Fix an RSS Feed That Stopped Updating

A feed that worked last month and shows nothing today is the most common problem RSS readers hit, and almost always the one people misdiagnose. The instinct is that the XML has broken. In practice the far more likely causes are a cached copy being served, a feed address that moved, or a reader that simply has not checked yet. Work the checks below in order and you will usually find the cause in under two minutes.

First, understand the five kinds of broken

These look identical in your reader and have nothing in common, so knowing which one you have saves all the guesswork that follows.

Check 1: open the feed URL in a browser

Paste the feed address into a tab of its own, away from your normal pages. What you are looking for is simple: readable XML containing your newest item, or an unmistakable error. If you get an HTML page, a login wall or a 404, the problem is at the source and nothing else matters yet.

If you would rather not install anything, curl answers the same question:

A chain of redirects that never settles is effectively a broken feed for most readers, even though it opens fine in a browser.

Check 2: rule out caching before anything else

This is the single most common cause and it takes seconds to rule out. Publishers often serve feeds through a CDN or a caching plugin, and a cached copy can persist for minutes or hours after something new is published. Open the feed URL with cache bypassed and see whether the new item appears. If it appears only when you bypass the cache, the feed is healthy and the cache is the whole story. There is nothing for you to fix, and the new item will arrive at your reader on its next normal check.

A feed that is correct but not yet updated is a waiting problem, not a broken problem.

Check 3: know your reader's check interval

Readers poll on a schedule, not continuously. Minutes for some, an hour for others, longer on free tiers. A post published thirty seconds ago has usually not been looked for yet. Before concluding that anything is wrong, find out how often your reader checks, then wait one full interval.

This also explains feeds that look intermittent. Nothing is broken, you are simply watching a slow poller.

Check 4: did the site change recently?

If the feed worked and stopped, and it stopped soon after a site update, the update is the suspect. A theme change, a plugin update or a site migration can alter the markup the feed was generated from, or change the address entirely, without changing anything a visitor sees. If the feed was working a week ago and stopped the day after an update, stop debugging the reader and look at the site.

The same thing happens at the other end. A publisher may move a feed from one address to another and leave the old path returning an error rather than a redirect. That is the case our guide to finding a feed for any website exists to solve, including sites that hide the address in the page source.

Check 5: validate the XML last

Only after the first four come back clean is malformed XML worth chasing, and it is genuinely rarer than people expect. Run the feed through the W3C feed validator. It reports the first structural problem rather than making you read the document yourself, and it tolerates none of the leniency a browser shows you.

If you own the site and the validator complains, the usual culprits are an unescaped ampersand in a title, a stray control character inside an item, or an empty required field. If you do not own the site, this is where your fixing stops and reporting begins.

What you can fix, and what only the publisher can

Being clear about this saves a lot of wasted effort. As a reader you can fix a wrong saved address, a slow polling interval, a feed hidden behind a filter in your own reader, and a stale cache on your side. You cannot fix a site that has deleted its feed, a publisher that has stopped syndicating, a server blocking your reader's requests, or markup that a recent update broke.

Deleting and re-adding the feed is the most-repeated bad advice in RSS support, and it almost never works. Everyone following a feed shares it, so if it is broken for you it is broken for every other subscriber too. Re-adding only helps when the address you saved has changed, and then you need the new address rather than the same one entered again.

Replace a feed that is genuinely gone

When a source has moved or died, the fix is to find the current feed and swap it. Open the site's home page in a browser and look for the orange feed icon in the address bar, or read the source for a line containing application/rss+xml. If neither is there, the publisher may publish a page URL that your reader can convert into a feed on its own, which most readers will offer when a plain address fails.

Follow the new feed, confirm the first few items look right, then delete the dead one. Keep a copy of any folders you had built around it, because the new feed is a different subscription and will not inherit them.

This is where having an OPML export turns a chore into a minute. Export your list before you start, so if several feeds have died at once you are restoring rather than hunting. The mechanics are in Back Up Your Feeds with OPML, and the reason it matters is the one the Google Reader shutdown taught an entire generation of readers.

FAQ

Summary

Work the checks in order and stop at the first one that explains what you are seeing. Open the feed URL first, rule out caching second, confirm your reader's polling interval third, look for a recent site change fourth and validate the XML only if all four come back clean. Most stalled feeds turn out to be a cache or a moved address, and both are quick to resolve once you know which one you have.

Related Articles