# RSS Feed Not Updating? 3 Checks That Find the Cause

> RSS feed not updating? Three checks find the cause in two minutes: open the feed, read the newest item's date, get the status code. Real examples.

Published by Hushread, 2026-10-18. Canonical: https://hushread.com/blog/rss-feed-not-updating

*Disclosure: Hushread is my app. I've tried to be fair, and I say where the others are the better choice.*

When an RSS feed is not updating, the cause is almost always one of three things. The publisher stopped writing to the feed (it still loads, but the newest item is old). The site now refuses your reader (a 403, 429 or 5xx error instead of a feed). Or the feed is fine and your reader simply hasn't checked it yet. Open the feed URL, look at the date of the newest item, and get the status code: those three checks tell you whether to wait, fix the URL, or replace the source.

## Check 1: open the feed URL in a browser

Copy the feed address out of your reader (most show it under the feed's settings or info) and paste it into a browser's address bar. You should see XML: a page of tags starting with `<rss>` or `<feed>`, or a styled version of it.

If you see the site's home page, a login page or a "Just a moment..." screen instead, the reader is getting the same thing, and it isn't a feed. Skip to check 3.

## Check 2: read the newest item's date, not the feed's date

A feed that loads can still be dead. The giveaway is the item dates. In RSS, each item's `<pubDate>` is "when the item was published", while the channel's `<lastBuildDate>` is only "the last time the content of the channel changed" ([RSS 2.0 specification](https://www.rssboard.org/rss-specification)). Software can bump the second without adding a single story, so trust the first. In an Atom feed, look at each entry's `<published>` or `<updated>`, not the `<updated>` at the top of the feed.

CNN's international feed is the clearest example I know. `rss.cnn.com/rss/edition.rss` answers instantly with 50 stories and a `lastBuildDate` of August 22, 2024. Nearly all the stories are from April 2023 (one 2023 golf feature carries an April 2024 date), led by "Trump pleads not guilty to 34 felony counts".

![CNN's edition.rss feed opened in Chrome, showing a lastBuildDate of 22 August 2024, a channel pubDate of 5 April 2023 and a ttl of 10 above a first item headlined "Trump pleads not guilty to 34 felony counts"](/blog-img/rss-feed-not-updating/cnn-frozen-feed.webp)
*CNN's international edition feed as Chrome shows it, 18 October 2026. Source: [rss.cnn.com](http://rss.cnn.com/rss/edition.rss)*

The same feed still tells readers to come back every 10 minutes (`<ttl>10</ttl>`), which is why it keeps looking healthy in reader apps. CNN's old feeds directory at `cnn.com/services/rss` now redirects to the CNN home page, so there is nowhere official to find a replacement.

To check dates without scrolling, press Ctrl+F (Cmd+F on a Mac) in the browser and search for `pubDate`, or run this in a terminal:

```
curl -s -A "Mozilla/5.0" https://example.com/feed | grep -o "<pubDate>[^<]*" | head
```

If the newest date is weeks or months old and the website itself has new stories, the feed is frozen at the source. No reader setting will fix it.

### A valid feed can still be a dead one

Validators check that the XML follows the rules, not that anything new is in it. CNN's frozen feed gets the green badge:

![The RSS Advisory Board validator result for rss.cnn.com/rss/edition.rss saying "This is a valid RSS feed", with only a missing atom:link recommendation](/blog-img/rss-feed-not-updating/w3c-validator.webp)
*The RSS Advisory Board validator on CNN's feed, 18 October 2026. Source: [rssboard.org](https://www.rssboard.org/rss-validator/check.cgi?url=http%3A%2F%2Frss.cnn.com%2Frss%2Fedition.rss)*

A validator such as the [W3C Feed Validation Service](https://validator.w3.org/feed/) is the right tool when a reader says a feed can't be parsed, and no help at all with a feed that parses perfectly and never changes.

