Mac Mini a1283 has SSD.
Unfortunately, that model maxes out a 8GB RAM if Iâm not mistaken, so youâll be stuck with a slow IBD.
That said, there is probably something else going on with the file IO to explain the > 1 month time frame even with an SSD. It may have gotten into a swap death spiral. In my (admittedly limited) experience, that time-frame is more typical of an HDD or swapping to an external drive over USB.
If you havenât already, you could try stopping Knots, and once it finishes then restarting the system, and then starting Knots back up.
As for your question about linking to systems running StartOS (presumably you mean satisfying the dependency on Knots from another computer), I do not believe that is possible. The RPC for Knots is exposed, but there is no configuration in (for example) the electrs service to configure an external RPC to connect to.
Iâm well aware 
It finally finished last night so looking fwd to configuring DATUM with a bitaxe I have to solo mine!
I do love the remote web interface over Tor that is native in StartOS to manage things anywhere via a browser, and Iâm guessing some background service to ensure Tor keeps things available is part of what is not jiving with the RPC of bitcoin knots download specifically, because as I said, the same hardware with other Linux installs downloads things much quickerâŚ
By âthingsâ you mean the IBD when running Knots on another Linux distro is quicker? If so, I imagine it comes down to how much memory the OS itself requires, and how it configured the swap. I have indeed seen that on the same Raspberry Pi with external SSD, the IBD is significantly longer when running Bookworm than with Bullseye (everything else identical). I believe it is specifically because Bookworm consumes more RAM for the OS than Bullseye.
Just to put a finer point on it â the IBD has far less to do with download speed, and far more to do with RAM. It is not simply a large download.
HmmmâŚin the past twelve hours the IBD has only progressed from 91.39% to 91.69. And it is sitting at the error message of âTimed out. Retrying soonâŚâ for longer and longer periods of time. Should I try Paulscodeâs suggestion of stopping IBD restarting the machine and restarting the IBD? Or should I just leave it alone?
Unfortunatley I have run into a new problem. It stayed stuck at âTimed out. Retrying soonâŚâ for the whole day. And not I cannot even shut it down. It appears that it is just sitting stuck. My connection from my computer to the server appears to be lost. My internet connection is fine. And I am connected to other systems. What do I do now? Power off the server and restart? Will this ungraceful shutdown ruin anything? Or is there something else I should do?
There is a risk of corrupting the blockchain data with a forced shutdown (which would be unfortunate since that would require reindexing). It might be your only option though. If you have a spare monitor and mouse that you can hook up to the server, it might reveal something useful, or give you an opportunity to properly restart if the direct interface is still responsive.
I finally got the computer to log back into it. I think I am going to be able to the shutdown from here. And then restart it. If not I can hook up a monitor and a mouse.
Well I was able to stop it. And then start it. From inside of Start OS. And it is not stuck on RPC and IBD process at restarting in blue. I am really not sure where to go now. Should I stop the server and do a power down and power up? Any ideas? I am at a loss of what to do. I canât imagine this server does not want to complete the IBD process?
I couldnât quite understand your description, but perhaps check the logs to see if there is anything useful?
Ok. Let me try to explain a little better.
I did a stop and start of StartOS. And it stopped and went into start mode. It never even logged me out. When I did the start it has two process on the screen under Bitcoin core Health Checks. RPC and Synced. Both seem to be sitting at the error message âTimed out. Retrying soonâŚâ in yellow. Eventually the RPC health check goes to âSuccess. RPC server is ready for connectionsâ in green.
However the Synced health check seems to stay stuck in Timed out.
I looked at the logs but cannot find anything that I understand. I will attach the log file here so you can see it.
I was asking in the previous message if I should stop the server and then power it down and up? Not sure where to go next.
I am not sure how to attach the log file?
Oh wow. The Synced health check finally just showed 91.95% progress. So I guess it is now running again. Its still cycling through progress and the timed out error message. But seems to be running again.
Yes, my suggestion in case of a swap death spiral was:
That said, I donât think that is the problem in this case, since it sounds like the system is responsive and is showing a percentage now (that problem presents itself as the system having extremely delayed responses, or worst case completely unresponsive).
The main thing to look for in the log is if you see lines that have a progress mentioned (will be a number with several decimal places that approaches 1.0000 for fully synced). Since it has a lot of decimal places, you can see the progress going up easier there than in the UI which only shows a couple decimal places IIRC.
Yes. I think you are right. It appears to show the progress that you said about several decimal places. And it is excruciatingly slow. But still progressing. Thanks for your help. I will let it cook for tonight and check again tomorrow.
It appears that in the service logs. That the progress is still incrementing very very slowly. I still see the progress log entry and 0.920566. So I guess I am ok?
Iâd say as long as the system is still responsive and the percentage is still going up, it is now just a waiting game. Looks like the Latitude 7000 series also maxes at 8GB RAM, so canât upgrade it to speed things up.
âEDIT-- A RAM upgrade might actually be possible. The spec actually says â2 slots supporting 1G, 2G, 4GB, 8GB DIMMsâ, which would imply max 16GB. You could always try dropping in a $10 8GB DDR3 stick if you get impatient.
Actually my patience and a long term fix might warrant this. I think it does go up to 16GB. I will check into it. Would it be ok to do a graceful shutdown. Power off the box. Then restart? And it will carry on where I left off? Or will I have to start over? When powering on the box does anything change as far as logging in or starting the IBD process, or does it just start working automatically?
I would follow this process:
- Stop the Knots service (wait for it to shut down)
- Gracefully shut down StartOS (System > Power > Shutdown)
- Install the RAM
- Power up the PC
- Start Knots
Knots should pick up the IBD where it left off (and will run considerably faster)
Great. Thanks for your help. I will start researching the RAM issue.
1 Like
Ok I ordered a new 8 GIG memory stick from Memorystock.com. They did have the top brands at the best prices. And they did call me to follow up and make sure they had it right. Crucial much more expensive. Anyway now I wait and install when it gets here. One other question. When I power up my machine does it automatically execute Start OS or is there some command line stuff I have to do?