Thank you for these details which I may have missed… But these rewards depend on minting a block!
Thank you very much man! I’m registered and ready to roll because of you and your detailed guide.
What I cannot decide now is how to escalate from here. I’m at 1 bp and 1 relay, where do you recommend me to start reading regarding infrastructure optimization? As in: What should be my pool objective be to better improve the chances of minting a block for my delegators? more relays? More nodes? Better cpu/ram ratios?
Again, thanks a lot for your time!
there is a recommendation for 2 relays but you can start with 1 relay as well (as long the relay will have an uptime close to 100%).
the pool objective… I don’t know, you should think at what you want to do, want to help projects (ready to donate 10-15% from your profit for charity organizations), or want to reinvest the profit in something… it’s up to you. To attract delegators (the hard part) you must make your pool known on social media like yt, fb, twitter. etc
Regarding the hw upgrade, you must keep your nodes under monitoring for a while and see if you find any issues like servers restart, cpu high/overloaded, etc
for the moment 8G ram should be ok
Hey everyone, thanks a lot for sharing your experiences and expertise - it’s enabled me to set up a Staking Pool, for which I’ve created a nice monitoring dashboard in Grafana. I’m happy to give back to the community by helping anyone that needs assistance!
Stéphane
5 Likes
thank you for helping the community,
it looks amazing!
you can create a new topic, how to set-up a grafana dashboard;
this way it will be more easy for other people to find it 
cheers,
2 Likes
WOW that looks amazing! I was just looking right at some answers in this post, and found some mentions at something like this by Alex I believe, but I never expected it to be so beautiful XD
I too look forward to a guide, or maybe we can do it together if you wish. I can help you in following the step by step to make sure it works. My setup is blank right now, right where Alex’s guide ends, maybe we can write something as a folloup to this guide.
1 Like
There is one mystery I’m still trying to solve. My relays are shown on the image, one is a quad-core Intel i5 with 10GB RAM and the other is a much older 2-core Centrino with 8GB of RAM. I don’t understand why the quad-core Relay1 seems to use a lot more CPU and RAM than the Relay2, logically it should be the opposite. I set both up at the same time with a clean install of Ubuntu 20.04, and nothing else is running other than cardano-node and related processes. I tried running Relay-1 with the +RTS -N4 -RTS run option in order to benefit from using the 4 cores, but then it consumes even more CPU power, with all 4 CPUS tracking at 60%. The only difference between the two is that Relay1 was installed with cardano-node 1.26.0 whereas Relay2 is running 1.25.1 . Any thoughts are welcome!! This monitoring dashboard screenshot shows when I tried the -N4 option you can see all four CPUS tracking at 60%.
eeee, why u installed 1.26.0? wasn’t released yet; perhaps this is the issue
rebuild the node with 1.25.1 … it will take only ~30 min…
I followed the instructions and following the
git fetch --all --tags && git tag
it said to use the latest tag that appears in the list, which was 1.26.0.
Ok i will rebuild that one with 1.25.1 to see if it solves the issue.
Thanks!!
1 Like
Success!
Re-installing 1.25.1 solved the CPU issue. Now all four CPUs are running at < 15%.
Is there a bug in 1.26.0? How to report it?
Here is the link to my dashboard: Grafana
It’s not a bug because it wasn’t released yet 
it’s in testing I believe
Alex, I am stuck in step 4, can’t get my relay and bp connected. Relay is connected to node fine, I’ve updated the topology file, and set the custom peers, and set firewall rule? Im at a loss.
question: the nodes are 100%synced?
if yes
from Producer:
telnet Relay_IP Relay_port
from Relay:
telnet Producer_IP Producer_port
u should see connected… if yes then check the topology of Producer (check if u wrote the correct IP + port and no other marks…)
Yes they are 100% synced, both producer and relay say they are unable to connect
then it’s a FW issue
on Producer
sudo ufw allow proto tcp from Relay_IP to any Producer_port
On relay
sudo allow proto tcp from any to any Relay port
Check again with telnet if now it’s saying connected if not then perhaps u have another cloud FW from your Provider… u will need also to open the ports on it
I have back tracked it and my core node doesn’t show that it is listening on the port set in the env file?
Not listening because it’s not opening for any … it is opened only for the relay…
Ok, I will keep looking. The core connects to the relay, got to get the relay to connect to the core.
Producer:
sudo ufw status… the rule is correct? (the IP for relay is correct)
Yes it is correct, I even deleted it and re applied the rule.