Manual Setup

If you can’t (or prefer not to) run the configuration step, you can follow the instructions below to configure your application.

Create Initialization Config Files

Create two files in the root directory of your project, sentry.client.config.js and sentry.server.config.js. In these files, add your initialization code for the client-side SDK and server-side SDK, respectively. We've included some examples below.

For each configuration:

import * as Sentry from "@sentry/nextjs";

const SENTRY_DSN = process.env.SENTRY_DSN || process.env.NEXT_PUBLIC_SENTRY_DSN;

  dsn: SENTRY_DSN || "",
  // We recommend adjusting this value in production, or using tracesSampler
  // for finer control
  tracesSampleRate: 1.0,
  // ...
  // Note: if you want to override the automatic release value, do not set a
  // `release` value here - use the environment variable `SENTRY_RELEASE`, so
  // that it will also get attached to your source maps

If you want to instrument Next.js API Routes, which run on serverless, you need to wrap your handler in our withSentry wrapper in order to be able to capture crashes:

import { withSentry } from "@sentry/nextjs";

const handler = async (req, res) => {
  res.status(200).json({ name: "John Doe" });

export default withSentry(handler);

You can include your DSN directly in these two files, or provide it in either of two environment variables, SENTRY_DSN or NEXT_PUBLIC_SENTRY_DSN.

Create a Custom _error Page

In serverless deployment environments, including Vercel, the Next.js server runs in a "minimal" mode to reduce serverless function size. As a result, some of the auto-instrumentation done by @sentry/nextjs doesn't run, and therefore certain errors aren't caught. In addition, Next.js includes a custom error boundary which will catch certain errors before they bubble up to our handlers.

To capture these errors in Sentry, you can use the Next.js error page customization option. To do this, create pages/_error.js, and include the following:

import NextErrorComponent from "next/error";

import * as Sentry from "@sentry/nextjs";

const MyError = ({ statusCode, hasGetInitialPropsRun, err }) => {
  if (!hasGetInitialPropsRun && err) {
    // getInitialProps is not called in case of
    // As a workaround, we pass
    // err via _app.js so it can be captured
    // Flushing is not required in this case as it only happens on the client

  return <NextErrorComponent statusCode={statusCode} />;

MyError.getInitialProps = async ({ res, err, asPath }) => {
  const errorInitialProps = await NextErrorComponent.getInitialProps({

  // Workaround for, mark when
  // getInitialProps has run
  errorInitialProps.hasGetInitialPropsRun = true;

  // Running on the server, the response object (`res`) is available.
  // Next.js will pass an err on the server if a page's data fetching methods
  // threw or returned a Promise that rejected
  // Running on the client (browser), Next.js will provide an err if:
  //  - a page's `getInitialProps` threw or returned a Promise that rejected
  //  - an exception was thrown somewhere in the React lifecycle (render,
  //    componentDidMount, etc) that was caught by Next.js's React Error
  //    Boundary. Read more about what types of exceptions are caught by Error
  //    Boundaries:

  if (err) {

    // Flushing before returning is necessary if deploying to Vercel, see
    await Sentry.flush(2000);

    return errorInitialProps;

  // If this point is reached, getInitialProps was called without any
  // information about what the error might be. This is unexpected and may
  // indicate a bug introduced in Next.js, so record it in Sentry
    new Error(`_error.js getInitialProps missing data at path: ${asPath}`)
  await Sentry.flush(2000);

  return errorInitialProps;

export default MyError;

Extend Next.js Configuration

Use withSentryConfig to extend the default Next.js usage of Webpack. This will do two things:

  • Automatically call the code in sentry.server.config.js and sentry.client.config.js, at server start up and client page load, respectively. Using withSentryConfig is the only way to guarantee that the SDK is initialized early enough to catch all errors and start performance monitoring.
  • Generate and upload source maps to Sentry, so that your stacktraces contain original, demangled code.

