It looks like your docker installation had cached the :release tag. To avoid the need for a docker pull in future, we might be able to add some logic to our launcher script.
The app images are handled differently, so there won’t be any caching issue there.
Emails sent during the outage should have been returned to the sending server with a “temporary fail”. Those servers should retry the mail periodically, so hopefully you will see the missing emails arrive in the next few hours.