451 4.4.1 connection timed out: when to wait, when to check DNS
A 451 4.4.1 means your server could not get an answer from the recipient's mail host in time. Wait if it is brief; otherwise check DNS, port 25 and MX.

On this page(7 sections)
A 451 4.4.1 reply means the connection to the recipient's mail server did not complete: nobody answered, or the answer arrived too late. It is a temporary failure, so your server should queue the message and try again. Whether you should do anything depends on how many domains are affected and for how long.
This post separates the cases where waiting is correct from the cases where the fault is in DNS, in your network or in the recipient's MX records.
What does 451 4.4.1 actually mean?
It means a deferral at the network level. The two numbers carry different information, and the SMTP reply code and the enhanced status code come from different RFCs.
| Part | Meaning | Source |
|---|---|---|
451 |
Requested action aborted: local error in processing. A 4xx class reply, so try again later. | RFC 5321 section 4.2 |
4.4.1 |
Class 4 (persistent transient failure), subject 4 (network and routing), detail 1: no answer from host. | RFC 3463 |
RFC 3463 describes 4.4.1 as an outbound connection attempt that was not answered, either because the remote system was busy or was unable to take a call. Servers phrase it differently: you may see "Connection timed out", "read error" or "no answer from host" next to the same code. Some servers use 421 or 450 with the same 4.4.1 detail. The meaning is the same.
For the full class.subject.detail structure, see SMTP enhanced status codes decoded.
Is this error a bounce I need to act on?
Usually not yet. A 4xx is a deferral, not a bounce, and your sending system is expected to keep the message and retry. RFC 5321 section 4.5.4.1 says retries should not be more frequent than every 30 minutes and that a sender should give up only after at least 4 to 5 days.
So the first question is not "how do I fix this" but "has it been deferred once, or for hours?" A single 4.4.1 in a log that later shows a successful delivery is normal Internet weather. Treat it as a problem when the same message, or the same recipient domain, keeps failing across several retry attempts. How to design the retries is covered in retry policies for outbound mail, and how classification should treat 4xx replies is in hard vs soft bounces.
When should you just wait?
Wait when all of these are true:
- The failures are for one recipient domain, not all of your mail.
- Mail to other domains from the same server is flowing.
- The first deferral is less than a few hours old.
- Your queue shows the message still has retries left.
In that case the likely causes are on the receiving side: a busy or overloaded server, a short outage, or throttling of new connections. Greylisting can also look like a refusal at connection time on some systems, which greylisting and 4xx deferrals explains. Do not resend manually. A manual resend creates a second queued copy and, if the first one also succeeds, a duplicate for the recipient.
When is the problem on your side?
Suspect your own network when many unrelated domains defer with 4.4.1 at the same time. That pattern points at the sending host, not at the recipients.
| Symptom | Likely cause | What to check |
|---|---|---|
| Every domain times out | Outbound port 25 is blocked | Hosting provider policy, firewall egress rules |
| Only large providers time out | Your IP is being slowed or dropped | Blocklist status, reverse DNS, reputation |
| Timeouts began after a network change | New firewall, NAT or MTU problem | Firewall SMTP inspection, packet loss, MTU |
| DNS lookups fail intermittently | Resolver trouble on your host | dig from the mail server itself |
Test from the server that sends the mail, not from your laptop. Many cloud hosts block outbound port 25 by default, and a laptop on a different network will not show it:
# Can this host open a TCP connection to the recipient's MX?
nc -vz -w 10 mx.example.net 25
# Where does the path stop?
mtr -rwn -c 20 mx.example.net
# Does the resolver on this host return the MX at all?
dig +short MX example.net
If nc times out for every destination, the issue is egress filtering or your provider, and no amount of waiting will fix it. Some firewalls that "inspect" SMTP content also slow the session until it times out, so turning that inspection off for port 25 is a standard test.
When is it the recipient's DNS or MX records?
Check the recipient domain when one domain fails for a long time while everything else works. Look up its MX records and confirm that each host has an address and answers on port 25:
- Run
dig +short MX recipient-domain.com. If nothing returns, the domain has no MX and your server falls back to the A record, as described in MX records explained. - Run
dig +short A mx-host.recipient-domain.comfor each MX host. An MX that points to a name with no address cannot be reached. - Connect to each MX on port 25 with
ncoropenssl s_client -starttls smtp -connect host:25. - Check whether the MX points at an old provider that has been shut down. A domain that moved email providers but left a stale MX with a low preference number produces exactly this timeout.
Mail servers try MX hosts in preference order, lowest number first, so a dead high-priority host can cause repeated timeouts before a working one is tried. If you do not control the recipient domain, tell the sender or the recipient's administrator which MX host does not answer. That is the actionable detail.
What should the sender tell the user?
If you run an app that sends mail, say what is true: the message has not been delivered yet, the system is still trying, and the next attempt is scheduled. Do not report a 4.4.1 as a failure to the end user until the retry window is exhausted. The final outcome, after the last retry, is what belongs in your delivery status.
With the Koltrix SMTP relay and the POST https://api.koltrix.com/api/v2/emails API, delivery events can be sent to your systems through webhooks, so you can show "still trying" separately from "failed". Details are in the Koltrix docs.
Bottom line
- 451 4.4.1 is a temporary network failure: no answer from the recipient's mail host.
- One domain, brief, still retrying: wait, and do not resend by hand.
- Many domains at once: check outbound port 25, firewall inspection, DNS and your IP's reputation from the sending server.
- One domain for days: check its MX records and each MX host's address and port 25 response.
- The final result after the last retry is the one to report to users.
Start with Koltrix
Your domain, one inbox, and an API that sends.
A team inbox where AI sorts and drafts (nothing is sent without your click), plus the transactional API and SMTP relay your product sends with. 7 days free, no card.

