max-smtp-out sets the maximum number of simultaneous outbound SMTP connections PowerMTA maintains to a destination domain (or MX host group, with mx-rollup). This is not a throughput setting — it is a concurrency limit. Higher max-smtp-out means more parallel connections, which allows more messages to be in delivery simultaneously. But ISPs enforce connection limits based on your sender reputation, and exceeding those limits produces 421 deferral responses that reduce effective throughput.
Section| ISP | Safe max-smtp-out (HIGH reputation) | Safe max-smtp-out (MEDIUM) | Safe max-smtp-out (Warming) | Notes |
|---|---|---|---|---|
| Gmail | 6–10 | 3–5 | 1–3 | 421 4.7.0 = reputation-based limit exceeded |
| Outlook.com | 4–6 | 2–4 | 1–2 | RP-002 = connection rate exceeded |
| Yahoo | 5–8 | 3–5 | 1–3 | TSS04 = throttle, reduce connections |
| GMX/Web.de | 2–4 | 1–2 | 1 | Very conservative — German ISPs strict |
| Orange.fr | 2–3 | 1–2 | 1 | Low tolerance for new/unknown senders |
| Generic (unknown) | 2–4 | 1–2 | 1 | Default: conservative until reputation established |
# max-smtp-out configuration per ISP reputation tier # HIGH reputation pool:Sectionvirtual-mta-pool high-rep-pool max-smtp-out 8 mx-rollup gmail.com # MEDIUM reputation — recovering from incident:virtual-mta-pool recovering-pool max-smtp-out 4 # Reduced until reputation recovers retry-after 20m mx-rollup gmail.com # Warming pool — brand new IPs:virtual-mta-pool warming-pool max-smtp-out 2 # Very low — no reputation history max-msg-rate 50/h # ~1,200/day in week 1 retry-after 30m mx-rollup gmail.com
max-smtp-out controls steady-state concurrency; max-conn-rate controls how fast new connections open. Both are needed. Setting max-smtp-out=10 without a max-conn-rate limit can cause PowerMTA to open all 10 connections simultaneously, which looks like a connection storm to the receiving ISP. Setting max-conn-rate=1/s with max-smtp-out=10 means connections open at 1 per second, reaching steady state of 10 over 10 seconds — a much healthier pattern.
# Combined configuration for controlled connection behaviorSectionmax-smtp-out 8 # Max concurrent connections max-conn-rate 1/s # Open 1 new connection per second max-msg-rate 300/h # Messages per hour (optional cap) max-msg-per-conn 100 # Messages per connection before reconnect # Effective throughput calculation: # 8 connections × 100 msgs/connection × reconnect every ~60s # = 8 × 100 / 60 ≈ 13 msgs/second = ~47,000 msgs/hour per pool # With 10 IPs in pool: ~470,000 msgs/hour to Gmail
# Signs that max-smtp-out is too high for current reputation:
# 1. 421 4.7.0 (Gmail) or 421 RP-002 (Microsoft) in dsnDiag
grep "421 4.7.0\|421 RP-002" /var/log/pmta/accounting.csv | \
awk -F, '{count[$7]++} END {for(d in count) print count[d], d}' | sort -rn | head -5
# 2. High deferral rate that increases when volume increases
# (indicates hitting ISP connection limits proportionally to volume)
# Response: reduce max-smtp-out by 30-50%
# Verify: deferral rate should decrease within 1-2 hours
# Resume: increase max-smtp-out gradually (10-15% per week) as reputation improves
# Monitor connection count in real-time:
pmta show vmta gmail-ip-1 | grep "connections"
# Compare to your max-smtp-out setting
The correct max-smtp-out value depends on your current Postmaster Tools reputation tier, not on a fixed "best practice" number. HIGH reputation at Gmail: 6-10. MEDIUM: 3-5. LOW: 1-2. New/warming IPs: 1-3. Exceeding ISP-appropriate limits at your current reputation tier produces 421 4.7.0 responses that compound reputation damage. Increase max-smtp-out incrementally: add 1-2 connections per week as reputation remains stable, not in large jumps.
