I am running a Raspberry Pi harvester with a small number of plots. I would like to keep running these solo while creating new pool plots on my plotter (PC) and then transfer everything to the harvester. Due to limitations on my ISP preventing port forwarding for 8444, I am not networking my PC and Pi. Should I copy my /db and /ca folders to the plotter from the harvester while I plot? The idea is to maintain sync for the OG plots when moving the completed pool drive to the harvester. Or will having those folders on two machines simultaneously running chia-blockchain cause problems? I am using the same keys and wallet.
My question is similar to this one here, which does not yet have an answer.
Hmmm I’m not sure if it is relevant to your situation, but what I’m doing is somewhat similar.
I am running a Pi 4 as full node, and it is currently farming my OG plots.
I am currently re-plotting old plots and converting them to NFT plots one at a time on my plotting machine.
The plotting machine acts as a harvester while re-plotting so I don’t lose out on proofs.
I am using the same keys on both machines, both upgraded to Chia 1.2. They are networked on my home network using the instructions from ‘Farming on many machines’ on the chia wiki page.
The Pi4 picks up all plots on both machines, even new NFT plots.
I have not altered my wallet in any way.
Not sure if that answers your question.
Here is the chia farm summary from my pi.
Local machine is my pi.
Remote harvester is my plotting machine, with replotted NFT plots.
All new plots pass chia plots check, and proofs are passing filters etcetc
That’s a really ideal situation. I used incorrect terms, which is causing some confusion. Because I can’t use port forwarding on 8444, I’m stuck with two machines running separately:
RPi4 harvester running as a full node with OG plots.
PC for plotting, which is not networked with the harvester.
These machines are not networked.
The question I am trying to resolve is: if I am running two separate machines, one for harvesting and the other for plotting, should my workflow be:
copy the /db from the harvester to the plotter
plot on plotter using copied /db while continuing to harvest with original /db
once plotting is completed, copy plotter /db to harvester to run OG and pool plots off of the Pi
in order to save sync time on the harvester?
Or will problems arise from running the two /db at the same time?
Why… do you need to copy the /db and /ca folder to the plotter if you’re not going to connect it to the network?
My understanding is you’ll only need to copy the /ca folder if you plan to run it as a remote harvester (which you’re not doing), and copy the /db folder to the plotter only if you plan to run the harvester as a full node (it only helps you sync faster).
So you want to run two separate full nodes? Im mean, you can… and they’ll run as two separate nodes, farming on the same wallet and the same keys.
Are these machines not on the same network? Or are they at two physically separate locations? Like your Pi is at home and your Plotter is at work?
Not sure why you can’t port forward 8444 on your router? Never heard of an ISP not allowing you to open a port on your router.
From your advice, it sounds like I do not need to copy the wallet or ca files. Both the harvester and the plotter will be running full nodes, as they are not networked. From what I understand, if I create a new /db folder on the plotter, and then replace the old /db on the harvester, my pool plots will sync faster, but I will need to resync my OG plots. I am hoping that by copying the harvester /db from the harvester to the plotter before pool plotting, both the OG and pool plots will sync faster once the new pool plots are migrated to the harvester.
I am behind a CGNAT, which limits the ports I am able to use for farming as the public IP is not static.
Afaik you don’t need the plotter to be connected to the chia network to create plots. Creating plots is a independent activity that only requires the correct keys (and chia software) to create cryptographically valid plots.
The plotting machine can then fill up a drive full of plots, and then you can connect that drive to your harvester, which is connected to the chia network to receive challenges and then search your plots for correct proofs. After connecting a new drive of plots to your harvester, your harvester doesn’t need to be resync’d (it should already be sync’d up since you’re farming OG plots on it anyway), and no files (other than plots) need to be transferred.
If you want your plotter to farm plots while it is creating plots your can turn on the chia farmer, and it’ll run as a separate node. So long as both machines are running the same keys and connected to the same wallet it should run just fine, except for consuming a little more bandwidth.
No problem, I hope I managed to help you in some way. Sorry if my responses are a little generalized. Pretty sure we’re still not on the same page lol.
