ember-cli-bugsnag

Ember-cli-bugsnag

Installation

Install the addon:

ember install ember-cli-bugsnag

💥 ember-cli-bugsnag 3.0.0 includes some breaking changes. Learn more.

Configuration

There are two ways to configure ember-cli-bugsnag:

1. Add POJO to config/environment:

{
  bugsnag: {
    apiKey: '',
    enabledReleaseStages: ['development', 'production']
  }
}

The releaseStage defaults to the current application environment, if you need to set a different releaseStage that diverges from the environment, you can pass and additional attribute to the bugsnag configuration called releaseStage. It would look like this:

{
  bugsnag: {
    apiKey: '',
    enabledReleaseStages: ['development', 'production', 'staging'],
    releaseStage: 'staging'
  }
}

1. Specify environment variables:

export BUGSNAG_API_KEY=''
export BUGSNAG_ENABLED_RELEASE='development,production'

Configuration options:

• config.bugsnag.apiKey / BUGSNAG_API_KEY -- required
• config.bugsnag.enabledReleaseStages / BUGSNAG_ENABLED_RELEASE -- optional, defaults to [] (never notify).
• config.bugsnag.releaseStage / BUGSNAG_RELEASE_STAGE -- optional, defaults to config.environment.
• config.bugsnag.endpoints / BUGSNAG_ENDPOINTS -- optional, defaults to what the libraryUrl uses.
• config.currentRevision -- any string representing the current version of the app, e.g. "1b8ef2c7" or "v1.2.4", optional.
  • Defaults to the version specified in package.json, e.g. 0.1.0.
  • This can be set automatically at build time with ember-git-version.

Customization

In order to send additional data along with errors reported to Bugsnag, generate a utility named bugsnag:

ember g util bugsnag

Custom Diagnostics (docs)

To send custom metadata, define a helper method getMetadata in the app/utils/bugsnag.js you created. getMetadata takes the error and the container as arguments, e.g.:

export function getMetadata(error, container) {
  return {
    // …some metadata
  };
}

ember-cli-bugsnag calls this method for every error and reports any data returned by it to Bugsnag as metadata for the respective error. The returned metadata should be formatted to correspond with tabs in your interface. E.g. for an Account tab:

return {
  account: {
    name: "Bugsnag",
    plan: "premium",
    beta_access: true
  }
};

Identifying Users (docs)

To correlate a specific user to an error and have the information appear in the User tab in the Bugsnag UI, send user data with each error data. Define a helper method getUser in the app/utils/bugsnag.js you created. getUser takes the container as an argument. For example, if you have a currentUser service that references a user model in your app:

import Ember from 'ember';

const {
  getProperties
} = Ember;

export function getUser(owner) {
  const currentUser = owner.lookup('service:current-user').get('user');
  const {
    email,
    id,
    fullName: name
  } = getProperties(currentUser, 'email', 'id', 'fullName');

  return {
    email,
    id,
    name
  };
}

Uploading Sourcemaps (docs)

Uploading sourcemaps to Bugsnag makes it easier to track down errors in your code because the stacktrace for each error in the Bugsnag UI highlights the exact line in your unminified source code. To send sourcemaps to Bugsnag, you can use the Ember CLI Deploy addon ember-cli-deploy-bugsnag.

Upgrading to 3.0

ember-cli-bugsnag 3.0 includes some changes to bring this addon in line with the latest from the bugsnag-js library and accompanying documentation.

1. Rename config.bugsnag.notifyReleaseStages/BUGSNAG_NOTIFY_RELEASE to config.bugsnag.enabledReleaseStages/BUGSNAG_ENABLED_RELEASE

2. config.bugsnag.endpoint/ BUGSNAG_ENDPOINT => config.bugsnag.endpoints/BUGSNAG_ENDPOINTS

3. Rename getMetaData => getMetadata in app/utils/bugsnag.js