This one unpingable satellite

Node Operators troubleshooting
, ,
1

Hey guys.

So, I am running a storage node on a server. I got really fast Internet, 600mbit up and 400mbit down. Pings to America (I’m in Europe) usually take about 10-40ms. Tried it with several domains, several states. All around the same values.

BUT: This one satellite (us-central-1) totally fricks up my uptime. On every other satellite I got 99.9% or 100% uptime. On this one I got 99.2 at the moment, decreasing every other day. I’m slowly starting to panic to be disqualified for this. I mean, seriously, this literally can’t be because of my connection. The server only restarts when the storagenode updates itself, that takes about 4 to 7 minutes everytime.

Any of you guys got the same problem? Devs got any ideas? Would be really thankful for any advice ^^

2

for us-central-1 my SN - 95.7%
others 99-100%

3

My node has been disqualified on us-central-1, maybe this is related.

4

I have 99.9% on all satellites, except saltlake(100%) and stefan-benten (99.7%).

5

The disqualification for downtime is currently disabled: Design draft: New way to measure SN uptimes

The TCP ping is not related to the dRPC ping, the last one is an exchange of messages, not just packets.

The percentage on the dashboard is not a good metric at all, the satellites uses the rate for disqualification (at the moment only audit rate): Script for Audits stat by satellites

6

Yea, but for some reason the dashboard only shows the success ratio instead of score. On the other hand, uptime scores for everyone are probably zero now (they are on my nodes), so the ratio is all there is right now.

7

So, that means that a node won’t be disqualified based on the Uptime checks, if I got that right. That’s good to hear.