search

near / stakewars-iii

Stake Wars: Episode 3 challenges and place to report issues
87 stars 177 forks source link

Observations: Node failed to catch up after adding network.public_addrs #98

Closed Thesephi closed 2 years ago

Thesephi commented 2 years ago

So.. my node was working well for the entire day (good uptime, normal CPU usage).

Then I got the announcement to add network.public_addrs to the config, which I did.

ISSUE 1

After restarting the node, I noticed this in the log:

encountered unrecognised field: network.public_addrs

So it looked like the node didn't understand the config. I'm sure there's no typo as I copied & pasted directly from the instruction. It's the literal string public_addrs

ISSUE 2

After restarting the node, I expected that it'd take maximum a minute for the node to catch up & continue validating.. But I was wrong and/or very unlucky. My node never caught up afterwards:

node-fails-to-catchup

I tried restarting several times. Took me hours of waiting in vain. My node is falling behind more and more.

My naive thinking is that this could have been a poor network state. But it could have been something else, as I notice these kinds of logs quite often:

mailbox-closed

and

connectionreset

Hope this issue helps shed some light or trigger some useful discussions 🙂

Thanks all!

Thesephi commented 2 years ago

After several discussions, I decided to upgrade storage speed, and things worked well since then. I'm still curious why it worked okay before, and only became an issue recently. Anyhow, if it works, it works :)

Issue 1 (the node doesn't seem to recognize the public_addrs config option) remains to be discussed (imho).

joesixpack commented 2 years ago

Use the latest commit and it will recognize it.

Thesephi commented 2 years ago

Confirmed that the issue about public_addrs not being recognized, has been resolved in a more recent release commit. Closing issue. Thank you for the insight above!