pull down to refresh

I run a core/CLN node on Start9. I don't have a real understanding of networks and servers. I usually get stuff to work by trial and error, without real understanding of WHY it works.

My question:

I use either the CLN built in GUI or Zeus as my wallet. After a week or two, sometimes longer, I lose connectivity between the wallets and the server, though the server and node are still running. The quick fix, I have discovered, is to restart the server. The issue is then always resolved, but the situation is annoying, and it's particularly difficult when I don't have access to my network. Did I set up something wrong? Is there a more permanent solution?

Thanks.

This is the nature of long-running services, things have memory leaks, subscriptions or listeners get lost to the operating system / garbage collection... from the hardware up through the operating system to the application is a very complex stack, it's never perfect.

There's not a production service in the world that doesn't have external monitoring to trigger reactive restarts, and/or scheduled proactive restarts.

You can schedule a chron job to proactively restart weekly or every few days, I don't have a start9 in front of me at the moment to see if they expose a GUI for this.

reply

Thanks very much for the information. I will see if I can automate restarts.

reply
41 sats \ 0 replies \ @Ohtis 1h

Ugh, this happens to me too Restarting always works, but yeah… would be nice if it just stayed connected.

reply
41 sats \ 0 replies \ @DarthCoin 2h

Only @Solomonsatoshi can answer so complex questions. He's the expert in LN nodes.

reply

I’m not sure what the best solution is, but you can always set up a task in the operating system to restart every week. Haha!

reply