Using a self-hosted feed reader

Running your own reader gives you the most control over your feeds. But do it wrong and it's also the best way to find out how many websites would rather you didn't.

A self-hosted feed reader is one you install and run yourself, whether that's on your own web server or on your laptop, desktop, or phone, even as a web browser extension. It gives you a level of control over your feeds you don't usually get with feed readers that aren't self-hosted.

These are the self-hosted readers we know about, and how you can run them.

Self-hosted readers get blocked more often than others and it's not really their fault. Each reader requests feed content on its own, so when thousands of people use the same reader, a website gets hit with thousands of separate requests for the exact same content, often at the same time.

One well-behaved reader hitting a website for its feed is nothing. It's the pile-up of many of them that does the damage. And while big websites shrug it off, smaller ones with smaller budgets take the hit. But it's nothing new. Feed readers have been hammering websites since feeds were invented. It's practically tradition.

But when a reader isn't self-hosted, it can make smarter decisions about how it requests feeds. If several of its users need updates to the same feed, it only has to ask the website once, then share what it got with everyone else.

Unfortunately, self-hosted readers will keep overwhelming websites, since that's just how they work when a lot of people are using them at the same time. But you can still use one in a way that doesn't make the situation worse.

A reader's "poll frequency" is how often it checks a feed for new content. Setting it too high is one of the best ways to overwhelm the feed's website and one of the fastest ways to get blocked.

Websites usually tell you the frequency they'd prefer, either in their documentation or in the response they send back. Each website will have its own limits, but a good rule of thumb is:

1. No more than one request to a feed per second, per website. For example, if you're subscribed to 4 feeds from the same website, your reader should check them one at a time, with at least a second between each request.
2. Wait at least 15-20 minutes before checking the same feed again for new content.

These are the same constraints we use for the feeds we offer, btw. Websites prefer it, and it keeps us from getting blocked too.

If you're unsure whether the reader you're using adjusts its requests according to what websites ask for, reach out to the developer to confirm. When in doubt, it's safest to just leave the polling frequency at its default value.

A reader's "user agent" is the name it gives a website when it asks for a feed. It's sent every time the reader checks for new content, and again when a feed's images or videos load.

Fortunately, most self-hosted readers already send a proper one. Some let you change it and it's tempting to do so, especially to try to get problematic feeds working. But if you don't have any specific reason to change it, you probably shouldn't. A user agent that's vague, incorrect, or inaccurate is something websites can detect, and they'll likely limit or block the reader for it.

If a particular website needs it to be different from its default value, you should only change it for that site's feeds rather than for every site. And if you do still end up having to change it, it should include the following:

1. The name of the reader
2. The reader's version (so potential issues can be traced properly)
3. A link to the reader's website or a support email address (in case the owner needs to be contacted)

A good user agent would be something like MyRSSReader/1.23.24 (https://example.com/docs).

If you're already using a self-hosted reader and unsure of what the user agent is, or don't know how to change it, you should contact the developer of the reader (or us, we don't mind).

Websites judge how suspicious a feed reader's requests are by the IP address they come from, and your home internet connection is likely the address they trust most.

You can run a self-hosted reader on a rented web server or behind a VPN to use a different IP address. But those are usually the same addresses most bad traffic comes from, so plenty of websites treat them with suspicion by default, and some block them outright.

If you go that route, pick a reputable provider. Just know that if the address it gives you has a history of abuse attached to it, your self-hosted reader will get blocked no matter how carefully it behaves.

When a website gets overwhelmed from too many requests, it starts turning readers away to protect itself, a concept called "rate limiting".

You'll know it when you see it. The feed takes forever to load, or quietly stops updating, or just sits there. Then you open the same feed in your browser and it's fine, new posts and all, making you think something's broken on your end. But the website's just telling your reader "no". And it usually isn't shy about telling your reader exactly why. If it's nice, it'll even say exactly how long to wait before asking again.

When it happens, you might be tempted to change the user agent, switch IP addresses, or even try a different reader, thinking that the site won't notice. But it really just makes you look even more suspicious, and that never ends well. Just wait it out. Blocks like this tend to lift on their own, and the website usually tells you when. If the block doesn't resolve in a reasonable amount of time, you'll likely need to reach out to the website owner to get it fixed.

Self-hosting your reader may give you more control, but it doesn't make you invisible, and it certainly doesn't guarantee that everything you do is 100% secure.

The feeds in a self-hosted reader have to come from somewhere, and getting them means telling websites certain things about yourself. On their own, none of these say much. But collectively, they're enough to pick you out of a crowd.

Most readers you run yourself are open source, which means the code is public and anyone can read it, spot problems, and suggest fixes. But all that openness doesn't mean the reader is actually safe to use, well built, or that it even works correctly.

Open source feed readers are usually worked on by a handful of people in their spare time, sometimes just one person. So it's not unusual for them to have all sorts of bugs that sit there, missed or ignored, for years.

This doesn't mean "don't self-host". It means being open source isn't a guarantee of privacy or that your data is safe, because a reader can still send out information about you, show you feed content without checking it first, or load scripts that let websites track you. So always use a self-hosted reader that's actively maintained, and make sure you install updates when they're available.

Last Updated: 2 days ago