## Check 3: get the status code

The status code names the problem. Run this in a terminal (Windows has included curl since Windows 10 build 17063, [Microsoft says](https://devblogs.microsoft.com/commandline/tar-and-curl-come-to-windows/)):

```
curl -sI -A "Mozilla/5.0" https://example.com/feed
```

The first line of the reply is the status, such as `HTTP/2 200` or `HTTP/1.1 502 Bad Gateway`. Keep the `-A "Mozilla/5.0"` part: some sites block curl's default identity. Reddit answered `403` to plain curl on October 18 and `200` with a browser's user agent, so without it you can mistake your own test for the problem. No terminal? Open the browser's developer tools (F12), pick the Network tab and reload the feed; the Status column shows the same code.

Status codes checked on 18 October 2026:

| Status | What it means | Example | What to do |
|---|---|---|---|
| 200, old items | Feed works, publisher stopped updating it | CNN `edition.rss` | Find another feed or follow the site another way |
| 301 / 302 | Feed moved | `cnn.com/services/rss` → home page | Update the URL if the redirect lands on a feed; if it lands on a home page, the feed is gone |
| 403 | Refused | The Economist; Reddit to plain curl | Usually a bot wall; only the site owner can lift it |
| 404 / 410 | Not there | | Look for a new feed URL on the site |
| 429 | Too many requests | Reddit, second request within a minute | Refresh less often; it clears on its own |
| 500 / 502 / 503 | The site's servers are failing | The Washington Post's world feed, `502` | Wait; if it lasts weeks, treat the feed as dead |

A 410 is the clearest signal of all: the resource is gone "and that this condition is likely to be permanent", and clients "should not repeat requests" ([MDN](https://developer.mozilla.org/en-US/docs/Web/HTTP/Reference/Status/410)). A 429 is the opposite, a temporary limit on how many requests you sent "in a given amount of time", sometimes with a `Retry-After` header saying how long to wait ([MDN](https://developer.mozilla.org/en-US/docs/Web/HTTP/Reference/Status/429)). Reddit's limit is tight enough that a second subreddit fetched straight after the first got a 429; the [subreddit RSS guide](/blog/subreddit-rss-feed) covers working round it until Reddit switches its feeds off on November 13.

### Feeds that block readers but let browsers through

The Economist's `latest/rss.xml` answers feed readers, curl and even a headless browser with a 403 and a Cloudflare "Performing security verification" page. The response even carries a header that says so: `cf-mitigated: challenge`. A person can tick the box. A feed reader can't.

![The Economist's RSS address showing a Cloudflare "Performing security verification" page with a "Verify you are human" checkbox instead of a feed](/blog-img/rss-feed-not-updating/economist-challenge.webp)
*What a headless browser gets from economist.com/latest/rss.xml, 18 October 2026. Source: [economist.com](https://www.economist.com/latest/rss.xml)*

This is why a feed can open fine on your laptop and fail in a web-based reader. Feeder's help pages put it plainly: "Bot protection often lets people in browsers through but blocks servers, and Feeder checks feeds from its servers" ([Feeder](https://feeder.co/help/rss/troubleshooting/why-a-feed-stops-updating/)). Inoreader's fetcher page notes that Cloudflare's Bot Fight Mode on the free plan "challenges verified bots too" ([Inoreader](https://www.inoreader.com/feed-fetcher)). No reader, Hushread included, gets past a challenge page.

## When the feed is fine and the reader is the problem

If the feed shows today's stories in a browser and returns 200, your reader hasn't fetched it yet. Reader schedules are slower than most people assume:

- **Feedly** polls Basic (free) accounts anywhere from every 30 minutes to once a day, depending on how popular the site is and how often it publishes. Pro is every 15 minutes to an hour, Pro+ as often as every 7 minutes ([Feedly help](https://docs.feedly.com/article/212-how-often-does-feedly-update)).
- **Inoreader** checks an active feed every 30 minutes to a few hours. A feed with no new content for a month drops to about once a day, and after a year to about once a week. Feeds no active user reads aren't fetched at all.