Monitor this configuration area through the PowerMTA accounting log's dsnDiag field. Filter accounting records for the specific ISP domains affected by this configuration and group dsnDiag responses by first 60 characters to identify the dominant error patterns. A deferral rate above 5% at any single ISP warrants investigation; above 15% requires immediate volume reduction and configuration review.
The dlvSourceIp field in the accounting log enables per-IP analysis within this configuration context. Comparing per-IP deferral rates identifies whether a configuration issue affects all IPs in a pool uniformly (configuration problem) or just specific IPs (reputation or IP-specific problem). This distinction determines the correct remediation path.
The parameter values documented in this reference are appropriate for established, warmed IPs with HIGH reputation at the target ISP. New or warming IPs, and IPs with MEDIUM or LOW reputation, require more conservative values. Move up incrementally as reputation signals confirm the infrastructure can sustain additional throughput. Review ISP-specific configuration monthly — Postmaster Tools reputation tier changes and SNDS status changes are the primary triggers.
Implementing this PowerMTA configuration correctly in production requires testing the specific parameter values against your actual IP reputation history, ISP distribution, and sending volume. The values documented here represent proven starting points, not fixed constants — your optimal configuration may differ based on your infrastructure's operational history.
After applying any configuration change, monitor the accounting log for the first 2-4 hours to verify the change produced the expected effect on deferral rates. A configuration change that was expected to reduce deferrals but shows no change (or increased deferrals) indicates either: the change addressed the wrong variable, or there is a confounding factor that needs investigation before continuing.
The Cloud Server for Email infrastructure team manages PowerMTA environments daily, applying the configuration principles documented in this reference series across clients with varied volume levels, ISP distributions, and reputation histories. Contact us at infrastructure@cloudserverforemail.com for a technical assessment of your specific PowerMTA configuration requirements.
This PowerMTA reference is part of the Cloud Server for Email technical documentation series covering production configurations and operational procedures from managed infrastructure environments. Configuration values are production-validated starting points; optimal settings depend on your IP reputation tier, ISP distribution, and sending volume. Browse the complete PowerMTA reference series, the MailWizz technical FAQ, and over 130 engineering notes.
For infrastructure-specific guidance — IP reputation analysis, configuration audit, or managed PowerMTA deployment — contact the Cloud Server for Email team at infrastructure@cloudserverforemail.com or +372 602-7190. Technical assessments are conducted at no obligation and produce environment-specific configuration recommendations. The Cloud Server for Email infrastructure team manages PowerMTA environments daily, applying the configuration principles documented in this reference series across clients with varied volume levels, ISP distributions, and reputation histories. Each managed environment receives monthly configuration review, daily monitoring, and incident response as part of the service. Contact us to discuss your specific PowerMTA requirements and receive an assessment of your current configuration against production best practices.
Related PowerMTA configuration topics in this series: PowerMTA Technical FAQ (53 configuration articles covering every aspect of production PowerMTA operations), Gmail delivery configuration, Outlook/Hotmail delivery, DKIM key management, accounting log format, and virtual MTA pool configuration. The operational notes series provides complementary engineering perspective on the production patterns that emerge from running these configurations at scale, including ISP-specific behavior, reputation management principles, and infrastructure architecture decisions that inform configuration choices. Organizations running PowerMTA in production benefit from both the configuration reference (this article) and the operational notes (production context). Cloud Server for Email clients receive both: configuration reference documentation as part of onboarding, and the operational intelligence that comes from managing PowerMTA environments across dozens of production deployments.
Cloud Server for Email operates fully managed PowerMTA infrastructure from EU-based dedicated servers. Daily monitoring, per-ISP domain block optimization, IP warming management, and incident response included.