SSR

React Query supports two ways of prefetching data on the server and passing that to the queryClient.

• Prefetch the data yourself and pass it in as initialData
  • Quick to set up for simple cases
  • Has some caveats
• Prefetch the query on the server, dehydrate the cache and rehydrate it on the client
  • Requires slightly more setup up front

Using Next.js

The exact implementation of these mechanisms may vary from platform to platform, but we recommend starting with Next.js which supports 2 forms of pre-rendering:

• Static Generation (SSG)
• Server-side Rendering (SSR)

React Query supports both of these forms of pre-rendering regardless of what platform you may be using

Using initialData

Together with Next.js's getStaticProps or getServerSideProps, you can pass the data you fetch in either method to useQuery's' initialData option. From React Query's perspective, these integrate in the same way, getStaticProps is shown below:

1 export async function getStaticProps() {
2   const posts = await getPosts()
3   return { props: { posts } }
4 }
5 
6 function Posts(props) {
7   const { data } = useQuery(['posts'], getPosts, { initialData: props.posts })
8 
9   // ...
10 }

The setup is minimal and this can be a quick solution for some cases, but there are a few tradeoffs to consider when compared to the full approach:

• If you are calling useQuery in a component deeper down in the tree you need to pass the initialData down to that point
• If you are calling useQuery with the same query in multiple locations, you need to pass initialData to all of them
• There is no way to know at what time the query was fetched on the server, so dataUpdatedAt and determining if the query needs refetching is based on when the page loaded instead

Using Hydration

React Query supports prefetching multiple queries on the server in Next.js and then dehydrating those queries to the queryClient. This means the server can prerender markup that is immediately available on page load and as soon as JS is available, React Query can upgrade or hydrate those queries with the full functionality of the library. This includes refetching those queries on the client if they have become stale since the time they were rendered on the server.

To support caching queries on the server and set up hydration:

• Create a new QueryClient instance inside of your app, and on an instance ref (or in React state). This ensures that data is not shared between different users and requests, while still only creating the QueryClient once per component lifecycle.
• Wrap your app component with <QueryClientProvider> and pass it the client instance
• Wrap your app component with <Hydrate> and pass it the dehydratedState prop from pageProps

1 // _app.jsx
2 import { Hydrate, QueryClient, QueryClientProvider } from 'react-query'
3 
4 export default function MyApp({ Component, pageProps }) {
5   const [queryClient] = React.useState(() => new QueryClient())
6 
7   return (
8     <QueryClientProvider client={queryClient}>
9       <Hydrate state={pageProps.dehydratedState}>
10         <Component {...pageProps} />
11       </Hydrate>
12     </QueryClientProvider>
13   )
14 }

Now you are ready to prefetch some data in your pages with either getStaticProps (for SSG) or getServerSideProps (for SSR). From React Query's perspective, these integrate in the same way, getStaticProps is shown below.

• Create a new QueryClient instance for each page request. This ensures that data is not shared between users and requests.
• Prefetch the data using the clients prefetchQuery method and wait for it to complete
• Use dehydrate to dehydrate the query cache and pass it to the page via the dehydratedState prop. This is the same prop that the cache will be picked up from in your _app.js

1 // pages/posts.jsx
2 import { dehydrate, QueryClient, useQuery } from 'react-query';
3 
4 export async function getStaticProps() {
5   const queryClient = new QueryClient()
6 
7   await queryClient.prefetchQuery('posts', getPosts)
8 
9   return {
10     props: {
11       dehydratedState: dehydrate(queryClient),
12     },
13   }
14 }
15 
16 function Posts() {
17   // This useQuery could just as well happen in some deeper child to
18   // the "Posts"-page, data will be available immediately either way
19   const { data } = useQuery(['posts'], getPosts)
20 
21   // This query was not prefetched on the server and will not start
22   // fetching until on the client, both patterns are fine to mix
23   const { data: otherData } = useQuery(['posts-2'], getPosts)
24 
25   // ...
26 }

As demonstrated, it's fine to prefetch some queries and let others fetch on the queryClient. This means you can control what content server renders or not by adding or removing prefetchQuery for a specific query.

Caveat for Next.js rewrites

There's a catch if you're using Next.js' rewrites feature together with Automatic Static Optimization or getStaticProps: It will cause a second hydration by React Query. That's because Next.js needs to ensure that they parse the rewrites on the client and collect any params after hydration so that they can be provided in router.query.

