Official FutureBit Moonlander 2 Driver and Support Thread

[wttbs]

Did anyone change the fans or modded them somehow? I have 4 of them running, and they are really loud and vibrating. I checked the screws, but it didn't change much.
I am thinking of replacing the fans with "silent fans" or maybe add some rubber washers to the stock ones. I also thought about removing them completely and put a 120mm fan next to the 4 to cool them down.

Are you Moonlanders also running quite loud? Any setup advices?

I have no problem with the fans, they make the sound a fan makes, nothing extreme. Did no mods on the miners.

[jstefanop]

Some guy made a Crypto Christmas Tree using a whole bunch of Moonlander 2 USB miners


https://youtu.be/Et3jgHnTjZ8

hahah that is brilliant...nice find.

[Bitfort]

Some guy made a Crypto Christmas Tree using a whole bunch of Moonlander 2 USB miners

https://youtu.be/Et3jgHnTjZ8

hahah that is brilliant...nice find.

Yea, I like it also. I want such tree as well. But given the number of MLDs it's pretty expensive tree.

Is someone here willing to help out this new set up his new moonlander stick?

Read the OP and come back with specific questions/issues.

Did anyone change the fans or modded them somehow? I have 4 of them running, and they are really loud and vibrating. I checked the screws, but it didn't change much.
I am thinking of replacing the fans with "silent fans" or maybe add some rubber washers to the stock ones. I also thought about removing them completely and put a 120mm fan next to the 4 to cool them down.

Are you Moonlanders also running quite loud? Any setup advices?

I have same issue with one of them. But I place the stick behind window so the noise is not issue anymore (just hope the fan won't die)

[JustAnotherPLT]

Alright, I've been playing with this thing all day. I've targeted it down this issue.

Code:

[2017-12-10 16:15:41] Network difficulty changed to 20.4k (146.3G)

Any time this happens the miner stops running. You can hear the fan wind up a little as if there's an increase in juice that was being used for the processor. Now the yellow light just flashes every few seconds. Once at this point, unless I restart everything, it won't work on any changes. I'm at a loss... please help.

[jwinterm]

Anyone have a good solution on Windows to just kill bfgminer and restart from bat file every 30 or 60 minutes?

https://stackoverflow.com/questions/15602687/batch-file-restart-program-after-every-20-minutes

[dem0nlord]

Arent you guys worried about heat? Even at 600 the cooling thing at the bottom gets pretty hot no?

I didn't think they were very hot at 600. At 900 they are hot, but they keep chugging away happily doing 5.1Mh/s.

[flamman]

Here is a question for our Linux/Raspbian crowd:

My plan was to run this off my RPI 3 as a headless system. I can putty in, do admin tasks, start, etc.
Trouble starts when I try to figure out how to log out & leave the pi mining.
Anyone have a good method for this?

I tried some tests with nohup command:
nohup ./startmoonlander.sh &
-
This does load bfgminer in background (with red LEDs flashing),
But not sure how to bring if forward again and worse the mining still stops when I close putty session.

Yes - I modified a script from the original bfgminer repository and saved it as miner.sh

First install screen..

Code:

apt-get install screen

Then modify in below snippet the /path/to/bfgminer, -u <username> and -p <password>

Code:

#!/bin/sh
n="
"
startscreen() {
	name="$1"; shift
	cmd="$1"; shift
	if ! screen -ls | grep -q "^[[:space:]]\+[0-9]\+\.$name"; then
		screen -dmS "$name"
	else
		for i in 1 2 3; do
			screen -x "$name" -p 0 -X stuff $(echo 'x' | tr 'x' '\003')
		done
		screen -x "$name" -p 0 -X stuff "stty sane$n"
	fi
	screen -x "$name" -p 0 -X stuff "$cmd$n"
}

startscreen miner "/root/bfgminer_5.4.2-futurebit2_linux_armv6/bfgminer --scrypt -o stratum+tcp://litecoinpool.org:3333 -u <username> -p <password>,d=128 -S ALL --set MLD:clock=600 --api-listen --api-network"

Save it to a file and make the file executable. Above script starts & restarts bfgminer.
You can check if it is running..

