502 Bad Gateway nach Upgrade auf 2.7.0beta2

Aktualisierung über die CLI:

cd /var/discourse
git pull
./launcher rebuild app

Alles lief problemlos, aber ich erhalte einen 502 Bad Gateway-Fehler.

docker ps

CONTAINER ID   IMAGE                 COMMAND        CREATED          STATUS          PORTS                  NAMES
6598e9407e29   local_discourse/app   "/sbin/boot"   35 minutes ago   Up 22 minutes   0.0.0.0:2045->80/tcp   app

./launcher logs app

(105) exiting
ok: run: redis: (pid 47) 17s
ok: run: postgres: (pid 44) 17s
supervisor pid: 114 unicorn pid: 118
config/unicorn_launcher: line 71: kill: (118) - No such process
config/unicorn_launcher: line 15: kill: (118) - No such process
(114) exiting
ok: run: redis: (pid 47) 21s
ok: run: postgres: (pid 44) 21s
supervisor pid: 123 unicorn pid: 127
config/unicorn_launcher: line 71: kill: (127) - No such process
config/unicorn_launcher: line 15: kill: (127) - No such process
(123) exiting
ok: run: redis: (pid 47) 26s
ok: run: postgres: (pid 44) 26s
supervisor pid: 132 unicorn pid: 136

curl localhost:2045

<html>
<head><title>502 Bad Gateway</title></head>
<body>
<center><h1>502 Bad Gateway</h1></center>
<hr><center>nginx</center>
</body>
</html>

Discourse läuft in einem LXC-Container. Mein zweites Forum, das ebenfalls in einem LXC-Container läuft, funktioniert nach dem Update einwandfrei. Sehr seltsam.
Irgendwelche Ratschläge?

Update:
Ich sehe, dass ein Ruby-Prozess noch läuft und viel CPU verbraucht:

/usr/local/bin/ruby /usr/local/bin/bundle exec config/unicorn_launcher -E production -c config/unicorn.conf.rb

Update2:
Gelöst durch Ändern des LXC-Containers von unprivilegiert auf privilegiert.