I just found out I can update the firmware through the Remote Avalon app (https://play.google.com/store/apps/details?id=com.canaan.avalon).

What surprised me is the following: when I add the device based on IP, it just connects to the machine, even though I have put in a password on the web app. I get immediate access to the machine, can update the firmware and change all the details, including miner configuration. How is the app access this, without ever putting in the password?

The behaviour is the same after the firmware upgrade to 24071801_42c628d. I do get a warning in the app that I should change the password, which I did a long time ago, when I got the device. If I want to change the password, I do need the previously set password, but nothing changes. The app just has access to the device, password or not.

Secondly, after a few weeks of uptime, the machines go into a high fan noise sequence with a terrible hashrate, maybe 1/10th of normal operation. The fan can be heard spinning up to full speed for a while and then spinning down again, with way more noise than normally. After a reboot it's fine again, so it doesn't look like an overheating scenario. We're going into winter here, so ambient temp is actually getting lower.

I understand how the pool works, it has to tune to a balance between number of shares reported and bandwidth. It looks like ckpool is steering to once ever 2 seconds or so. Starting with 10000 isn't that bad, and for a high performing miner that's probably too low, for a low powered miner that's too high, but it's a good middle ground. For a BitAxe it takes between 1-2 minutes to find an acceptable share.

To not accept a suggested difficulty from a miner I can understand, and is not mandatory for a pool. However, if the pool suggests a lower difficulty, it's weird that it doesn't accept this new difficulty until about 40 seconds after this message. In the stratum protocol it states: "The miner should begin enforcing the new difficulty on the next job received.", but switching to the lower difficulty at that point first will result in several shares accepted with 'Above target'. Or I'm missing something in the protocol, that could also be the case.

I'm lottery mining with a BitAxe Supra on the solo pool, and I noticed something strange on startup.

The initial difficulty after a connection is set to 10000. This takes a little while for a BitAxe to find a share, and submits. After a few shares, the pool sets the difficulty down to something around ~500, and the BitAxe starts submitting shares on this difficulty after that, but the first few shares are not accepted due to 'Above target'. This takes about 40 seconds. After that, the shares are suddenly accepted.

There are some logs with stratum communication on https://github.com/skot/ESP-Miner/issues/212.

I don't think that's possible, unless your specific miner actually finds a block, which is really really really rare. The only useful notification would be if the pool you're mining for found a block, but your local miners won't know this.

Try the Raspberry Imager.

NVMe. Don't know if it'll accept sata.

If you don't like the paperclip trick, you can try something like this: https://aliexpress.com/item/1005002524343143.html

With Slushpool, I set the payout to daily with minimum value of 0.01. This is the minimum amount with no payout fee. I've been mining for 5 months now and close to my first payout. This is with 3 units on max.

Can this be done on the command line?

I'm running 3 hash boards on balanced mode on a be quiet! Dark Power 12 750W, in single rail mode. I mainly selected on modularity and number of cables. It had enough cables to power the 3 boards, so with one extra cable I could add one more hash board, it's not close to max power.