Max Memory Restarting on Traditionally Sharded Bot
What do people here use for facilitating automatic max memory restarts with a traditionally sharded bot?
4 Replies
• What's your exact discord.js
npm list discord.js and node node -v version?
• Post the full error stack trace, not just the top part!
• Show your code!
• Explain what exactly your issue is.
• Not a discord.js issue? Check out #useful-servers.Doesn't work in this case, or at least I haven't figured out how to get it working, because PM2 only monitors a single process so the additional processes created from traditional sharding are not monitored :(
I think what you want is pm2's cluster mode https://pm2.keymetrics.io/docs/usage/cluster-mode/
Based on my understanding, PM2 cluster mode creates multiple processes in itself, similarly to how it would be done for a web application. So doing it with PM2's cluster mode would bypass the sharding manager entirely if I am understanding it correctly, which is unfortunately not a solution to my problem.
I'm not quite sure I understand
You mean I can use cluster mode with one process 
the problem is that the one process only uses like 100mb of RAM
Whereas the other shards combined use a lot more RAM
So I can't get an accurate RAM measurement to properly restart the whole bot.
How can I see the allotted maximum? It's been crashing my whole program when I reach too much RAM.
I believe my issue is more on the collective RAM usage of all my shards rather than individual shard usage unfortunately.
And I believe passing in the max-old-size parameter would only apply to the main sharding process and not to each individual shard process.
the problem is that the one process only uses like 100mb of RAM
Whereas the other shards combined use a lot more RAM
So I can't get an accurate RAM measurement to properly restart the whole bot.
How can I see the allotted maximum? It's been crashing my whole program when I reach too much RAM.
I believe my issue is more on the collective RAM usage of all my shards rather than individual shard usage unfortunately.
And I believe passing in the max-old-size parameter would only apply to the main sharding process and not to each individual shard process.