About this article: This article is reviewed against the cited sources. This is source research, not lab validation. Check the stated versions and your own requirements before making changes.
A 421 4.7.28 reply from Gmail is a temporary rate limit. Gmail has decided that the volume or pattern of mail from one of your identifiers (the sending IP, its netblock, a DKIM domain, an SPF domain, a domain in a link, or a repeated Message-ID) is unusual, and it will not accept more for now. The 421 means "try later", so your mail server should queue and retry.
What the error means
Google's SMTP error reference lists the 4.7.28 messages. Each begins 421 4.7.28 Gmail has detected and differs in what it blames. The wordings below are Google's, each cut where Google's text goes on to point at its bulk sender guidelines (the Message-ID one points to "Why has Gmail blocked my messages?"):
| Message (abridged by scope) | Scope that is limited |
|---|---|
421 4.7.28 Gmail has detected an unusual rate of email. To protect our users from spam, email has been temporarily rate limited. |
Unspecified |
421 4.7.28 Gmail has detected an unusual rate of email originating from your IP address ip-address. To protect our users from spam, email sent from your IP address has been temporarily rate limited. |
Sending IP |
421 4.7.28 Gmail has detected an unusual rate of unsolicited email originating from your IP Netblock ip-address. To protect our users from spam, email sent from your IP Netblock has been temporarily rate limited. |
The network block around your IP |
421 4.7.28 Gmail has detected an unusual rate of unsolicited email originating from your DKIM domain domain-name. To protect our users from spam, email sent from your domain has been temporarily rate limited. |
DKIM signing domain |
421 4.7.28 Gmail has detected an unusual rate of unsolicited email originating from your SPF domain domain-name. To protect our users from spam, email sent from your domain has been temporarily rate limited. |
SPF domain |
421 4.7.28 Gmail has detected an unusual rate of unsolicited email containing one of your URL domains. To protect our users from spam, email with the URL has been temporarily rate limited. |
A domain that appears in links in the message |
421 4.7.28 Gmail has detected an unusual amount of unsolicited email originating from your IP address. To protect our users from spam, email sent from your IP address has been temporarily blocked. |
Sending IP, described as "temporarily blocked" |
421 4.7.28 Gmail has detected this sender exceeded the quota for sending messages that have the same Message-ID:. To best protect our users, the message has been temporarily rejected. |
Reuse of one Message-ID |
Google's sender guidelines explain the model: "Gmail tracks volume, feedback, and limits per domain and IP address: DKIM and SPF quotas are specific to your domain. If you hit the limit, other domains that send from the same IP may not be affected. The IP address quota is shared for all senders that use that IP address. When the IP address hits its quota, all domains that send from that IP address stop sending emails."
That last sentence matters for hosted servers: if the IP is shared with other senders, you can be limited by someone else's volume.
Likely causes
| Cause | What you see | How to confirm |
|---|---|---|
| A burst, such as a campaign, a retry storm after an outage, or a queue flushed in one go | Limit appears soon after a large release; earlier mail was fine | Count Gmail-bound deliveries per hour in your mail log, and compare with the previous week |
| Too many parallel connections to Gmail | Limit appears even at moderate volume | Your mail server's per-destination concurrency setting |
| A new IP or new domain with no sending history | Limits on the first days of use | Age of the IP and domain; Google's advice to "start with a low sending volume" |
| Other senders on a shared IP, or a neighbouring range | Message names the IP or the netblock; your own volume is low | Ask the provider who else sends from it; check the IP in Postmaster Tools |
| Recipients marking mail as spam, or unwanted mail | Spam rate shown in Postmaster Tools is high; the DKIM or SPF domain is named | Postmaster Tools spam rate against Google's 0.10% and 0.30% thresholds |
| A website form, script or account sending spam | Unexpected message volume to unrelated recipients | Mail log and queue contents; see "554 5.7.1 Client host blocked using zen.spamhaus.org: which list, and how to get off it" for the same signs |
| One Message-ID reused across many messages | The Message-ID variant of the error | Compare headers of several queued messages |
Prove the cause before changing anything
All of these only read.
