Our website is in Dutch; support is available in English.

Reading bounce messages: from error to cause

Two real bounces worked through, and why forwarding is more often the cause than your settings.

Updated 11 minute read

Contents of this article
  1. Start with the first digit, not the text
  2. Two bounces we see regularly
  3. Forwarding is more often the cause than your settings
  4. Why you are suddenly seeing this
  5. When a Dutch or Belgian provider rejects your mail
  6. Blocklists: which ones matter and which don't
  7. The order to tackle it in
  8. What we can and can't do here
  9. Further reading

A message comes back with an error and a code. Using two real bounces from our own mail log, this page shows how to get from that code to the actual cause, and why that cause is more often the forwarding than your settings.

Start with the first digit, not the text

The sentence in a bounce message is almost always the least useful part of it. What you need to know first is whether the message is still on its way or has been given up for good, and that is in a single digit. A code starting with 4 is temporary: your mail server holds on to the message and offers it again by itself, for days. A code starting with 5 is permanent.

That distinction decides whether you need to act. With a 4, the right thing to do is usually nothing. With a 5, something has to change, because sending it again unchanged gets the same rejection. What each individual code means is in our bounce explainer (in Dutch); below we look at the two cases it most often comes down to in practice.

Two bounces we see regularly

Gmail: "receiving mail at a rate that prevents additional messages"

This one arrives as a permanent bounce, with dsn=5.2.1 and status=bounced:

Code
550-5.2.1 The user you are trying to contact is receiving mail at a rate
550-5.2.1 that prevents additional messages from being delivered. For more
550-5.2.1 information, go to
550 5.2.1  https://support.google.com/mail/?p=ReceivingRatePerm

This is not about you. It is a limit on the receiving side: that one mailbox is getting so much mail right now that Gmail won't accept any more. Your SPF, your DKIM, your IP address and your reputation have nothing to do with it, and there is no setting that fixes it. Google also does not publish the threshold at which this happens. If you come across a specific number of messages per hour somewhere, it did not come from Google.

The awkward part is that the code starts with a 5 while the cause is temporary. So your server gives up, even though the same message would have been delivered an hour later. If you see this in waves for one recipient, consider that the person may be being flooded: someone signed up to hundreds of newsletters at once to bury a real notification produces exactly this picture.

Yahoo: "temporarily deferred due to unexpected volume or user complaints"

Same day, a different recipient, and a very different kind of message:

Code
421 4.7.0 [TSS04] Messages from 192.0.2.10 temporarily deferred due to
unexpected volume or user complaints; see
https://postmaster.yahooinc.com/error-codes

Here the code starts with a 4, so the message is still in the queue. By saying "volume or user complaints", Yahoo deliberately doesn't tell you which of the two it is: it is a single reputation signal, not separate counters. The tag in square brackets refers to an internal rule, and Yahoo does not publish what any individual number means; the same sentence also comes with [TS01]. Anyone who gives you an exact definition of TSS04 did not get it from Yahoo.

A small practical note: the URL in the message is out of date. Yahoo's postmaster site is now called Sender Hub, and postmaster.yahooinc.com redirects you to senders.yahooinc.com.

Forwarding is more often the cause than your settings

Both bounces above involved forwarded mail, and that is no coincidence. Forwarding email (an address on your own domain that passes everything on to a Gmail or Outlook address) is the most underestimated cause of delivery problems. Three things go wrong at once.

SPF no longer matches, and the fix costs you DMARC

Your forwarding server is not in the original sender's SPF record, so SPF fails. The usual fix is called SRS: your server rewrites the envelope address to your own domain, so SPF passes again. That works, but at a price. After the rewrite, SPF aligns with your domain rather than the domain in the visible From address, and DMARC requires exactly that alignment. For forwarded mail, DMARC therefore depends entirely on DKIM.

Worth knowing: SRS never became a standard. The 2003 proposal expired and the term does not appear in the SPF standard itself. What the standard does say is that an intermediary that rewrites the sender address must make sure SPF passes afterwards.

DKIM survives forwarding, but not if the message is touched

DKIM is the only check that can survive forwarding, because the signature is on the message itself. But it breaks as soon as the message changes along the way, and that happens more often than you might think. Adding a footer, putting [SPAM] in front of the subject, rewriting links for click protection, or a virus scanner re-encoding the content: any one of those four is enough.

A persistent misunderstanding is that DKIM's relaxed mode handles this. It only tolerates differences in whitespace and in the letter case of header names; added text is not covered. A disclaimer below the message breaks the signature in relaxed mode too.

The complaints land on you

This is the most underestimated point. If the recipient marks forwarded junk as spam, that complaint counts against your server, because it is the party that delivered the message. Google says so in as many words: if forwarded messages from your domain are marked as spam, future messages from your domain are more likely to be marked as spam too. SRS even makes this worse, because the envelope address then points squarely at you.

