Hi,

I'm running a relay on debian jessie using packages from deb.torproject.org.

I want to share the problems I had so others are aware of them when upgrading their relays.

While upgrading from 0.2.7.6 to 0.2.8.6 via apt-get, I did a

tail -f syslog

to make sure I notice problems during the upgrade. (I expected a simple restart of all running tor instances)

I use debian's multi instance systemd service file.

When upgrading, all running tor instances are stopped (not restarted, as expected)

syslog shows:

Interrupt: we have stopped accepting new connections, and will shut down in 30 seconds. Interrupt again to exit now.

Clean shutdown finished. Exiting.

(problem is reproducible)

Side note (unrelated to the upgrade but also relevant for the debian tor package from deb.torproject.org repo):

Stopping the default instance stops all instances due to

/lib/systemd/system/tor@.service:

[...]

PartOf=tor.service
ReloadPropagatedFrom=tor.service

How about using the same way as the RPM maintainer does - so one can enable the default instance without affecting all others?

PartOf=tor-master.service
ReloadPropagatedFrom=tor-master.service