Short answer. Commerce Manager shows “Your data feed file couldn't be accessed” (or “Failed to connect to server”, “Fetching feed from server failed”, “Internal server error”, “HTTP authentication failed”) when Meta cannot fetch your file from its URL. Open the exact URL in a private browser window: it must download the file, not a page; then allow Meta’s crawler facebookexternalhit in robots.txt and your firewall, remove any login unless the file truly needs one, and set a Google Sheet to “Anyone with the link”.
| Shown in Commerce Manager as | Your data feed file couldn't be accessed |
|---|---|
| Effect | Whole feed blocked |
| Where to find it | Catalog → Products → a product’s Status and issues; feed-level errors under Catalog → Data sources → Download issue report |
| Free check | Not something a feed file shows; see below for what the check can tell you instead |
What Commerce Manager means by “Your data feed file couldn't be accessed”
Meta fetches a scheduled feed from the URL you gave it, and this family of messages on its data feed troubleshooting page all mean the same thing: the fetch did not return your file. The page lists the causes to check, in order: the URL must open or download the file in a browser and must start with http://, https://, ftp:// or sftp://; the server must not block Meta’s crawler, named facebookexternalhit; a username and password go in the feed settings only if the file is really protected; a Google Sheet must be shared with anyone who has the link.
Nothing in the catalogue changes while fetches fail, which is the quiet part: prices and stock stay as they were at the last good upload. Meta also pauses a schedule that has failed several times, so after the fix the feed’s Settings tab may need a Resume.
Why it happens
- The URL returns a page, not a file: a 404, a maintenance notice, a login screen or a homepage redirect.
- The crawler is blocked by robots.txt, a firewall or a CDN rule that challenges unknown user agents.
- Credentials entered for a public file, which makes the fetch fail with “HTTP authentication failed”.
- A restricted Google Sheet, shared only with specific people.
- A server error (5xx) at fetch time, or a host that rejects requests from Meta’s regions.
How to fix it
In Shopify
Shopify stores usually sync through the Facebook & Instagram channel rather than a feed URL, so this error means a third-party feed app’s URL is failing. Open that URL in a private window; if it asks for a store password or shows a page, fix the app’s settings or the storefront password.
In WooCommerce
Open the feed URL from your plugin in a private window and confirm it downloads XML. If it shows a 404 page, the feed’s pretty URL has no rewrite rule yet: our free plugin answers at its address regardless, but other plugins may need Settings → Permalinks saved once. Then allow facebookexternalhit through robots.txt and any security plugin, and remove maintenance mode.
In a CSV or XML feed
Follow Meta’s list: test the URL in a browser; check robots.txt for a disallow on facebookexternalhit and add it to the allow list; check the server log for Meta’s requests and their status; clear the username and password fields unless the file needs them; for Google Sheets, set sharing to Anyone with the link. Then re-upload or Resume the schedule.
How long until Commerce Manager clears it
The next successful fetch, scheduled or manual, replaces the error with an upload summary. If the schedule was paused after repeated failures, resume it from the feed’s Settings tab; it does not restart on its own.
What FeedRobin does about it
The free feed check fetches your feed URL the way Meta does, from a server with no cookies, so a URL that fails for Meta usually fails for the check too and the report says why (unreachable, a web page instead of a feed, an HTTP error). For connected stores, the Meta feed is served from our servers at a stable URL that answers every fetch, and if your store goes down we keep serving the last good version.
Questions merchants ask
Which user agent does Meta fetch feeds with?
Meta’s page names the crawler facebookexternalhit (facebookexternalhit/1.1) and says to make sure it is not on your robots.txt disallow list. Firewalls and CDN bot rules that challenge unknown agents block it too.
My feed URL opens in my browser. Why can’t Meta access it?
Your browser sends cookies and a referrer that Meta does not. A private window is a fair test; if it still opens there, check the server log for Meta’s requests and their status codes, and whether your host blocks by user agent or region.
Can the feed URL need a username and password?
Yes, and Commerce Manager has fields for them in the feed’s settings. Meta’s page adds the reverse warning: do not enter login details for a file that is not password-protected, or the fetch fails with “HTTP authentication failed”.
Sources
- Troubleshoot data feed errors in your catalog — Meta Business Help Centre
- Product data specifications for catalogs in Commerce Manager — Meta Business Help Centre