Probably in line with other recent reports here, my debian buster install did break upon the following update:
jicofo/stable 1.0-508-1 -> 1.0-539-1 amd64
jitsi-meet-prosody/stable 1.0.3729-1 -> 1.0.3928-1 all
jitsi-meet-web-config/stable 1.0.3729-1 -> 1.0.3928-1 all
jitsi-meet-web/stable 1.0.3729-1 -> 1.0.3928-1 all
jitsi-meet/stable 1.0.4101-1 -> 1.0.4335-1 all
The log file used by nginx gets filled with entries like this (where 192.168.1.78 is my firewall running HAProxy):
2020/03/29 00:13:28 [error] 3013#3013: *217 connect() failed (111: Connection refused) while connecting to upstream, client: 192.168.1.78, server: 0.0.0.0:443, upstream: “127.0.0.1:4445”, bytes from/to client:0/0, bytes from/to upstream:0/0
Please be so kind to share information on how to proceed. In the meantime, I did revert the VM and disabled unattended upgrades.
My setup is (not ideal): I am running meet behind HAProxy which does handle the certificate and ssl uploading. As I did not want to mess with the configuration, I did forward to the SSL port of the debian VM running meet with a private certificate. Of course, it would be more efficient to run meet unencrypted then, but I did not find a simple way to do that. Is there a best practice for this that will outlast (regular) updates?