[Guide] Dogie's Comprehensive ASICMiner Cube Setup [HD]

[cdogster]

Can someone help me debug a restarting cube?

Every 8 minutes or so (8 minutes and 40 seconds it seems), the cube will restart automatically and the effect is an overall lower hashing rate.  I have tried running the proxy on 2 different machines but I get the same results.  I notice a lot of warning messages in the proxy logs and BTCGuild will estimate my hashing at about 24GH/s.  The heatsinks are all tight, it's in a well ventilated area and all looks good except for the "Started before:" time never going that much higher than 8 minutes.  I've created brand new workers and done a factory reset.  What can I try next?  

For the proxy, I am running it with:

Code:

mining_proxy.exe -o stratum.btcguild.com -p 3333

Proxy logs:

Code:

2013-12-23 10:26:02,490 WARNING proxy getwork_listener._on_submit # [0ms] Share from 'btcguild_5' accepted, diff 32
2013-12-23 10:26:02,506 INFO proxy getwork_listener._on_authorized # Worker 'btcguild_5' asks for new work
2013-12-23 10:26:02,515 INFO proxy jobs.submit # Submitting a895e301
2013-12-23 10:26:02,515 WARNING proxy getwork_listener._on_submit # [0ms] Share from 'btcguild_5' accepted, diff 32

Cube config:
http://imgur.com/8PpXxI5

I am using the Corsair CX500M which should be plenty of power for one cube and it is always cool to the touch.  Setup: http://imgur.com/VZf7i0d

Someone please help me.  There has to be a reason why this cube is restarting every 8 minutes and 40 seconds.  After each restart, it will hash away just fine until about an hour or so later it will turn all chips to X's where I can power cycle the cube to return it to all O's and the 8 minute cycle.  Any ideas?

[1714733916]

1714733916

[jesse11]

Can someone help me debug a restarting cube?

Every 8 minutes or so (8 minutes and 40 seconds it seems), the cube will restart automatically and the effect is an overall lower hashing rate.  I have tried running the proxy on 2 different machines but I get the same results.  I notice a lot of warning messages in the proxy logs and BTCGuild will estimate my hashing at about 24GH/s.  The heatsinks are all tight, it's in a well ventilated area and all looks good except for the "Started before:" time never going that much higher than 8 minutes.  I've created brand new workers and done a factory reset.  What can I try next?  

For the proxy, I am running it with:

Code:

mining_proxy.exe -o stratum.btcguild.com -p 3333

Proxy logs:

Code:

2013-12-23 10:26:02,490 WARNING proxy getwork_listener._on_submit # [0ms] Share from 'btcguild_5' accepted, diff 32
2013-12-23 10:26:02,506 INFO proxy getwork_listener._on_authorized # Worker 'btcguild_5' asks for new work
2013-12-23 10:26:02,515 INFO proxy jobs.submit # Submitting a895e301
2013-12-23 10:26:02,515 WARNING proxy getwork_listener._on_submit # [0ms] Share from 'btcguild_5' accepted, diff 32

Cube config:
http://imgur.com/8PpXxI5

I am using the Corsair CX500M which should be plenty of power for one cube and it is always cool to the touch.  Setup: http://imgur.com/VZf7i0d

Someone please help me.  There has to be a reason why this cube is restarting every 8 minutes and 40 seconds.  After each restart, it will hash away just fine until about an hour or so later it will turn all chips to X's where I can power cycle the cube to return it to all O's and the 8 minute cycle.  Any ideas?

Do this: Proxy mining_proxy.exe -o stratum.btcguild.com -gp 8332

getWork ports are -gp 8332

[zeuzzz]

Hmm, stumped now. Maybe I should try a different pool? I know the password for multipool is not used, they generally say either leave blank or set to x, and I've tried it with both in the cube settings: sytherz.1:x,sytherz.1:x and also leaving the x out sytherz.1:,sytherz.1:

Happy xmas everyone

Anyone got any advice for the above? It's still failing even though I've done everything as advised here ...

Would people recommend I install CGminer or BFGminer instead to try through them? I've not compiled much on this laptop as I needed to use my main desktops PSU to power the cube ... but I could get round to it later.

[uozer]

changing the clock setting does not seem to impact anything at all, does anyone else have the same issue?

[Damnsammit]

Hmm, stumped now. Maybe I should try a different pool? I know the password for multipool is not used, they generally say either leave blank or set to x, and I've tried it with both in the cube settings: sytherz.1:x,sytherz.1:x and also leaving the x out sytherz.1:,sytherz.1:

Happy xmas everyone

Anyone got any advice for the above? It's still failing even though I've done everything as advised here ...

Would people recommend I install CGminer or BFGminer instead to try through them? I've not compiled much on this laptop as I needed to use my main desktops PSU to power the cube ... but I could get round to it later.

Merry Christmas!  The only thing that I can think of is that multipool doesn't support proxy mining... but that doesn't really make sense.

I'd try Slush's pool or BTC Guild and if that doesn't work then I am stumped as well.  Doesn't look like the proxy miner is connected to the pool though...

