error codes, providing descriptions of their meanings and suggested actions. This enhancement improves your error-handling code and helps to narrow down whether the issue lies with the database, Realtime service, or client error.

Realtime Error Codes

RLS AI Assistant v2

We’ve improved the prompt and output of our RLS AI Assistant by including best practices found in our RLS docs and upgrading to OpenAI’s newest GPT-4o. We’ve also introduced numerous test scenarios to make sure you’re getting the right security and performance recommendations by comparing parsed SQL with the help of pg_query.

Pull Request

Quick product announcements

  • [Functions] JSR modules are supported in Edge Functions & Edge Runtime [Announcement]

  • [Functions] Debug Edge Functions with Chrome DevTools [Docs]

  • [Functions] Use HonoJS web Framework with Edge Functions [Docs]

  • [Analytics] Log Drains is in Private Alpha [Announcement]

  • [Realtime] Realtime Authorization Early Access [Announcement]

  • [Docs] SQL to PostgREST API Translator [Docs]

  • [Client libs] Supabase JavaScript SDK Sentry Integration now supports Sentry SDK v8 [Commit]

Made with Supabase

  • GroupUp - organize social gatherings to hang out with friends [Website]
  • HabitKit - track habits, view daily progress, and stay motivated as you work towards your goals [Website]
  • Meteron AI - LLM and generative AI metering, load-balancing and storage [Website]
  • EQMonitor - An app that displays and notifies earthquake information in Japan [Website]
  • GitAuto - AI software engineer that writes, reads, and creates pull requests [Website]
  • GenPPT - Free AI powerpoint presentation generator to help you create beautiful slides in minutes [Website]

Community highlights

  • Make your queries 43,240x faster [Video]
  • Exploring Support Tooling at Supabase: A Dive into SLA Buddy [Article]
  • FlutterFlow SuperApp Complex Template : Developing Feed with Supabase [Video]
  • How We Use Supabase in Betashares Direct [Video]
  • AI Assistant to Chat with Supabase Database [Video]
  • How to use wrappers in Supabase [Video]
  • Build Realtime Apps with Next.js and Supabase [Video]
  • SvelteKit & Supabase Project Build [Video]
  • Next.js 14 x Supabase — Build a Team component using shadcn [Article]
  • Create a Real Time Chat App with Supabase and Angular [Article]

This discussion was created from the release Developer Updates - May 2024.

You might be wondering what we've been up to the past few weeks when we'd usually have some cadence in our weekly updates with our GitHub discussion updates - the team at Supabase had decided to commit to a month of consolidation by putting our efforts into each of the following pillars: Alerting, Testing, and Scaling. Let's jump right in to see what this means for the Dashboard! 💪🏻

🚨 Alerts: Enabling proactive resolution to issues that users run into

We want to find out about bugs before you do. In an effort to lower the time between deploying a bug and fixing it, we’ve set up some alerts that will enable us to be more proactively catch bugs, in particular for show-stopping problems (e.g that awful “A client side exception has occurred” screen)!

  • Added alerts for critical points of failures (PR) This includes any errors that will completely prevent you from using the Supabase dashboard.
  • Added alerts for full page client crash events with a custom error boundary (PR) Having a custom error boundary also allows us to potentially provide contextual hints/resolutions to the problem faced, such as what we did in this PR.

🛠️ Testing: Identifying and covering critical points of failure

We believe in ensuring that minimally our critical paths are covered by tests as opposed to ensuring our dashboard has 100% test coverage. Anything that might completely block our users from doing what they need to do should be caught as much as possible before reaching production.

