In the Cluster Management Console I have configured a smtp connection with a sender email address, e.g. email@example.com [Home -> Settings -> Emails -> SMTP server]
Cluster Management emails come through my smtp service and are received from this sender email address.
In my Environment, there are no settings for sender email address. [Settings -> Integration -> Problem Notifications -> Set up notifications -> Email]
Problem notifications coming through this integration seem to be going out to Dynatrace Mission Control and are received from an external address 'firstname.lastname@example.org'.
Can we configure integration/email/problem-notifications to use the cluster smtp connection, and use the sender email address configured therein?
I've converted this to an idea.
Here is a better problem statement:
Problem notifications going to an email integration go external through Mission Control, then back to the emails recipients I have set. This raises privacy and confidentiality risks for the information in my notifications.
If the email integration was revised to use a configurable smtp connection, then I could send these problem notifications internally, avoiding any hops outside.
The cluster notifications have configurable smtp connection...
Hi I've converted that to the question 🙂
Your expectations are right. If you configure SMTP server for notifications this configuration should be used. So problem notifications should come "from" exactly same address as you configure in CMC.
E-mail notifications will come via Mission Control and "From" address will be converted and enforced to 'email@example.com' only if the SMTP server cannot be reached (as a fail-back). That needs to be investigate - I'd go with a support case here. If everything is good at your side, then it's a bug in Dynatrace - support case is appreciated here as well 🙂
Hope this helps!
Hi Radoslaw - thank-you very much for the advice. You are correct, the problem notification will use the CMC configuration. I ran another test and confirmed this.
It's embarassing now, but I had configured the CMC smtp on our cluster and inadvertently was testing the problem notifications on a separate proof-of-concept system that didn't have the CMC smtp config.