So a blog that posts monthly can sit a day behind in Inoreader even after it publishes.

Most readers also tell you what went wrong if you know where to look. Feeder turns a feed red after four failed fetches in a row and shows the reason under **More > Check error**. NetNewsWire has an Error Log under **Window > Error Log** (Cmd-9) on the Mac, and **Troubleshooting > Error Log** in settings on iPhone ([NetNewsWire help](https://netnewswire.com/help/error-log.html)). Read the error before you start changing things.

Deleting and re-adding the feed rarely helps with a web-based reader. Feeder's help explains why: "everyone who follows a feed shares it", so you get the same copy back.

## If it's your own site's feed

When the feed that stopped is yours, the usual culprit is a cache serving an old copy: a caching plugin, a CDN, or file permissions that stop the cache from refreshing. The WebKit project's blog feed stalled for about ten weeks in 2018 until someone fixed the cache file permissions ([WebKit bug 190785](https://bugs.webkit.org/show_bug.cgi?id=190785)). Fetch your feed with the curl command above and compare it with the newest post on the site. If you publish through FeedBurner, check the source feed it points to; Google kept that setting when it cut FeedBurner back to core features in 2021 ([FeedBurner help](https://support.google.com/feedburner/answer/10483501)).

## When the feed is dead: what to do next

Once the checks say a feed you follow is frozen, gone or walled off, you have four ways forward.

1. **Look for a newer feed on the site.** Publishers move feeds without redirecting the old address. The steps in [how to find the RSS feed of any website](/blog/find-rss-feed-of-any-website) take a couple of minutes, and [the best RSS feeds for news](/blog/best-news-rss-feeds) lists working feeds for most big outlets.
2. **Run a broken feed through a fixer.** Open RSS rebuilds feeds that parse badly: add `openrss.org/` to the start of the page address. Its guide says it corrects invalid syntax and removes "improper caching implementations that prevent a feed from updating" ([Open RSS](https://openrss.org/guides/fix-an-rss-feed-using-open-rss)). It can't help with a feed the publisher stopped writing to.
3. **Follow the site itself instead of its feed.** In Hushread, paste the site's address into **Add source**. When there's no feed to use, it offers **Follow this page**, checks the page regularly and turns new articles into feed items ([Hushread docs](/docs/add-a-source)). You stop depending on a feed the publisher may never fix. The free plan covers 5 sources.
4. **Tell the publisher.** A frozen feed is often a broken cache or plugin nobody has noticed. An email with the feed URL and the date of its newest item is the most useful bug report a site owner can get.

## FAQ

### Why is my RSS feed not updating?

Either the publisher stopped adding items, the site is refusing your reader with an error such as 403 or 502, or your reader checks that feed only every few hours. Open the feed in a browser and look at the newest item's date: if it's old, the problem is at the source.

### Why did my RSS feed stop working all of a sudden?

Usually the site changed something: it moved the feed, turned on bot protection, or broke the software that writes it. Run `curl -sI -A "Mozilla/5.0"` on the feed URL; a 404 means it moved or was removed, a 403 often means a bot wall, and a 5xx means the site is failing.

### How do I know if an RSS feed is dead?

Compare the newest `<pubDate>` in the feed with the newest story on the website. If the site has published since and the feed hasn't changed in weeks, the feed is dead, even if it still loads and a validator calls it valid.

### Why is my feed reader not updating one feed when the others work?

That feed either returns an error to the reader's servers or is on a slow schedule. Inoreader drops feeds without new posts for a month to about one check a day, and Feedly's free plan can be as slow as once a day. Check the reader's error message for that feed.

If the feed works and your reader is slow, a paid plan or a reader that fetches more often is the fix, and you don't need to switch anything else. If the feed is dead and the site has no other, following the page directly is the way to keep reading it, which [Hushread](https://hushread.com/#download) can do.