  • Moved to Vitest + Integrating packages to make writing tests easier (PR) We’ve revamped our current test set up to use Vitest instead of Jest and have integrated Mock Service Workers (MSW) and next-router-mock. These are all to support writing better tests easier, and also to reduce the amount of configuration we need to set up the tests.
  • Expanded on Playwright for E2E tests (PR) We previously had a couple of Playwright tests written over in our tests repository that we’ve decided to shift over to our main repository. We’re also in the midst of writing more tests and making them public

📈 Scaling: Supporting bigger workloads as the project grows

The dashboard should be able to keep up with your projects, no matter how big they grow - and even if unable to, it should never block you, or leave you feeling confused. This effort involved several solutions: gracefully failing if unable to handle large workloads, better observability, or even just better error handling to allow our users to self-service most errors. We were able to get improvements out for several common issues. Check out each individual PR for more information 😄

  • Rendering large column data within the Table Editor (PR) We’re now limiting rendering each text/JSON based column data in the table editor at a max of 10kB when initially viewing the table, which should alleviate browser performance issues when rendering tables with large row data. You may load the entire column data on demand instead.
  • Handling large results within the SQL editor (PR) A limit parameter will now be added as a suffix to select queries that are run in the SQL editor, which should prevent the browser performance from degrading when trying to query too much data. The limit parameter can be optionally removed if your intention might be to pull data in greater quantity.
  • View ongoing queries in the SQL editor and support terminating them (PR) This came in response to a user who accidentally ran a large query through the SQL editor which ended up in the API request timing out, giving the impression that the query stopped running when in fact it was still running on the database in the background. This should help surface such problems and give the tooling required to abort erroneous queries.
  • Contextual errors when something goes wrong (PR) While the attached PR was more of a POC, this is the direction we want to move towards to by providing contextual information and suggest possible resolutions whenever users run into an error

Of course consolidation is not a one time event, but an ongoing effort that our team is committing to in parallel to shipping new and exciting features. We hope that as our users, you will all be able to benefit from this with less blockers to keep you from building cool stuff 🙂

As always, if you have any feedback for us - we're just a message away through the feedback widget to the top right corner of the dashboard! We're always listening 👂🏻 (as in reading 👀).

Go here for latest update

Hey everyone,

I'm Stojan a member of the Supabase Auth team, bringing some updates about what's next with @supabase/ssr. This is the recommended package that helps you use the Supabase JavaScript client with SSR frameworks such as NextJS, Remix, SvelteKit and others.

We've been quite busy recently gathering feedback, reviewing common complaints and bugs with the package, and using it in the popular SSR frameworks. We've identified a few areas needing improvement and we've already started implementing them.

The package is still on major version 0, indicating its beta status. We plan to move it to major version 1 in the coming months making it the preferred way of using the Supabase JavaScript library in your favorite SSR framework.

First, we'll extract @supabase/ssr's code from the auth-helpers repository into its own. We’re doing this because:

  • @supabase/auth-helpers-x (like for NextJS) is no longer supported by the team, as we expect people to move to @supabase/ssr.
  • It's no longer about "auth-helpers," but rather about how you can most effectively and ergonomically use the Supabase Client in various SSR and CSR contexts.
  • A standalone repo makes it easier for the community to contribute and for us to track issues.

We're going to release a fairly ground-up reimplementation of the package. This should come as version 0.4.0 around mid-June. We've received a lot of signal in the past months from developers and the community about possible improvements for developer ergonomics and better handling for edge cases.

The reason for this change is because @supabase/ssr is really just a thin layer for cookie management on top of @supabase/supabase-js. We've identified some improvements that reduce odd and difficult-to-diagnose behavior. The new implementation will boast over 90% test coverage, including testing for issues that we’ve seen commonly reported so far.

As part of the new implementation, we are changing the API. The old API will be deprecated when we reach v1.0.0. This is to ensure the best possible experience for both developers and users. For most use cases and happy paths, the deprecated API will continue working during the phase-out, but we encourage switching as soon as possible. Once we release v1.0.0, major version 0 will no longer be maintained.

The change in the API is quite simple, so here’s an example of how it will look like. Instead of defining three cookie access methods get, set and remove like so:


_13
createServerClient(SUPABASE_URL, SUPABASE_ANON_KEY, {
_13
cookies: {
_13
get: async (name) => {
_13
// ...
_13
},
_13
set: async(name, value, options) => {
_13
// ...
_13
},
_13
remove: async(name) => {
_13
// ...
_13
}
_13
}
_13
})

You would need to define two — getAll and setAll cookie access methods like so:


_10
createServerClient(SUPABASE_URL, SUPABASE_ANON_KEY, {
_10
cookies: {
_10
getAll: async() => {
_10
// return all cookies you have access to
_10
},
_10
setAll: async(cookiesToSet: { name: string; value: string; options: CookieOptions; }[]) => {
_10
// set the cookies exactly as they appear in the cookiesToSet array
_10
}
_10
}
_10
})

Note that for createBrowserClient nothing needs to be done in most cases, it automatically works with the document.cookie API.

