starting 6.3, the internal mail server will do STARTTLS. this env
allows apps to configure themselves appropriately for pre 6.3 and
post 6.3 appropriately.
we trigger a re-configure which ensures that the new env gets put
in the database and then in the container.
we export using the old addon containers, which has a bug that it crashes
when db is missing. so, we have to skip them already. the crash then causes
future exports to also fail because it is restarting.
this happens because we have some bug in sftp container causing uninstall(s) to
fail. the database of those apps are gone but the export logic then tries to export
them and it all fails.
restart service does not rebuild automatically, we should add a route
for that. we need to figure where to scale services etc if we randomly
create containers like that.
Girish Ramakrishnan