Include the following in your next.config.js:

// This file sets a custom webpack configuration to use your Next.js app
// with Sentry.

const { withSentryConfig } = require("@sentry/nextjs");

const moduleExports = {
  // your existing module.exports

const sentryWebpackPluginOptions = {
  // Additional config options for the Sentry Webpack plugin. Keep in mind that
  // the following options are set automatically, and overriding them is not
  // recommended:
  //   release, url, org, project, authToken, configFile, stripPrefix,
  //   urlPrefix, include, ignore

  silent: true, // Suppresses all logs
  // For all available options, see:

// Make sure adding Sentry options is the last code to run before exporting, to
// ensure that your source maps include changes from all other Webpack plugins
module.exports = withSentryConfig(moduleExports, sentryWebpackPluginOptions);

Make sure to add the Sentry config last; otherwise, the source maps the plugin receives may not be final.

Configure Source Maps

By default, withSentryConfig will add an instance of SentryWebpackPlugin to the webpack plugins, for both server and client builds. This means that when you run a production build (next build), sentry-cli will automatically detect and upload your source files, source maps, and bundles to Sentry, so that your stacktraces can be demangled. (This behavior is disabled when running the dev server (next dev), because otherwise the full upload process would reoccur on each file change.)

To configure the plugin, pass a sentryWebpackPluginOptions argument to withSentryConfig, as seen in the example above. All available options are documented here.

Disable SentryWebpackPlugin

If you want or need to handle source map uploading separately, the plugin can be disabled for either the server or client build process. To do this, add a sentry object to moduleExports above, and set the relevant options there:

const moduleExports = {
  sentry: {
    disableServerWebpackPlugin: true,
    disableClientWebpackPlugin: true,

If you disable the plugin for both server and client builds, it's safe to omit the sentryWebpackPluginOptions parameter from your withSentryConfig call:

module.exports = withSentryConfig(moduleExports);

In that case you can also skip the sentry-cli configuration step below.

Use hidden-source-map

(New in version 6.17.1)

If you would like to use hidden-source-map rather than source-map as your webpack devtool, so that your built files do not contain a sourceMappingURL comment, add a sentry object to moduleExports above, and set the hideSourceMaps option to true:

const moduleExports = {
  sentry: {
    hideSourceMaps: true,

Note that this only applies to client-side builds, and requires the SentryWebpackPlugin to be enabled.

Widen the Upload Scope

(New in version 6.19.1)

If you find that there are in-app frames in your client-side stack traces that aren't getting source-mapped even when most others are, it's likely because they are from files in static/chunks/ rather than static/chunks/pages/. By default, such files aren't uploaded because the majority of the files in static/chunks/ only contain Next.js or third-party code, and are named in such a way that it's hard to distinguish between relevant files (ones containing your code) and irrelevant ones.

To upload all of the files in static/chunks/ anyway, add a sentry object to moduleExports above, and set the widenClientFileUpload option to true:

const moduleExports = {
  sentry: {
    widenClientFileUpload: true,

Configure sentry-cli

The SentryWebpackPlugin uses sentry-cli to manage releases and source maps, which can be configured in one of two ways - using configuration files, or with environment variables - both of which are discussed below. For full details, see the CLI configuration docs.

If you choose to combine the two approaches, the environment variables will take precedence over values set in the configuration files. One common approach is to set sensitive data (like tokens) in the environment and include everything else in the configuration files added to your VCS.

The URL, organization, and project properties identify your organization and project, and the auth token authenticates your user account.

Use Configuration Files

You should commit all the properties to your VCS, except the auth token. You can accomplish this by using two files: including the properties of your organization and project, and .sentryclirc including your auth token. This is the approach taken by the wizard and it allows you to commit the former while ignoring the latter in your VCS.

Here is an example:
# cli.executable=../path/to/bin/sentry-cli

Add the token to .sentryclirc:


And don't forget to ignore .sentryclirc in your VCS.

Use Environment Variables

Alternatively, the cli can be configured using environment variables.

Property nameEnvironment variable
Help improve this content
Our documentation is open source and available on GitHub. Your contributions are welcome, whether fixing a typo (drat!) to suggesting an update ("yeah, this would be better").