The change should be trivial for most SSR frameworks, and we'll be sure to update the guides to instruct you on how to change your code into this new way of accessing cookies.

Thanks for all your feedback! Feel free to ask any questions below!

Log drains is currently private alpha, and is available for Teams and Enterprise customers. We are still firming up the pricing and documentation, however it will likely involve a flat fee and variable egress usage. This will be announced separately through official channels.

We will be supporting Datadog as our initial provider.

The following destinations are in the works:

  1. Elastic/Filebeat
  2. Syslog

We are currently onboarding interested customers manually, so please fill out this form to get started: https://forms.supabase.com/logdrains.

We've released a fix to the deployment instructions for the supabase-grafana monitoring application.

If you're ingesting the metrics endpoint into your pre-existing managed infrastructure without using the supabase-grafana app, this change does not affect you. If you're running the supabase-grafana app using the docker-compose mechanism, you are also not affected.

Fly applications launched off the repository between December 10, 2023, and May 16, 2024 are impacted, and will experience:

  • Fly application being shut down after periods of inactivity of a few minutes (default Fly.io behaviour)
  • Historical data will not be persisted after such a shutdown

The fix to the deployment instructions ensures that a persistent volume is created to store the data on, which prevents loss in case of a machine shutdown or restart. Additionally, autoshutdown behaviour is disabled, in order to prevent the app from being paused due to inactivity.

In order to fix an existing, already deployed Fly application, you can edit its configuration to disable auto_stop_machines, and create a persistent volume, and mount it at /data (similar to the updated deployment instructions). Please note that as the newly created persistent volume will be empty to start, any existing metrics data will not be preserved as part of this change. If doing so is necessary, you can initially mount it at a separate path, copy the data over, and finally mount it at /data.

If you need further help, please reach out to Support via https://supabase.help

Conversational AI assistant in the SQL Editor

This was previously behind a feature flag but we're now making this available by default, which will replace the existing single prompt UI that you saw previously at the top of the SQL editor. Once again, thank you all so much for the feedback that you've left us - we really appreciate them and they definitely do help in guiding us towards the ideal dashboard experience for everyone. 🙂🙏

We're also aware that the feature preview functionality is missing in the local set up - rest assured we're looking into it and hope to get a fix out soon for everyone!

PR: https://github.com/supabase/supabase/pull/23142

Link: https://supabase.com/dashboard/project/_/sql

A step towards slightly more contextual error messages

A topic that came up in one of our discussions internally was regarding self-serviceability, and we realised that our error messages could do a much better job than just informing users what the error is about - especially when their errors from Postgres directly and the messages could be slightly cryptic for those not familiar with Postgres (yet 😉). The PR linked here is just a small idea and example for what we plan to do with error messages in the future, by giving users more context about the errors like possible solutions and links to relevant documentation. Hopefully this will make using the dashboard slightly more easier 🙂

PR: https://github.com/supabase/supabase/pull/23135

Link: https://supabase.com/dashboard/project/_/editor

Other improvements and bug fixes

Branching

  • Disable branch reset while branch is initializing (PR)

Database

  • Allow searching for schema and tables when creating indexes (PR)
  • Allow SQL language for writing database functions (PR)

Here’s everything we shipped during our GA week:

Day 1 - Supabase is officially launching into General Availability (GA)

Supabase has moved to General Availability (GA) with over 1 million databases under management and over 2,500 databases launched daily. We’ve been production ready for years and now we are fully confident that we can help every customer become successful, from weekend projects to enterprise initiatives at organizations like Mozilla, 1Password, and PwC.