The result is missing referential equality for all the hydration data, which for example triggers whereever your data is used as props of components or in the dependency array of useEffects/useMemos.

Using Other Frameworks or Custom SSR Frameworks

This guide is at-best, a high level overview of how SSR with React Query should work. Your mileage may vary since there are many different possible setups for SSR.

If you can, please contribute your findings back to this page for any framework specific guidance!

On the Server

• Create a new QueryClient instance inside of your request handler. This ensures that data is not shared between different users and requests.
• Using the client, prefetch any data you need
• Dehydrate the client
• Render your app with the client provider and also using the dehydrated state. This is extremely important! You must render both server and client using the same dehydrated state to ensure hydration on the client produces the exact same markup as the server.
• Serialize and embed the dehydrated cache to be sent to the client with the HTML
• Clear the React Query caches after the dehydrated state has been sent by calling queryClient.clear()

SECURITY NOTE: Serializing data with JSON.stringify can put you at risk for XSS-vulnerabilities, this blog post explains why and how to solve it

1 import { dehydrate, Hydrate, QueryClient, QueryClientProvider } from 'react-query';
2 
3 async function handleRequest (req, res) {
4   const queryClient = new QueryClient()
5   await queryClient.prefetchQuery(['key'], fn)
6   const dehydratedState = dehydrate(queryClient)
7 
8   const html = ReactDOM.renderToString(
9     <QueryClientProvider client={queryClient}>
10       <Hydrate state={dehydratedState}>
11         <App />
12       </Hydrate>
13     </QueryClientProvider>
14   )
15 
16   res.send(`
17     <html>
18       <body>
19         <div id="root">${html}</div>
20         <script>
21           window.__REACT_QUERY_STATE__ = ${JSON.stringify(dehydratedState)};
22         </script>
23       </body>
24     </html>
25   `)
26 
27   queryClient.clear()
28 }

Client

• Parse the dehydrated cache state that was sent to the client with the HTML
• Create a new QueryClient instance
• Render your app with the client provider and also using the dehydrated state. This is extremely important! You must render both server and client using the same dehydrated state to ensure hydration on the client produces the exact same markup as the server.

1 import { Hydrate, QueryClient, QueryClientProvider } from 'react-query'
2 
3 const dehydratedState = window.__REACT_QUERY_STATE__
4 
5 const queryClient = new QueryClient()
6 
7 ReactDOM.hydrate(
8   <QueryClientProvider client={queryClient}>
9     <Hydrate state={dehydratedState}>
10       <App />
11     </Hydrate>
12   </QueryClientProvider>,
13   document.getElementById('root')
14 )

Tips, Tricks and Caveats

Only successful queries are included in dehydration

Any query with an error is automatically excluded from dehydration. This means that the default behavior is to pretend these queries were never loaded on the server, usually showing a loading state instead, and retrying the queries on the queryClient. This happens regardless of error.

Sometimes this behavior is not desirable, maybe you want to render an error page with a correct status code instead on certain errors or queries. In those cases, use fetchQuery and catch any errors to handle those manually.

Staleness is measured from when the query was fetched on the server

A query is considered stale depending on when it was dataUpdatedAt. A caveat here is that the server needs to have the correct time for this to work properly, but UTC time is used, so timezones do not factor into this.

Because staleTime defaults to 0, queries will be refetched in the background on page load by default. You might want to use a higher staleTime to avoid this double fetching, especially if you don't cache your markup.

This refetching of stale queries is a perfect match when caching markup in a CDN! You can set the cache time of the page itself decently high to avoid having to re-render pages on the server, but configure the staleTime of the queries lower to make sure data is refetched in the background as soon as a user visits the page. Maybe you want to cache the pages for a week, but refetch the data automatically on page load if it's older than a day?

High memory consumption on server

In case you are creating the QueryClient for every request, React Query creates the isolated cache for this client, which is preserved in memory for the cacheTime period. That may lead to high memory consumption on server in case of high number of requests during that period.

On the server, cacheTime defaults to Infinity which disables manual garbage collection and will automatically clear memory once a request has finished. If you are explicitly setting a non-Infinity cacheTime then you will be responsible for clearing the cache early.

To clear the cache after it is not needed and to lower memory consumption, you can add a call to queryClient.clear() after the request is handled and dehydrated state has been sent to the client.

Alternatively, you can set a smaller cacheTime.