Darknode Alerts

Hi

What kind of alerts darknode should be created on Digital Ocean just in case something goes wrong?

CPU/Disk/Memory/Bandwith?

thanks

This is what I use:

15 mimute load average running high
(15 minute load average above 70 for 5 minutes)

CPU running high
(CPU above 70 for 5 minutes)

Memory utilization running high
(Memory utilization above 70 for 5 minutes)

I am likely going to add these exact alerts too because I see other people doing it and that has been an instinct that has gotten me through much of life.

As a super new node operator I would like to piggyback a bit here with some ELI5 questions:

What would you do if these alerts went off? I assume a darknode opperator would probably consider upgrading the node’s vps, right? Am I correct in thinking that these alerts would basically indicate that the network is getting heavy traffic and that is pushing the node toward its purchased capacity? What happens if a node is underpowered?

Hi,
if you have an alert the first thing to do is find out why it alerted , then address the problem. If the load, cpu, etc is too high, the node will crash.

2 Likes

Hi FanaticalFishing,

For instance:

  1. alert for load cpu I can go and see D.O log and from there figure out what to do

Can one just restart the darknode to fix? I mean how long can a darknode can stay down? up to 24h?

thanks

1 Like

yes you can look at the logs to see what is going on. the darknode program so far has been very stable so I doubt we would have any problems. But yes , a reboot probably would fix it, at least temporarily.
Loongy said that a darknode in Sao Paulo went offline due to a network problem on AWS some time ago, It didn’t affect anything. I’m not sure how long a node can be offline before it gets removed.