The following example handles server and connectivity errors for javascript fetch requests.

 1function handleErrors(response) {
 2    if (!response.ok) throw new Error(response.status)
 3    return response
 4}
 5
 6fetch('https://jsonplaceholder.typicode.com/todos/1')
 7    // handle network err/success
 8    .then(handleErrors)
 9    // use response of network on fetch Promise resolve
10    .then(response => console.log("ok"))
11    // handle fetch Promise error
12    .catch(error => console.log('found an error: ', error))

handleErrors can be updated for handling different types of errors in different ways. This function is called whenever we get a response from the server (successful or failed). When we have a different type of error, like a DNS one, the catch block is called directly.

All errors are caught by the catch section, because server errors end up in throwing a new Error instance. Such Error instance can be extended to meet specific needs.

References

• Amirmohammad Moradi. Fetch API error handling.