SMTP and Mail Flow
Delivery paths, relays, and smarthosts: diagnosing SMTP connections, transport encryption, and sender checks in the mail flow.
NewHow Long Does an SMTP Session Stay Open? ConnectionTimeout 00:10:00 in Exchange and the Systems for Which That Is Too Short
Exchange terminates every incoming SMTP session after ten minutes, even if it is currently transferring data. Which senders stay on a connection that long, how to determine the actual session duration from the protocol log, and when ConnectionTimeout and ConnectionInactivityTimeout should be adjusted on a relay connector.
NDR, DSN, Bounce: How to correctly distinguish non-delivery notifications
NDR, DSN, bounce, reject, backscatter: The terms surrounding failed delivery are often used interchangeably, but they refer to different things. What the RFCs define, who generates which notification, how a DSN is structured, and why the distinction between reject and bounce determines backscatter.
Sending mail through a relay: Checking TLS and authentication
A one-pager for Application Managers whose applications send mail through a relay: Which three application settings matter (port, TLS mode, authentication), what the options are called in common environments, and how a single test email’s Received header proves that the connection is actually encrypted and authenticated.
SMTP Load Testing with Numbered Recipients: Send Every Email Traceably
A load test is only as good as its evaluation. With the -N option, smtp-source numbers every email via the recipient address without sacrificing throughput. Learn how to structure the run, how many sessions make sense, and how to automatically find missing numbers.
smtp-source without installing Postfix: Extract load-testing tools from the RPM
smtp-source and smtp-sink are part of Postfix, but they also run without an installed mail server. Learn how to extract the two tools from the package on RHEL, why running them from /tmp can fail because of the noexec mount option, and which libraries need to be included.
Analyze Email Headers Without Uploading the Email: Locally in the Browser Instead of in a Web Tool
Email headers contain internal hostnames, IP addresses, and personal data. Anyone who pastes them into an online tool sends this information to a third-party server. Why analysis does not need a server and what a tool running locally in the browser can do.
F5 BIG-IP as an Outbound Proxy for Bulk Email Delivery: Persistence, SNAT, Timeouts, and DNS Resolution
A bulk delivery of 1,000 emails per minute runs through a BIG-IP as an outbound proxy to the provider relay. This article explains why sticky sessions provide no benefit here, how to properly resolve the provider hostname using an FQDN node, and which SNAT, timeout, and connection-limit settings actually determine throughput.
Determining a Mail Server’s Load Profile: Bursts, Peak Rates, and Recipient Structure from Message Tracking
How many emails per minute does your mail server really process, and how high are the peaks? How to use PowerShell and Exchange Message Tracking to determine the real load profile: rates per minute and hour, burst duration, recipient structure, message sizes, and common analysis mistakes.
Planning email load tests: Tools for 10,000-email bursts under Linux and Windows compared
Anyone migrating a gateway or sizing an email environment needs reliable figures rather than gut feelings. Which tools generate bursts of tens of thousands of emails, what a clean test plan looks like, and how to evaluate the results from logs.
SMTP Load Testing with Apache JMeter in Practice: 10,000 Emails, Five Rule Paths, One HTML Report
A complete load test from A to Z: a test plan with a message mix along the ruleset paths of an encryption gateway, a portable setup without installation, 10,000 emails in a burst, and analysis using the JMeter HTML report—including the issues that actually occurred.
Analyzing Exchange Mail Flow: Message Tracking, SMTP Logs, and Receive Connectors
How to systematically determine where a message ended up in Exchange On-Premises, Hybrid, and Exchange Online: queries with sample output, how to read the SMTP log correctly, and the issues that regularly lead to incorrect conclusions.
Rebuilding Apache James Rule Sets in a Structured Way: Tools and Method
Mailet rule sets that have grown over the years contain dead paths that no one recognizes anymore. Learn how to analyze the rule set as a graph, reliably find unreachable code, and design the migration so that a single Mailet keeps the rollback path open.
What We Can Learn from Science for IT Troubleshooting
Falsifiability, control groups, confounding variables, and sampling bias: the method natural sciences have used for centuries solves exactly the problems where IT troubleshooting regularly fails, illustrated with examples from mail flow.
Who Is Actually Delivering to Your Tenant? Aggregating Sending IP Addresses
A single report shows which systems actually deliver mail to your tenant: forgotten connectors, applications sending directly, and service providers nobody documented, including the typical analysis errors involving pagination logic and interpretation.
Guide for DNS Administrators: MX, SPF, DKIM, DMARC, and Common Pitfalls
Anyone managing a zone will usually receive ready-made mail records and only needs to publish them. What regularly goes wrong: the 255-byte limit for DKIM, duplicate SPF records, the lookup limit, MX pointing to a CNAME, automatically appended zone suffixes, and policies nobody enforces anymore.
Renewing a Certificate on the Cisco SMA
Certificates can only be installed on the Cisco SMA through the CLI, and current AsyncOS versions validate the entire chain during import: without a stored root CA, the import fails. This article shows the paths to a new key pair, the OpenSSL approach in detail, how to handle OpenSSL 3's RC2-40-CBC error, and how to import the internal root CA into the appliance's trust store.
Testing SMTP on Linux: From the TCP Connection to Delivered Email
When an appliance stops delivering email, a manual SMTP test is more useful than any log. Learn how to check each layer using built-in tools, what different error patterns mean, and why a load balancer can distort diagnosis.