Code:

screen -list

Which then shows something like..

Code:

There is a screen on:
	2139.miner	(09/12/17 16:12:05)	(Attached)
1 Socket in /run/screen/S-root.

You can then safely log out.
You can see the terminal output when you SSH back in by running..

Code:

screen -r

And by doing crtl+a and ctrl+d while in screen you detach again while keeping bfgminer running.
Within screen crtl+c will stop bfgminer if necessary

" --api-listen --api-network" can be left out - that is just because there I want to be able to query the API

[JustAnotherPLT]

Alright, I've been playing with this thing all day. I've targeted it down this issue.

Code:

[2017-12-10 16:15:41] Network difficulty changed to 20.4k (146.3G)

Any time this happens the miner stops running. You can hear the fan wind up a little as if there's an increase in juice that was being used for the processor. Now the yellow light just flashes every few seconds. Once at this point, unless I restart everything, it won't work on any changes. I'm at a loss... please help.

Honestly looking for some help please.

[Megaquake]

Alright, I've been playing with this thing all day. I've targeted it down this issue.

Code:

[2017-12-10 16:15:41] Network difficulty changed to 20.4k (146.3G)

Any time this happens the miner stops running. You can hear the fan wind up a little as if there's an increase in juice that was being used for the processor. Now the yellow light just flashes every few seconds. Once at this point, unless I restart everything, it won't work on any changes. I'm at a loss... please help.

Honestly looking for some help please.

I sent you a PM

[dora-gt]

Hi,

I'm wondering if supplying 5V power on this plug works or not because my usb-hub is unstable.
Of course power for fan is supplied too.

https://gist.githubusercontent.com/dora-gt/cec006d70356b1bb4684126a7b12b2db/raw/1cfe3d8a4fd01386675d02744dc58210023eedab/1.png

Did you guys already try this?

[jstefanop]

Alright, I've been playing with this thing all day. I've targeted it down this issue.

Code:

[2017-12-10 16:15:41] Network difficulty changed to 20.4k (146.3G)

Any time this happens the miner stops running. You can hear the fan wind up a little as if there's an increase in juice that was being used for the processor. Now the yellow light just flashes every few seconds. Once at this point, unless I restart everything, it won't work on any changes. I'm at a loss... please help.

Honestly looking for some help please.

Whenever network diff changes  new work gets sent to the ASIC and its restarted, sounds like your USB port/Hub is not providing enough power. When new work is sent to the ASIC is when it has the highest chance to get stalled because of power transients the hub can't keep up with.

Beta testing of my new update that resolve this issue and similar issues that cause the sticks to go dead and need restarting is going well and should have a binary release tomorrow. This should also keep sticks alive from poorly tuned/unstable voltages etc.

Pic below is a bunch of sticks running the new driver that I have purposely undervolted so they are unstable....driver has kept them alive and hashing for over a day now with no manual restarts needed!



changes are already pushed to my GitHub for anyone that dosent want to wait and can compile on their own.

[dem0nlord]

changes are already pushed to my GitHub for anyone that dosent want to wait and can compile on their own.

Thanks! New code running here.
https://preview.ibb.co/jdK3aG/moonlanders_new.jpg

[aarons6]

Alright, I've been playing with this thing all day. I've targeted it down this issue.

Code:

[2017-12-10 16:15:41] Network difficulty changed to 20.4k (146.3G)

Any time this happens the miner stops running. You can hear the fan wind up a little as if there's an increase in juice that was being used for the processor. Now the yellow light just flashes every few seconds. Once at this point, unless I restart everything, it won't work on any changes. I'm at a loss... please help.

Honestly looking for some help please.

Whenever network diff changes  new work gets sent to the ASIC and its restarted, sounds like your USB port/Hub is not providing enough power. When new work is sent to the ASIC is when it has the highest chance to get stalled because of power transients the hub can't keep up with.

