This is super weird.
You are using separate servers for relays and BPs. Right? So total you have 3 separate servers? Right?
And on BP you see 12 connection to new-Relay? What is source IP? Is this BP? BP should have only 1 connection to each relay.
Correct 3 servers total – 1 BP and 2 relays
Relay 1 is 207.180.224.185
Relay 2 is 38.105.209.35
I put x.x.x.x for the BP IP below
:~$ netstat -n -a | grep 6000
tcp 0 0 0.0.0.0:6000 0.0.0.0:* LISTEN
tcp 0 0 x.x.x.x:35837 207.180.224.185:6000 ESTABLISHED
tcp 0 0 127.0.0.1:40842 127.0.0.1:6000 ESTABLISHED
tcp 0 0 x.x.x.x:38085 38.105.209.35:6000 ESTABLISHED
tcp 0 0 x.x.x.x:6000 38.105.209.35:34453 ESTABLISHED
tcp 0 0 x.x.x.x:6000 38.105.209.35:37629 ESTABLISHED
tcp 0 0 x.x.x.x:6000 38.105.209.35:40113 ESTABLISHED
tcp 0 0 127.0.0.1:6000 127.0.0.1:40842 ESTABLISHED
tcp 0 0 x.x.x.x:6000 38.105.209.35:42087 ESTABLISHED
tcp 0 0 x.x.x.x:6000 38.105.209.35:46195 ESTABLISHED
tcp 0 0 x.x.x.x:6000 38.105.209.35:43611 ESTABLISHED
tcp 0 0 x.x.x.x:6000 38.105.209.35:39145 ESTABLISHED
tcp 0 0 127.0.0.1:40844 127.0.0.1:6000 ESTABLISHED
tcp 0 0 127.0.0.1:6000 127.0.0.1:40844 ESTABLISHED
tcp 0 0 x.x.x.x:6000 38.105.209.35:35437 ESTABLISHED
tcp 0 0 x.x.x.x:6000 38.105.209.35:35793 ESTABLISHED
tcp 0 0 x.x.x.x:6000 38.105.209.35:44429 ESTABLISHED
tcp 0 0 x.x.x.x:6000 38.105.209.35:34497 ESTABLISHED
tcp 0 0 x.x.x.x:6000 38.105.209.35:44215 ESTABLISHED
tcp 0 0 x.x.x.x:6000 207.180.224.185:41917 ESTABLISHED
Hmmm, several cardano-node instances running on relay 2 (by accident)?
No, I just see 1 cardano-node process on relay2 and the topology.json looks correct as well. It has one producer address pointing at the BP and a bunch of other producers based on the topology pull.
Something is not ok with your relay - 38.105.209.35
it should not have more then 1 connection toward BP. Maybe check logs of it? What cardano-node version you are running there?
Yes, I think there is something wrong here.
I’m seeing lots of IPSubscription Error and connection refused errors. It appears the node is running just fine and gLiveView looks fine, too.
Glad I had this discussion with you all. I need to do some sleuthing. When I get it figured out, I’ll respond back.
Just an update…
I did some comparisons of the cardano-node service logs between my relay1 and relay2. They actually were both receiving lots of SubscriptionIp errors and connection refused errors. So, I can’t say if relay2 was acting funny or not as it seemed consistent with relay1. But, after running the relay-topology-pull.sh script, and restarting the node, relay2 logging looks a lot quieter. I need to do the same for relay1, but am going to wait until tomorrow.
I also rebooted relay2, so I could clear up all the incoming connections on the BP node. Even after rebooting, the BP node still shows all the connections if I run netstat command. It seems surprising that netstat results don’t refresh. I will probably reboot the BP node tomorrow to make sure the incoming connections are reset, when I run netstat.
The good news is that when I run netstat command on relay2 and grep for the BP IP address, I only see 2 connections-- 1 incoming and 1 outgoing I assume.
I think I’m good for now and with any luck, my block will get minted in about 3 hours. Fingers crossed.
You can distinguish it by port. If it is 6000, then it is incoming. Someone else contacted the port opened by the BP. If it is some random high port (and the one on the other side is 6000), then it is outgoing.
Actually works the same for all kinds of network protocols. If you netstat your desktop computer, you will see a lot of high random ports connecting to the 443s and 80s of web servers you are browsing at the moment.
U are using the testnet magic for scripts right?
--testnet-magic 1097911063
Topic switched to mainnet, today:
aaa ok my bad then the mainnet magic should be used
Yep, that’s exactly what I see
tcp 0 0 38.105.209.35:41691 <BP_IP>:6000 ESTABLISHED
tcp 0 0 38.105.209.35:6000 <BP_IP>:44081 ESTABLISHED
I believe it’s all good now.
In my experience, to use topology updater with testnet add the following key value pair to the URL that the TopologyUpdater.sh script constructs:
&magic=1097911063
CHG