Understanding storage quota

All browsers impose an upper limit on the amount of storage that your web app's origin is allowed to use. You can configure Workbox to automatically clean up the data it caches at runtime in order to avoid running into storage quota limitations that may impact the caching efficiency and reliability of your website.

What configuration options are supported?

When setting up a route and runtime caching strategy, you can add in an instance of ExpirationPlugin from workbox-expiration configured with settings that make the most sense for the type of assets you're caching.

For instance, the following configuration might be used for caching images at runtime, with both explicit limits as well as automatic cleanup if quota is exceeded:

import {registerRoute} from 'workbox-routing';
import {CacheFirst} from 'workbox-strategies';
import {ExpirationPlugin} from 'workbox-expiration';

registerRoute(
  ({request}) => request.destination === 'image',
  // Use a cache-first strategy with the following config:
  new CacheFirst({
    // You need to provide a cache name when using expiration.
    cacheName: 'images',
    plugins: [
      new ExpirationPlugin({
        // Keep at most 50 entries.
        maxEntries: 50,
        // Don't keep any entries for more than 30 days.
        maxAgeSeconds: 30 * 24 * 60 * 60,
        // Automatically cleanup if quota is exceeded.
        purgeOnQuotaError: true
      })
    ]
  })
);

You need to set maxEntries, maxAgeSeconds, or both when using ExpirationPlugin. purgeOnQuotaError is optional.

maxEntries

This imposes an upper limit on the number of entries (that is, unique URLs) for a given cache.

Setting this is normally a good idea, unless you know that there are only a small number of possible URLs that ever might be handled by a given strategy.

maxAgeSeconds

Entries that were added to the cache more than this number of seconds ago will be considered stale, and automatically cleaned up the next time the cache is accessed.

This is not as effective in managing storage quota as maxEntries, since your caches can grow arbitrarily large as long as the entries were all added within a small period of time. It's most useful when you know that there's an upper limit on freshness that you want to impose, and keeping around older entries has little value for your web app.

purgeOnQuotaError

This option allows you to mark a given cache as being safe to automatically delete in the event of your web app exceeding the available storage.

This option currently defaults to false. Runtime caches should, in general, be resilient in the face of deletion, so setting this option to true is a good practice, and helps ensure your web app can automatically recover in the face of storage constraints.

How much data are you allowed to store?

Each browser has its own upper limits on storage, so there's no single answer. Additionally, some browsers have a dynamic limit which varies based on the amount of storage that's free on a given device, so the effective upper limit might change without notice.

Some browsers expose an interface for querying the approximate amount of storage that your origin is using, along with the upper limit, via navigator.storage.estimate(). The "Estimating Available Storage Space" article has more information on how you can use that in your own web apps.

Special Chrome Incognito considerations

Opening a web app in Chrome's Incognito mode imposes a special restriction on storage that doesn't apply to normal browsing contexts: there's a quota limit of around 100 megabytes, regardless of free space available on your device.

Beware of opaque responses!

A common source of unexpectedly high quota usage is due to runtime caching of opaque responses, which is to say, cross-origin responses to requests made without CORS enabled.

Browsers automatically inflate the quota impact of those opaque responses as a security consideration. In Chrome, for instance, even an opaque response of a few kilobytes will end up contributing around 7 megabytes towards your quota usage.

You can quickly use up much more quota than you'd anticipate once you start caching opaque responses, so the best practice is to use ExpirationPlugin with maxEntries, and potentially purgeOnQuotaError, configured appropriately.