Beta testing of my new update that resolve this issue and similar issues that cause the sticks to go dead and need restarting is going well and should have a binary release tomorrow. This should also keep sticks alive from poorly tuned/unstable voltages etc.

Pic below is a bunch of sticks running the new driver that I have purposely undervolted so they are unstable....driver has kept them alive and hashing for over a day now with no manual restarts needed!



changes are already pushed to my GitHub for anyone that dosent want to wait and can compile on their own.

since you used the same driver name, is the v1 working with this version?
edit nevermind i see you used a different branch name..

is this right?
git clone -b futurebit2_driver https://github.com/jstefanop/bfgminer

edit.. it took forever to compile on the rpi.. it seems to be working tho. ill see if they need restarted.

[dem0nlord]

edit.. it took forever to compile on the rpi.. it seems to be working tho. ill see if they need restarted.

Not sure what options you are using for configure, but probably compile fastest with:
./configure --disable-other-drivers --enable-scrypt --enable-futurebit --disable-sha256d
Also you're more than likely using a quad core rpi so you should do make -j4
My single core atom does the make in 51 seconds, I bet you can do it faster than that.

[Astrali]

i compiled with extra options --enable-scrypt --enable-futurebit
and left all other drivers included.

it compiled on my rpi within ~10 minutes i would think (without using multiple cores).


if you ever compiled a kernel, or an android build .. 10 minutes is just a blink of the eyes
and you are not supposed to do it daily.

the new version works good so far, but i guess my 3A psu is just to small for my 2 moonlanders (maybe low efficiency).
running them on ~4-4.2 mh/s and they get in trouble every now and then when diff changes or pool gets changed.

Thanks for the nice work @jstefanop

[thebosz]

Whenever network diff changes  new work gets sent to the ASIC and its restarted, sounds like your USB port/Hub is not providing enough power. When new work is sent to the ASIC is when it has the highest chance to get stalled because of power transients the hub can't keep up with.

Beta testing of my new update that resolve this issue and similar issues that cause the sticks to go dead and need restarting is going well and should have a binary release tomorrow. This should also keep sticks alive from poorly tuned/unstable voltages etc.

Pic below is a bunch of sticks running the new driver that I have purposely undervolted so they are unstable....driver has kept them alive and hashing for over a day now with no manual restarts needed!



changes are already pushed to my GitHub for anyone that dosent want to wait and can compile on their own.

Woo-hoo! I can't wait!

You're awesome jstefanop!

I've already ordered 2 more in-stock units and 3 more preorders for the next batch! These are great!

[hilti30]

Hi i'm total newb and not sure if i am doing it right but can't get this to work on my mac followed the work around but when i try to edit the start-moonlander-2 file with my pool details it says its read only and won't allow me to change it.

So how do i enter my details?

Thanks

[RdStrcklnd]

Linux question - Once the new driver is released for Pis, what specifically do we need to type to install and use it?

[sunk818]

What macOS version are you using?

How are you editing the file?

In Finder, go to the folder, click on the file, and do Command+I (or right-click Get Info). See what your Sharing & Permissions is set at. You (me) should have Read & Write privilege.

Make sure you quit out of the script (terminated) is not being used (running).

Hi i'm total newb and not sure if i am doing it right but can't get this to work on my mac followed the work around but when i try to edit the start-moonlander-2 file with my pool details it says its read only and won't allow me to change it.

So how do i enter my details?

[sunk818]

I'm used to cgminer where the accepted share is a multiplication of the share difficulty. Is it me, or is bfgminer weird about how it displays accepted shares? It shows it as one share regardless of difficulty. But then with rejects and hardware errors, I'm not really sure what the numbers really mean.

Is there a way to fix this, or is this just wonky programming by bfgminer?

Also, adjunct... --shares 1000 parameter doesn't seem to work. When I quit with 800 accepted shares, it tells me only 21 shares were accepted before quitting.