1. Read the exact reply, and which identifier it names. In Postfix the reply text appears in parentheses after status=deferred in the mail log, and in postqueue -p output. The table above tells you what is limited. If the message "does not specify if you exceeded your DKIM, SPF, or IP quota, assume that all three are affected."
2. Look at the queue by destination. qshape deferred (run as the superuser) groups the Postfix queue by domain and age. Most of the count at gmail.com and Google-hosted domains confirms the limit is specific to Google. "Postfix mail stuck in the queue as deferred: read the reason" covers these commands in detail.
3. Check Postmaster Tools. Google's setup page says Postmaster Tools "dashboards have detailed information about spam rate, reputation, message authentication, and delivery errors", and that the data "only applies to messages sent to personal Gmail accounts" (addresses ending @gmail.com or @googlemail.com). Add the sending domain, using the DKIM d= domain or the SPF (Return-Path) domain, and verify it. Google's guidelines say: "Keep spam rates reported in Postmaster Tools below 0.10% and avoid ever reaching a spam rate of 0.30% or higher."
4. Confirm the basics Google requires of all senders. The guidelines require "SPF or DKIM email authentication", "valid forward and reverse DNS records", a TLS connection, and spam rates below 0.30%; senders of more than 5,000 messages a day to Gmail accounts have further requirements. Authentication faults produce different Gmail codes (4.7.26, 4.7.27, 4.7.29 and 4.7.30 are in the same reference), and "Application Email Rejected or Junked After a Server Move: SPF, DKIM, DMARC and PTR" covers them.
Example (illustrative): app01 (203.0.113.25) sends invoice notices through Postfix. After a billing run releases 40,000 messages at once, postqueue -p shows many entries deferred with a reply starting 421 4.7.28 Gmail has detected an unusual rate of email originating from your IP address. qshape deferred shows nearly all of the count at gmail.com. In Postmaster Tools the domain's spam rate is 0.04%, below the 0.10% guideline, and SPF and DKIM pass. The cause is the burst, not reputation or content, so pacing the queue is the fix.
Fix
Pause and restart as Google describes
Google's steps:
- "Do not send email for at least 10 minutes."
- "Identify the cause."
- "After 10 minutes, send emails from a single connection. If the single connection fails, wait another 10 minutes. If successful, increase the number of connections one at a time until your desired number of connections are established."
Disruptive step: pausing delivery to Gmail delays that mail until you resume it, and it stays queued meanwhile. In Postfix, postsuper -h QUEUEID holds only the named message (a message on hold is not retried, and does not expire while on hold) and postsuper -H QUEUEID releases it. The documents read give no one-command hold for a whole destination, so hold the Gmail-bound queue IDs you have listed. Do not delete the messages. A hold is reversible; deletion is not.
Pace the queue
Google recommends: "Send email at a consistent rate. Avoid sending email in bursts. Start with a low sending volume to engaged users, and slowly increase the volume over time." It adds: "Avoid introducing sudden volume spikes if you do not have a history of sending large volumes. For example, immediately doubling previously sent volumes suddenly could result in rate limiting or reputation drops." And: "If messages start bouncing or start being deferred, reduce the sending volume until the SMTP error rate decreases. Then, increase slowly again."
In Postfix, default_destination_rate_delay (default 0s) inserts a delay between deliveries to the same destination, and transport_destination_rate_delay overrides it for one transport. The Postfix manual's note says that with a non-zero delay you should "specify a transport_destination_concurrency_failed_cohort_limit of 10 or more to prevent Postfix from deferring all mail for the same destination after only one connection or handshake error." The transport(5) manual shows how to send one destination through a separate transport, with the line example.com slow:, and says the slow transport "could be configured to run at most one delivery process at a time". Define the transport in master.cf and its limits in main.cf following the Postfix documentation; this is a configuration change, so test it on a quiet system first.
Save first: before editing, copy both files (for example cp /etc/postfix/main.cf /etc/postfix/main.cf.bak-20261119 and the same for master.cf), and print the current value of each parameter you will change with postconf default_destination_rate_delay. After editing, read the result back with postconf -n. Then run postfix check; postfix(1) describes it as warning "about bad directory/file ownership or permissions" and creating missing directories, so it is not a full test of your new lines.
