[Global] Email delivery issues

Update timeline

1. investigating Mar 18, 2026, 03:41 PM UTC

   We’re investigating an issue affecting email delivery for messages processed by CodeTwo services to apply email signatures. We’re also checking whether this may be related to an Azure DNS problem. Please stay tuned for further updates.

2. investigating Mar 18, 2026, 03:52 PM UTC

   We’re seeing indications that Azure DNS issues could have affected email flow for some customers using CodeTwo services. DNS appears to be ok again, and several previously affected customers have already confirmed successful email delivery. We’re continuing to investigate to rule out any other causes. No configuration changes or updates were made to our services today.

3. investigating Mar 18, 2026, 04:10 PM UTC

   Microsoft has acknowledged the issue after receiving reports from multiple organizations. They are investigating on their side as well, but the root cause has not yet been identified. On our side, we are no longer receiving additional reports and all systems appear to be operating normally, but we’ll continue to monitor the situation and carry on with our investigation.

4. monitoring Mar 18, 2026, 04:46 PM UTC

   It looks like the issue is no longer occurring, but we are continuing to monitor the situation and are still investigating the root cause. Here’s a quick summary of events: 1. Some of our mail processing servers were temporarily unreachable, most likely due to an Azure DNS outage. As a result, some emails sent by our customers (using the cloud-mode for signatures) were rejected with a bounce‑back (NDR) by Exchange Online because our mail processing domain could not be resolved at that moment. 2. Some emails still went through. This happened either because they did not hit the Azure DNS issue (not all messages were affected), or because they were queued by Exchange Online for delivery and Azure DNS came back online before the system generated an NDR. 3. Once DNS became available again, our domain was visible and queued messages could be delivered normally. 4. Nothing in this incident occurred within CodeTwo’s systems. The issue happened entirely before any message reached our service pipeline, which is why our service logs did not show any abnormalities. 5. Emails that did not generate an NDR (did not bounce back) can be assumed to have been delivered successfully. Emails that generated an NDR (bounced back) will need to be resent by the original senders. We are continuing to investigate this issue with Microsoft (who has now confirmed the Exchange Online infrastructure experienced an issue) and are trying to determine why Azure DNS became unavailable for a short period. We'll let you know about the results of that investigation. All CodeTwo services are operating normally.

5. resolved Mar 18, 2026, 06:32 PM UTC

   We have confirmed that the situation has returned to normal, and Microsoft has acknowledged that the issue originated on the Microsoft 365 side. For now, we are closing this incident, but our 24/7/365 Monitoring Team will continue to keep a close eye on the service. We will share more details in the post‑mortem once Microsoft provides us with their RCA.

6. postmortem Apr 27, 2026, 01:07 PM UTC

   # Summary On March 18, 2026, a small subset of CodeTwo customers experienced temporary issues with email delivery. The incident was caused by a Microsoft networking and DNS‑related issue within Microsoft 365 infrastructure and was not the result of a failure in the CodeTwo service. # Impact * Only a limited number of customers were affected. * The issue occurred intermittently. * The impact was limited to customers whose mail flow depended on Microsoft 365 services affected by the incident. # Root cause Microsoft confirmed that the incident \(ID MO1255228\) was caused by a temporary issue with their networking infrastructure, where part of the infrastructure intended to handle production traffic was only partially reinstated. This led to DNS and load‑balancing issues, affecting services such as Exchange Online. The incident originated entirely within Microsoft’s infrastructure and was outside of CodeTwo’s control. # Resolution Microsoft manually reinstated the full networking infrastructure and confirmed service health recovery. Following this action, telemetry indicated stable and healthy service operation. # Why CodeTwo was affected CodeTwo services integrate directly with Microsoft 365 and Azure and rely on Microsoft‑managed DNS and networking components. DNS failures at the Microsoft platform level affect all services integrated with Microsoft 365, including third‑party solutions. There is no technically viable way for any Microsoft 365-integrated service to be fully immune to platform‑level DNS or networking failures on Microsoft’s side. # Preventive measures * CodeTwo services are monitored 24/7/365. * We operate redundant and failover‑capable infrastructure. * In case of Microsoft‑side incidents, we take all actions within our control to minimize customer impact, including traffic rerouting where technically possible. Microsoft stated that this type of incident has been extremely rare and committed to improving safeguards to prevent similar events in the future. # Current status Microsoft has confirmed that the affected services were fully restored and operating normally. No further issues related to this incident have been observed.