S3 redirect loses query string

Hi, I’m having the following issue:

Say I have a page called nice-page. Currently, whenever I access a url without a forward slash at the end of its path, e.g.:

https://mywebsite.com/nice-page

S3 tries to find an object with that name (nice-page) but instead finds nothing, so it falls back to trying to find a folder with that name, therefore sending a 302 to the client and redirecting it to

https://mywebsite.com/nice-page/

(notice the forward slash at the end). This opens the page correctly, however, when the redirect happens, the browser loses all query string parameters, making we lose campaign traffic informations like utm_source, utm_campaign etc.

Do you guys have any idea of how to fix this?

[JoshuaWalsh]

@philippeehlert While testing my new terraform-aws-gatsby module I wrote a Lambda@Edge function which I believe achieves what you want. Attach this function to the Origin Response event:

'use strict';

exports.handler = (event, context, callback) => {
    var cf = event.Records[0].cf;
    var request = cf.request;
    var response = cf.response;

    if (response.status[0] === "3" && response.headers.location && response.headers.location[0] && request.querystring) {
        response.headers.location[0].value += "?" + request.querystring;
    }
    
    return callback(null, response);
};

You will need to ensure that on your CF distribution, “Query String Forwarding and Caching is set either to Forward All, Cache Based on Whitelist or to Forward All, Cache Based on All” as mentioned here.

This lambda function will preserve the query string for all 3XX responses. If you want to be more selective, you will have to modify the function.

[Tushar-Sawdays]

@JoshuaWalsh: What needs to be done to get the “301” in your code?