@SouthKoreaLee check my comment #2
Could you please explain how to view the log?
If u know the slot number missed then inside the logs u should search for that slot
24hs later I have 13 missed blocks, would you consider this normal right?
1 Like
Should be fine
1 Like
I was looking into this last night as the missed slot count was annoying me. It seemed to be synchronised with the garbage collection timings. I wrote up a guide that may help people I found snippets of info elsewhere but not much in the way of any explanations so Iโve tried to clarify what the changes do and some more details on how to implement them.
Feedback welcome. Hopefully it helps. Since I made the changes Iโve only had 2 missed slot leader checks and it hasnโt increased in 15hrs
3 Likes
Hello,
I have the same problem. On my BP node I get the warning โConnection to node lost (1/3)โ in ./glliveview.sh. And woops I lost a leadercheck. I have 3 relay nodes and 1 BP node.
The BP is a VM @ stratoserver.net. Every node is on v1.30.1 and guildliveview v1.21.1. I am really worried. My pool is very small and I dont want to miss the 1 in a million chance to create a block.
what is the hardware configuration of the node? 1.29.0 requires ~9,5G RAM for each serverโฆ but if you will run also CNCLI with only ~10G RAM the node will crash
Hi Alex,
3 nodes are having 32GB RAM (1 BP and 2 relay) and one node 16GB RAM(relay). The last node is on a different provider and has the same issues.
All other nodes are on different server but all are having the same issue. In gliveview.sh they sometimes show โConnection to node lost (1/3)โ. I tried reinstall on the base setup, reinstall on ubuntu 20.4, the bashrc fix, topologieupdate with different relays.
when it will happen just check with journalctl -e -f -u cardano-node
or check the uptime of the node with sudo systemctl status cardano-node
check if the nodes are crashing or not
it has an uptime for 1h and 5 min
check with journalctlโฆ the command provided aboveโฆ do you see something with killing?
also type top and check how much RAM the server has,
I will wait for the next "Connection lost " and I will show you the result of the journalctl
Yes I was installing everthing new that night. I started the nodes and after 20 minutes boom โ โConnection to node lost (1/3)โ. I really donโt know what else to do
here is my htop
Try to monitor the server โฆ if u have 32G and 3 nodes up and running which are using ~30G then Could be the reassonโฆ try to start only 2 nodes for test and monitore for few hours/days
every node has 32gb 
Aaaaa, understood nowโฆ hmm
Ok, hmmm I see u are running 1.30.1 version which is still in tests โฆ the latest official version released is 1.29.0
ok, that sounds plausible. I always upgraded to the newest version. Oh i feel so stupid right now ^^
Downgrade to 1.29.0
And is fine, the version is working (1.30.1) but as u can see it can have bugs
Yes I will do it 
Thank you so much. I will tell you about the results
1 Like