Disruptive step: changing queue manager settings applies on postfix reload ("Re-read configuration files"), and the manual notes that the rate-delay timer state "does not survive postfix reload or postfix stop". To back out, copy the saved main.cf and master.cf back, run postfix check, and reload again. After the reload, send one test message to a Gmail address and read the mail log for its status=.
Fix the underlying cause
- Reputation or spam rate. Google's wording: "Make sure you send only to people who subscribe to your emails, and give recipients an option to unsubscribe." Clean the list.
- Shared IP. Google advises making sure the shared IP address "isn’t on any internet blocklist". See "554 5.7.1 Client host blocked using zen.spamhaus.org: which list, and how to get off it" for how to check an IP against a major blocklist.
- Message-ID reuse. Make the sending application generate a unique Message-ID for every message.
- A new IP or domain. Ramp up. Google's sender guidelines say to avoid sudden spikes if you have no history of large volumes. Do not move volume between IPs to evade the limit.
- Compromise. If the volume or recipients are not yours, treat it as a security incident: lock the account or form, then clear the queue deliberately.
Google's guidelines state that "Google and Gmail don’t accept allowlist requests from email providers", so the route is to bring volume and reputation inside the limits.
Confirm the fix
- After the pause, a single connection delivers to Gmail with
status=sentin the log, with no new 4.7.28. - As you add connections, the error does not return. If it does, drop back one step and wait another 10 minutes, as Google instructs.
qshape deferredshows the Gmail row emptying over several retry cycles. Postfix retries deferred mail on its own schedule (queue_run_delaydefault 300s), so allow time before flushing.- Postmaster Tools shows the spam rate under 0.10% over the next days. The guidelines add: "It can take time for improvements in spam rate to reflect positively on spam classification."
What to watch afterwards
- Keep a daily record of Gmail-bound volume and deferrals, and alert when deferrals start.
- Schedule bulk sends in steady batches, and plan a ramp-up for any new IP.
Read next
Related articles you may want next.
- HostingApplication Email Rejected or Junked After a Server Move: SPF, DKIM, DMARC and PTRApplication mail bouncing or junked after a migration? Read Authentication-Results, then fix SPF, DKIM, DMARC alignment, PTR and Microsoft 365 relay settings for the new path.Read article
- TroubleshootingPostfix mail stuck in the queue as deferred: read the reasonPostfix mail stuck as deferred? Read the reason in postqueue -p and the mail log, group by destination with qshape, fix the cause, then flush once.Read article
- Troubleshooting554 5.7.1 Client host blocked using zen.spamhaus.org: which list, and how to get off itMail rejected: Client host blocked using zen.spamhaus.org? Check the ZEN return code, work out SBL, CSS, XBL or PBL, and take the right removal route.Read article
Risks and limits
- These commands have not been run or lab-tested. The Postfix commands come from its documentation, and the Gmail behaviour from Google's pages.
- Google does not publish numeric sending limits on the pages read. The ten-minute pause and the single-connection restart are Google's own guidance, and the spam-rate thresholds are Google's stated guidelines, not guarantees.
- The Postfix per-destination slowdown needs a separate transport and changes to
master.cf; the documents read show the parts but not one complete recipe, so the exact configuration is yours to assemble and test. - Postmaster Tools covers only mail to personal Gmail addresses. Google Workspace recipients are not in its data.
Sources
- Google Workspace Admin Help: Gmail SMTP errors and codes (the former support.google.com/a/answer/3726730 address redirects here) — Google, last updated 2026-11-10 UTC, accessed 18 November 2026
- Google Workspace Admin Help: Email sender guidelines — Google, no date shown, accessed 18 November 2026
- Google Workspace Admin Help: Set up Postmaster Tools — Google, no date shown, accessed 18 November 2026
- Postfix: postconf(5), configuration parameters — Postfix documentation, no date shown, accessed 18 November 2026
- Postfix: transport(5) — Postfix documentation, no date shown, accessed 18 November 2026
- Postfix: postsuper(1), postqueue(1), qshape(1) and postfix(1) and postconf(1) — Postfix documentation, no date shown, accessed 18 November 2026
This article was originally published on .