Allow these addresses so mfr.io can read your storefront.
Production
Allow all of them. Which address a given import uses depends on where it runs, so allowing only one will work until it does not.
These are not used by routine imports of a store that is already connected — they belong to the environment we use while a store is first being connected and tested. But if one of them is what your access log shows, that is the environment your store is being connected against, and allowing it is what will unblock the import.
Not sure which applies to you? Your mfr.io contact will know, and can confirm before you change anything.
If you host your own store — Magento, WooCommerce, or a WordPress content site — it very likely runs rate protection such as fail2ban, mod_evasive, or a firewall rule that blocks an IP address after it makes a certain number of requests in a short window. That protection is doing its job. It just cannot tell our importer apart from a scraper.
When it blocks us, the symptom is one-sided and easy to misdiagnose: your site is up, fast, and fine for every visitor, while mfr.io reports that it cannot connect. Nothing is wrong with your store. Our address is simply on a block list.
Every request we make identifies itself in your access log:
robots.txt, and your XML sitemap — plus the product images and documents those pages link to.If someone else manages your server, this is the whole request. It lists every address you need for this site, so your host only has to make the change once:
If your access log shows one of the staging addresses listed above, send this one instead. It adds that address and still covers everything else, so your host only has to make the change once:
They do not rotate, and they survive our own infrastructure changes. You should only ever have to do this once. If they ever do need to change, we will update this page before the change takes effect.
Questions about an import? Contact your mfr.io representative, or reply to the notification that brought you here.