Full announcement | Video announcement | X space

Day 2 - Supabase Functions now supports AI models

Supabase Functions has added a native API that makes it easy to run AI models within your functions while removing nasty cold starts. You can use the gte-small embedding model to generate text embeddings or bring your own Ollama server to tap into many more embedding models and Large Language Models (LLMs) like Llama3 and Mistral. Soon we’ll provide hosted Ollama servers so you won’t have to manage them yourselves for a more seamless experience.

Blog post | Video announcement | X space

Day 3 - Supabase Auth now supports Anonymous sign-ins

Supabase Auth heard your requests and went to work building anonymous sign-ins which enable you to create temporary users that have yet to sign up for your application. This lowers the friction for visitors to use your application while making it easy to convert them to permanent users once they’re hooked.

Blog post | Video announcement | X space

Day 4 - Supabase Storage now supports the S3 protocol

Supabase Storage already has standard and resumable uploads and now supports the industry standard S3 protocol enabling multipart upload and compatibility with a myriad of tools such as AWS CLI, Clickhouse, and Airbyte for a wide array of use cases.

Blog post | Video announcement | X space

Day 5 - Supabase Security & Performance Advisor

Supabase has managed over 1 million databases over the last four years and has seen all manner of use cases with common pitfalls that we’re helping our customers address with our Security, Performance, and Index Advisors. These Advisors will help to surface and fix insecure database configurations and recommend database and query optimizations to keep your database secure and performant for your mission critical workloads.

Blog post | Video announcement | X space

GA Week Hackathon Winners

We are delighted that so many high quality projects were submitted but in the end there could only be one Best Overall Project. The decision wasn’t easy but the Supabase panel of judges chose vdbs (vision database SQL) for the honorific. Congratulations 👏 to @xavimonp who will receive the prize of Apple AirPods.

Full list of winners | All the submissions

One more thing from GA Week

Just kidding, there’s always more than one. Here’s more awesome things we shipped:

Community Highlights

  • Changing Databases 5 Times in 48 Hours Boosted Our Launch to 35,000 Views [Article]
  • Crazy new Supabase feature: Understand and learn about anonymous users [Video]
  • Support unstructured data in Postgres with JSON columns [Video]
  • Build an AI-powered blogging platform (Next.js, Langchain & CopilotKit) [Article]
  • How to Secure Your Supabase Database and Storage [Blog post]
  • Self-host Protomaps PMTiles on Supabase Storage [Video]
  • Supabase Realtime - How to deal with multiplayers in Next.js [Blog post]
  • The Hard Parts of Building an Application, Made Easy with Supabase [Article]

This discussion was created from the release Developer Updates - April 2024.

You can now use JSR packages in your Edge Functions. JSR is a modern package registry for JavaScript and TypeScript created by the Deno team. With JSR support, you can use the latest versions of popular Deno packages like Oak.

How to use:


_10
import { Application } from "jsr:@oak/oak/application";
_10
import { Router } from "jsr:@oak/oak/router";

For local development, you will need to update Supabase CLI for the version v1.166.1 or above.

Edge Functions also supports using NPM and deno.land/x packages. If you are already using them, no changes are needed.

Other improvements and bug fixes

We've been focusing on improving existing features on the dashboard and fixing some issues over the past week, so while we've got nothing shiny to shout out about, here's still a list of things that we've shipped! 🙂 As always, feel free to let us know if there's something that you guys really want to see in the dashboard - we'll see how we can make it happen 😉

General

  • Feedback widget will not clear its contents when closing until explicitly cleared or submitted [PR]

Table Editor

  • Reinstate link button for foreign keys in table editor side panel [PR]
  • Fix creating foreign key on new column after changing column's name [PR]

SQL Editor

  • Set column width of results to be relative to column content length [PR]

Authentication

  • Added Create policy CTA under each table for convenience [PR]

Storage

  • Added file size validation against project's upload limit when uploading files in dashboard [PR]

Database

  • Query performance: Fix searching via role and query [PR]
  • Query performance: Add db inspect docs link for visibility to aid in helping identify potential Postgres issues [PR]
  • Enumerated types: clean up form field when reopening create enumerated type panel [PR]
  • Tables: Add ellipses to table descriptions to prevent odd wrapping for long descriptions [PR]

Build in a weekend, scale to millions