Auto-scaling cluster system for discord.py

IPC is optional.

How to use

Replace L13 with your bots token, and ideally update bot.py with your actual bot.

Example log output:

[2019-10-0219:58:53,003 Cluster#Launcher/INFO] Hello, world![2019-10-0219:58:53,096 Cluster#Launcher/INFO] Successfully got shard count of 1((200,'OK'))[2019-10-0219:58:53,097 Cluster#Launcher/INFO] Preparing 4 clusters
[2019-10-0219:58:53,098 Cluster#Alpha/INFO] Initialized with shard ids [0,1,2,3], total shards 16[2019-10-0219:58:53,098 Cluster#Beta/INFO] Initialized with shard ids [4,5,6,7], total shards 16[2019-10-0219:58:53,099 Cluster#Charlie/INFO] Initialized with shard ids [8,9,10,11], total shards 16[2019-10-0219:58:53,100 Cluster#Delta/INFO] Initialized with shard ids [12,13,14,15], total shards 16[2019-10-0219:58:53,100 Cluster#Launcher/INFO] Starting Cluster#Alpha
[2019-10-0219:58:53,107 Cluster#Alpha/INFO] Process started with PID 24030[2019-10-0219:59:59,573 Cluster#Alpha/INFO] Process started successfully
[2019-10-0219:59:59,574 Cluster#Launcher/INFO] Done![2019-10-0219:59:59,575 Cluster#Launcher/INFO] Starting Cluster#Beta
[2019-10-0219:59:59,588 Cluster#Beta/INFO] Process started with PID 24054[2019-10-0220:01:04,002 Cluster#Beta/INFO] Process started successfully
[2019-10-0220:01:04,003 Cluster#Launcher/INFO] Done![2019-10-0220:01:04,004 Cluster#Launcher/INFO] Starting Cluster#Charlie
[2019-10-0220:01:04,017 Cluster#Charlie/INFO] Process started with PID 24083[2019-10-0220:02:09,205 Cluster#Charlie/INFO] Process started successfully
[2019-10-0220:02:09,217 Cluster#Launcher/INFO] Done![2019-10-0220:02:09,218 Cluster#Launcher/INFO] Starting Cluster#Delta
[2019-10-0220:02:09,238 Cluster#Delta/INFO] Process started with PID 24106[2019-10-0220:03:13,711 Cluster#Delta/INFO] Process started successfully
[2019-10-0220:03:13,713 Cluster#Launcher/INFO] Done![2019-10-0220:03:13,713 Cluster#Launcher/INFO] All clusters launched
[2019-10-0220:03:13,714 Cluster#Launcher/INFO] Startup completed in260.7107082050061s

Extra Features

Automatically restarts clusters that died with a non-zero exit status

[2019-10-0220:04:58,805 Cluster#Launcher/INFO] Cluster#Alpha exited with code -1[2019-10-0220:04:58,806 Cluster#Launcher/INFO] Restarting cluster#Alpha
[2019-10-0220:04:58,811 Cluster#Alpha/INFO] Process started with PID 24149[2019-10-0220:06:03,831 Cluster#Alpha/INFO] Process started successfully

Will turn off if all clusters close with exit code 0

[2019-10-0220:07:13,901 Cluster#Launcher/INFO] Cluster#Alpha found dead
[2019-10-0220:07:13,903 Cluster#Alpha/INFO] Shutting down with signal <Signals.SIGINT:2>[2019-10-0220:07:13,903 Cluster#Launcher/INFO] Cluster#Beta found dead
[2019-10-0220:07:13,904 Cluster#Beta/INFO] Shutting down with signal <Signals.SIGINT:2>[2019-10-0220:07:13,904 Cluster#Launcher/INFO] Cluster#Charlie found dead
[2019-10-0220:07:13,905 Cluster#Charlie/INFO] Shutting down with signal <Signals.SIGINT:2>[2019-10-0220:07:13,905 Cluster#Launcher/INFO] Cluster#Delta found dead
[2019-10-0220:07:13,905 Cluster#Delta/INFO] Shutting down with signal <Signals.SIGINT:2>[2019-10-0220:07:18,911 Cluster#Launcher/WARNING] All clusters appear to be dead
[2019-10-0220:07:18,914 Cluster#Launcher/INFO] Shutting down clusters

Proper KeyboardInterrupt handling and cleanup