The remedy Google recommends is simple and rarely followed: filter spam before you forward, not after. And consider whether forwarding is the right solution at all. Let the receiving side fetch the mail over IMAP or POP, and there is no forwarding server left to inherit the reputation and the bounces.

Why you are suddenly seeing this

Many administrators have recently started getting bounces on a setup that worked for years. That is not because the requirements got stricter, but because enforcement did. Google's and Yahoo's requirements have been fixed since early 2024: SPF or DKIM for everyone, and for bulk senders both, plus a DMARC record, alignment and a one-click unsubscribe link. What changed afterwards is the consequence of not complying: such mail used to land in the spam folder, now it is rejected during the SMTP conversation. Google announced that tightening for November 2025.

Microsoft joined in May 2025 with the same requirement for outlook.com, hotmail.com and live.com: from five thousand messages per day from one domain, SPF, DKIM and DMARC all have to be in order, and p=none is enough. Microsoft first puts non-compliant mail in the junk folder and later rejects it with code 5.7.515. Microsoft has not announced when rejection becomes the default.

Two things that are often repeated incorrectly. The threshold of five thousand messages per day belongs to Google and Microsoft; Yahoo deliberately uses no number. And Google's requirements only apply to personal Gmail addresses, not to business domains running on Google Workspace. That explains why your mail reaches a business customer but not their private address.

When a Dutch or Belgian provider rejects your mail

This is where it gets uncomfortable, and it is in almost no article on the subject: the big Dutch and Belgian providers have no postmaster page. Where Google, Microsoft and Yahoo all offer an environment in which you can see how they view you, KPN and Ziggo have nothing of the kind. You report a block by email to their abuse address, and that's it.

The messages are just as bare. KPN rejects with a plain 550 5.7.1 Mail rejected, with no reason and no reference, so you don't even know what the problem is. Ziggo's own customer service page for blocked senders advises a virus scan, changing passwords and waiting 24 hours. Of the Benelux providers, only Proximus has a self-service form to get a block lifted.

For comparison: GMX and WEB.DE include a unique link in every rejection, to a page that explains exactly why this message was rejected, and Orange publishes a list of about twenty of its own error codes, including solutions. That is not a detail. It decides whether you can solve a problem or only wait, and it is one of the reasons we look at the log ourselves instead of sending you to a form.

Blocklists: which ones matter and which don't

If the rejection refers to a blocklist, that is your lead. Three lists matter in practice. Spamhaus is the most important one and is queried directly by Yahoo, among others; you check and request removal at check.spamhaus.org. Barracuda has its own removal form, which requires a real explanation. SpamCop resolves itself: listings expire automatically once the reports stop.

Removal is free with all three. That is worth mentioning, because there is a market of parties charging you for a delisting you can request yourself at no cost.

Two things not to do. Don't panic about a listing on UCEPROTECT: at level two that list blocks an entire network block, and at level three an entire network, for the behaviour of one other party; it offers immediate removal for a fee; and Google, Microsoft and Yahoo don't query it. A listing there expires by itself once the actual abuse stops. And don't respond to a reference to SORBS: that service shut down in June 2024 and the domain no longer exists, so a server that still rejects on it is itself misconfigured. There is nothing to get removed.

One last trap that is often mistaken for a blocklist: a message like "blocked using zen.spamhaus.org; Error: open resolver" does not mean you are listed. Spamhaus refuses queries that come in through public DNS services, and the receiving server wrongly treats that error response as a listing. That is a problem on the receiving side.

The order to tackle it in

  1. Look at the first digit. If the code starts with a 4, do nothing and let your server do its job.
  2. Check whether the mail was forwarded. If there is an orig_to that differs, or an address starting with SRS0=, look for the cause there and not in your DNS.
  3. Read which party rejected the message and look the code up with that party. The same code can mean something different at Google than in the standard.
  4. If it is about authentication, start with DKIM. It survives forwarding; SPF does not.
  5. Then check your SPF record, especially the number of lookups: above ten, nothing gets a pass anymore, not even your own mail server.
  6. Only then look at blocklists and reputation. They are almost never the first cause, but they are the slowest to fix.

For step 3, use the bounce explainer; for step 5, the SPF checker. Both are free to use without an account (both in Dutch).

What we can and can't do here

We read these messages every day, for our own domains and our customers'. If you are a customer, we go through the mail log for your domain with you, find out where a series of bounces comes from, set up SPF, DKIM and DMARC, and request removal from a blocklist that lists our own addresses. That is simply part of managing your mail, not a separate service.

What we can't do, and what nobody can: get a limit at Google or Microsoft raised, force a provider to lift a block, or guarantee that mail reaches a specific recipient. Anyone who promises you that is selling you something they don't control.

We also don't have a bounce dashboard or blocklist monitoring in the control panel. If you see such a feature advertised somewhere, with us it is human work based on the log, and for most businesses that is exactly what helps.

Further reading

Was this article helpful?

Still stuck?

Send us a message. We will help you, usually within one working day.