No worries. Just post one, so that transition will be obvious. Actually, that time line is 24h, so you can just take one tomorrow morning, if that would work for you.
Really, I was not expecting that at all. I am in software development for really long time, and it is not often that you see such differences.
@Bones You should see the final results after switching to Flex client. Maybe all your pain on that old box was due to the same thing.
I don’t wanna pool, but if my unlucky streak persists I might change my mind.
Old box is ready now, once ram arrives I’ll try again.
It is nothing about pooling. Pooling is just a side effect here, basically just a measurement tool. The problem is with how well node works with the local resources, and how deficient Chia code is.
I am almost certain right now, that your problems were mainly due to the Chia code, not your box.
It runs perfectly on my plotter, even the dust storm only caused 8 missed blocks.
Time will tell.
I dont really wanna use flex, others can do as they choose.
Pvt keys may be safe, but them signing is centralising, no one can convince me thats not true, because it is.
Sorry, that i didn’t say it right. It is not about using Flex at all, but rather a performance difference that properly written software can have. Nothing else.
To end this day on a good note, maybe you can remember Wordstar. Here is a quote about the guy (Barnaby) that wrote it (including IBM’s assessment of how he performed, comparing to IBMers):
“In Oct. of 1978, a month after introduction, Barnaby began coding Wordstar with new features. According to Rubenstein, who carefully tracked Barnaby’s work, it took four months to code Wordstar. This was done in assembler from scratch. Only 10-percent of Wordmaster code was used. That was the text buffering algorithms. In four months Barnaby wrote 137,000 lines of bullet-proof assembly language code. Rubenstein later checked with some friends from IBM who calculated Barnaby’s output as 42-man years.”
It’s prob not you but me, just finished work, tired , and trying to install 2 plotted disks.
I’m that tired I’ve not even bothered creating a harvester, just chucked them in my old box as another node.
I read another post earlier, completely misunderstood it, said I was probably having a blonde moment, then grasped what was being conveyed and deleted my post.
Enjoy your day fella.
Glad to hear this problem was solved easily! Just to clarify we still use the old node with layers. It’s the farmer and the proof of space we’ve made ourselves.
Once we’ve finished our node we expect many to switchover to our full farmer node wallet package. With fees on the network 100% blocks are more profitable than 30% full ones. Obviously the node won’t be perfect from the start but at least any problems we can solve as we know how to fix our own node and can blame ourselves when there’s issues. A KV database should lead to a large performance improvement too.
As people know from our farmer we do listen to all feedback and are quite quick with fixes.
PS: highly encourage everyone to try our farmer first when there’s issues. Helps you figure out if the issue is with your setup or not. Plus your drives should last longer and use less power so it’s a win win win.
I guess I said goodbye to stale. After deep investigation, it seems that my issue was due to chia 1.2.10 and the look up times, basically the more plots you have the more you are affected…
I have to say, that to me, that is just mind boggling. I hope (or rather am certain) that you will never have any more issues.
Could you check, whether your logs still have those partial timeouts, and how those changed (that cat/grep command you did before)?
That would be a great update, as people are struggling with those timeouts tweaking everything, where it may be that those are again a reflection of Chia software not handling those results properly.
