SMTP error 421
Service not available, closing channel.
Updated Jul 1, 2026
SMTP 421 is a temporary "Service not available, closing transmission channel" reply. The server accepted your connection but is ending the session — usually too many concurrent connections, rate limits, or maintenance. It is a 4xx temporary failure, so queue and retry with backoff; do not treat it as a permanent bounce.
Back off and retry later; if it is constant, check you are not exceeding connection or sending limits.
What it means
421 is a transient reply the server can send to any command when it needs to shut the connection down. Because it is a 4xx code the message is not permanently rejected — RFC 5321 is explicit that a client should requeue and try again later. What sets 421 apart from a per-recipient deferral like 450/451 is that it tears down the whole channel, so every message in that session is affected. An enhanced code often tells you why: 4.3.2 (server overloaded / not accepting), 4.7.0 (Gmail reuses this for rate-limiting or reputation throttling), 4.4.5 (congestion).
Common causes
Amazon SES returns "421 Too many concurrent SMTP connections" when you exceed its open-connection limit — the commonest cause on bulk sending.
Load-balanced endpoints like SES recycle instances, so a connection held too long can be dropped mid-session.
Gmail/Workspace return "421 4.7.0 ... Try again later" when they see an unusual rate or low sender reputation.
The server is busy or shutting down and refuses to continue the conversation.
How to fix it
- Retry with backoff — do not hard-bounce
Requeue and try again on a growing delay (for example 5s → 10s → 30s → 20 min). Confirm your client is not treating 421 as permanent.
- Cap concurrent connections
If the text says "too many connections", reduce parallel SMTP connections to stay under the provider's ceiling.
- Recycle long-lived sessions
Close and reopen the connection after a set number of messages or a set age, so a stale one is not dropped mid-send.
- For Gmail 4.7.0 rate-limiting
Pause that lane 10+ minutes, resume with a single connection, and add concurrency back slowly. Fix the root cause — volume spikes and reputation — with SPF/DKIM/DMARC and Postmaster Tools.
- Check provider quotas
Log into your relay (SES, SendGrid, Mailgun, etc.) and confirm you have not hit a rate or daily-sending limit surfacing as a 421.
Paired code
A real bounce often shows both a 3-digit code and an enhanced code together. This one commonly pairs with:
Common questions
Is 421 permanent, or should I retry?
It is transient. The client should requeue and retry with a sensible backoff. Treating 421 as a hard bounce is a misconfiguration.
What does "421 4.7.0 Try again later" from Gmail mean?
Google is temporarily rate-limiting your IP, usually from a sending spike or weak reputation. Slow down, send on one connection, and work on authentication and reputation.
Why does SES return 421 after many emails on one connection?
Its endpoints sit behind a load balancer that recycles instances, so a long-held connection can be dropped. Recycle connections after a set count or age.