Sorry, I can’t say. I don’t know what the node does when the disk is full.
1 Like
Hey Chris, what are we looking for here in the logs?
If a node has been shunned, we inform that node before blocking all communication to it.
Its something like “considers (you) bad” or similar. The key word is “bad” not “shun”
Right, I’m not sure exactly. I would need to ask Roland. Will clarify for you both.
1 Like
But that is the old message, why is this in the changelog?
Because it was mentioned so much and they’ve done work on when to consider another node as bad
I am still pulling my hair out with this.
So previously a node was told when it was seen as bad, now nodes are being told when they are seen as bad.
The change is what?
we inform that node before blocking all communication to it.
Damn man I have been battling with this for too long.
They were always doing that. I am thinking its just about the changes they did to the method of determining bad more than the message layout
Yeah or that it is now in metrics… idk.
Ah, its because its in launchpad - for sure this is it
1 Like
I’ve asked Roland to reply to you. I’m sure he will be able to provide more information about this. I remember that this was one of his changelog contributions.
1 Like
Thanks Chris!
Sorry, I thought I saw you asking about the changelog. Yeah, that was something I really wanted to get in as part of the releases. It is manually curated, but I think it’s worth doing.
The automated changelogs we were generating really weren’t helpful.
1 Like
Yes I did, its great!! 
I realised I knew the answer so removed the question.
1 Like
Are you folks upgrading or nuking? full fresh start?
1 Like
I am upgrading, no nukes here
Nuke, I always do.
2 Likes
Yeah I considered that, but honestly I have become a little lazy now. So instead of wanting to bother with checking that logs are not overflowing I just want know it is all fresh and good.
I know I definitely also have some nodes in the naughty corner too.
3 Likes
The string in the logs for shunned is consider us as BAD. I’ve had one, although the node (via Launchpad) still seems to be running OK.
[2024-08-27T21:41:08.391974Z WARN sn_networking::event::request_response] Peer PeerId("12D3KooWKt1oHzqDFWGk6toRT8cg6dKsBNc9FoAEP9UHeKwYtUsy") consider us as BAD, due to "ReplicationFailure".
3 Likes
Both. I needed to change port numbers on some machines so I nuked those and upgraded others.
1 Like