Investigating delays with message delivery via email

Problem

Email notifications seem significantly delayed. What's happening? Is it on the xMatters side or ours?

Environment

All versions of xMatters, all mail server and client versions

Resolution

To resolve this issue, we'll need to collect a few samples of the messages that were slow to be delivered. We need the original messages so we can check the email headers - for tips on how to review e-mail headers, this site contains instructions on how to do just that.

If the email messages are in msg format and you use a Mac, you have two options:

• Use another app like MailReader; or,

• Log into Office 365 and create a draft uploading the samples. Once the draft is saved you will be able to see the samples in the online version of Outlook under the Drafts folder.

Copy the email headers (also known as raw source or message details) from one of the samples.

Use a website like https://mxtoolbox.com/EmailHeaders.aspx or http://www.levinecentral.com/mail_parse/ to analyze or parse the headers.

Check for any columns that contain the word "Delay" to determine which hop experienced the delay. In the following example we can see on the third row under Hop Delay that it was over 1 day and 1 hour. That indicates the issue was between the source and destination servers where see the longest hop delays and in that case you need to contact that provider or the mail server's administrator team who probably will need to check their logs to determine the root cause.

​

Cause

Email notification delays can occur due to several reasons:

• Connection limit restrictions at the email provider

• Throttling of messages when the number of concurrent email requests exceeds system limits

• Temporary issues with email gateways

• External hops in the SMTP delivery chain (outside of xMatters and the local mail server)

These delays can be caused by anything from antivirus scanning to anti-spam or email security solutions. In some specific cases, email providers may report 'internal connection limit reached' errors, which cause messages to be delayed but not lost.

When investigating, it's important to identify which hop in the delivery chain experienced the delay, as shown in the example above, and contact the appropriate provider or administrator team.