Hi guys. This is my second topic here.
I have around 100tb that I wish to run on 3 computers. Reason for that is simple, if one is off, the other 2 are running.
Yesterday one of my PCs had to be restarted, so this way 2/3 of my farm was still running.
I would like to ask you. Is it ok to run 3 full node computers at the same time?
Friend of mine told me it’s wrong, and I should run only one farmer and 2 harvesters.
But the original idea of Chia was simple. Have 12-24tb plotted, have 20 bucks per week and be happy. But now I have no option to run an external 12hdd box. I have two computers that can handle 12hdds each and I wish to fill them…
So… Should I change something in the settings to achieve the same chance of winning? For example if I would have 60,60,60tb full plotted 3 PC’s in future. Is that a problem or absolutely not?
Far I understand each plot is equal. So if my one pc with 1 plot has won, reward would be transfered to my wallet and would be shown on all ful nodes. Of course I run the same wallet full nod on each.
I know pro’s have no issues like that. I believe it’s much better to run 5 servers with 50disks as solo then 5 servers synced to one. I might be wrong, but keep in mind CPU power, response time, gb of ram, power failure etc etc.
Is it true, if I have only 22 plots for example and someone else has 2200 and we run the same block so my chances are out?
Please help. I believe I’m not alone in this.
I was seaching over internet and people have different opinions on this. So would apprechiate answer from community.
i have 4 full node on same wallet on 4 dif servers.
if you run multiple noed within the same network, be sure to disable upnp in the config except for your main full node
I run 4 computers in my living room. Every computer was set on full node. Every one plots and farm at the same time. Is it wrong?
Do I have to set something? Because these computers run on its own and are not connected to each other (only wifi).
Can you please be more specific what is upnp?
Thank you. Did you change any settings and do you have any problems with earnings?
I believe the issue was that you might find with UPnP disabled (and who wants it enabled anyway) that having multiple full node machines on one network might be a problem.Its a lot of traffic on your WAN connection all on port 8444.
There must be a way to limit port 8444 on whichever machine you want to assign as your primary full node, to external access and then have secondary machines sync from your primary node. If the secondary machines could use port 8445 that would be great.
As a windows GUI user so far, I personally don’t fully get the full node, harvester, farmer distinction. I’m running one full node which is farming, and two other windows GUI plotters which are blocked in firewall so they dont sync.
When you run full nodes it increases you resource usage in disk, ram, cpu, and bandwidth. If you monitor all these, especially disk usage when the nodes are syncing you can see constant writes to the disk, which can easily be avoided by running harvester only on the secondary computers.
On another post someone mentioned they saw bandwidth usage up to 70 MB/s. I did not measure this, but can say the constant syncing does use bandwidth as I saw constant ethernet activity light blinking. If the bandwidth has to be split up amongst multiple full nodes I can’t imagine it would help with the response time.
As far as future troubleshooting is concerned running dedicated harvesters would simplify things as all they do is create plots and point to the full node. The troubleshooting would be narrowed down to one PC.
To your point if you restart one PC, 2/3 of your farm is still running, on the flip side if two PCs are harvester only and the plots are stored in the full node, restarting or powering off the harvesters will leave you with 100% of your farm running.
Thank you. Do you think running farmer and harvester in 1 pc at the same time with 75% ram usage and 80% CPU can avoid rawards?