Mail sending stopped working - M365 Mail SMTP

We have
XIBO 4.4.2
docker 29.5.0
Ubuntu 26.04 LTS

Issue

Since friday the server does not send any mails.
Neither TFA nor Mail alerts are being sent anymore.

There have been no changes made, so I want to find the reason, but do not know how to debug a docker container or the mail sending of it.

We use Microsoft 365 for mail sending, I have set-up a connector for the xibo server and everything was working fine until friday.

The config.env:

Can anyone explain how to debug the issue? I have no idea.

I have no idea why, but after upgrading from 4.4.2 to 4.4.3 mail sending works now.

Prior I have stopped, restarted, looked into the logs etc. and did not find the reason.
Now it’s working again… This is strange.

Anyway, I have a second XIBO installation with the same settings that does not send mails since XIBO 4.0 and have no idea why.

Any idea how to find out why it is not sending mails would be nice.

After 24 hours the mails are again broken. Has anyone an idea how to fix this?

I’m having the same issue on 4.4.2. I upgraded to 4.4.2 on May 5th and it was working until May 13th. Now it’s not working. Maybe some Ubuntu update caused the issue?

Actually the only solution I have is to do a

sudo docker compose down && sudo docker compose -f cms_custom-ports.yml up -d

at around 08:45 a.m.

Editing config.env is not necessary, but this command only work after this time. Doing it earlier does not help. It still breaks. Do you also use a SMTP Relay (Connector)?

I think I fixed the issue by changing the config.env to the same domain that I’m sending email to.

##Rewrite domain (the domain your email will appear to come from)
CMS_SMTP_REWRITE_DOMAIN=

Then restarted the container.

sudo docker compose -f cms_custom-ports.yml up -d

I found this by checking the logs in /var/log/exim4 after bring up bash in the container.

docker compose -f cms_custom-ports.yml exec -it cms-web bash

I had the error message: “Mailing to remote domains not supported”

Let’s see if this keeps working after some amount of time.

Nice, I hope it works until tomorrow afternoon. Unfortunately my exim4 directory in the docker container is empty, so I can’t agree with the same error.

Please let me know if it works for more than 24 or 48 hours, because I also thought that a change in config.env worked, but next day it was always broken.

I tried a few changes of settings and every change that did not matter worked just until next day 08:45 a.m. CET

I am still searching…

Hi Aaron!

Did that solve your issue? If yes, how does your config.env look like?

No it did not fix the issue. After some amount of time after restarting the container, it stops working. I think the service might be crashing in the container. I need to do some more digging.

It would help if we heard from one of the developers on this issue.

The only information I have, is that mail sending via SMTP Relay without credentials is not supported by XIBO.
In my case I need a cron job that does a container down and up after 08:40 am every day. Then the TFA and mail delivery works.
The strange thing on my side is the time. Whenever I restart docker it always stops working at nearly exact this time (08:40 am).
On two different servers (one in my office for testing and the other is an Azure VM) the same behaviour. And why did it run for more than half a year?
The next try from my side is to move XIBO out of the docker container to see if that changes something and maybe I can debug it better.

Actually I have a open ticket at microsoft regarding this problem. Hopefully they find the reason.

We expect you to be sending email via SMTP on port 587 (ie authenticated). You need a user in your Microsoft domain to authenticate as to send email. You can’t relay unauthenticated via port 25 as your message won’t pass SPF/DMARC etc.

This issue seems to have been resolved for me in version 4.4.4.