[Adamlm]

Yes, it looks exactly as proxy error. I've tested some pools and not all work properly with mining_proxy.exe. Here is a list which I maintain:

https://bitcointalk.org/index.php?topic=383137.0

[mogrith]

I have used my cube at multipool. But that was with BFGminer 3.8.0 on a RasPi

[mothman57]

Merry Christmas everyone!  I got a quick question for everyone.  I planned on leaving my desktop on at home running while I am away.  Am I able to hook up my ASICMiner Cube to my desktop via Ethernet and run my proxy through Wi-Fi?  If so, what settings do i need to change.  I have been trying for quite some time to make this work.  Thank you!

[daddyfatsax]

Anybody had an issue with fuses going bad? I had one melt on cube this morning. I took it apart, cleaned everything out, and then re-assembled it. Everything is working fine now.

[kern]

Great discussion! I have learned alot about this cube I just got and am trying to get setup. My problem is that my router is a Netgear WNDR3400v2 and every device on my home network has an IP address something like 10.0.0.1 and to make matters worse when I am on the status page of the router it recognizes all the devices on the network with the exception of the cube. Anyone else have experience with the Netgear network?
Thanks,
Kern

[stex2009]

My cube was giving problems and I spent lot of time to debug the whole thing. I had to run each board at a time and it worked. I realized it was just 1 bad board. Did this happen to anyone else?

[mothman57]

I believe it is a network issue.  Trying to connect direct to the motherboard via ethernet will require me to share my internet from the WI-FI card to the ethernet ports.  Figuring that out is the hard part.

[cdogster]

Can someone help me debug a restarting cube?

Every 8 minutes or so (8 minutes and 40 seconds it seems), the cube will restart automatically and the effect is an overall lower hashing rate.  I have tried running the proxy on 2 different machines but I get the same results.  I notice a lot of warning messages in the proxy logs and BTCGuild will estimate my hashing at about 24GH/s.  The heatsinks are all tight, it's in a well ventilated area and all looks good except for the "Started before:" time never going that much higher than 8 minutes.  I've created brand new workers and done a factory reset.  What can I try next?  

For the proxy, I am running it with:

Code:

mining_proxy.exe -o stratum.btcguild.com -p 3333

Proxy logs:

Code:

2013-12-23 10:26:02,490 WARNING proxy getwork_listener._on_submit # [0ms] Share from 'btcguild_5' accepted, diff 32
2013-12-23 10:26:02,506 INFO proxy getwork_listener._on_authorized # Worker 'btcguild_5' asks for new work
2013-12-23 10:26:02,515 INFO proxy jobs.submit # Submitting a895e301
2013-12-23 10:26:02,515 WARNING proxy getwork_listener._on_submit # [0ms] Share from 'btcguild_5' accepted, diff 32

Cube config:
http://imgur.com/8PpXxI5

I am using the Corsair CX500M which should be plenty of power for one cube and it is always cool to the touch.  Setup: http://imgur.com/VZf7i0d

Someone please help me.  There has to be a reason why this cube is restarting every 8 minutes and 40 seconds.  After each restart, it will hash away just fine until about an hour or so later it will turn all chips to X's where I can power cycle the cube to return it to all O's and the 8 minute cycle.  Any ideas?

Do this: Proxy mining_proxy.exe -o stratum.btcguild.com -gp 8332

getWork ports are -gp 8332

Thanks for the comment jesse11, but it's still not working for me.  In addition to passing the flag "-gp 8332" to the proxy exe, I have also tried passing the username/password as "-cu btcguild_5 -cp 123".  I have also tried disabling my firewall on my computer and on my router.  I have also changed my default gateway from 0.0.0.0 (default) to 192.168.1.1 (my router).

I can add a few more details...  After about 7 minutes and 30 seconds of hashing, the cube will stop asking for more work according to the proxy logs.  Then it restarts at 8 minutes and 40 seconds (almost exactly) and then it starts hashing away again.  After an hour, this restarting cycle will result in all chips turning to "X" where it requires me to power cycle the cube to begin hashing again.  The power supply and the cube are both cool to the touch the entire time.

Help!  What can I do?

[cdogster]

I had a very similar issue.  It ended up being network hardware related.  Try bringing your proxy machine and cube to a friend's house and connecting there.

Could you elaborate on the network hardware issue?

I am having a very similar issue and I have noticed that right before it automatically restarts, long polling will become inactive, then it restarts at almost 8 min 40 seconds exactly.  I don't see how a network hardware issue like a router could cause the cube to restart like that.

Could an ISP be causing this somehow?  I can try at a friend's house but if the ISP is the reason, then I won't be ruling that out.  Thanks for the help, much appreciated.

bark0de, did you fix your issue?

[Damnsammit]

cdog.... that sounds like such a bizarre issue.  I wonder if you have a faulty main board.

Have you already tried running it one blade at a time to see if it can stabilize with less than 6 blades?

I'd contact your supplier and see if they can get a replacement.  Very strange issue you are having

[MikeMike]

Great discussion! I have learned alot about this cube I just got and am trying to get setup. My problem is that my router is a Netgear WNDR3400v2 and every device on my home network has an IP address something like 10.0.0.1 and to make matters worse when I am on the status page of the router it recognizes all the devices on the network with the exception of the cube. Anyone else have experience with the Netgear network?
Thanks,
Kern

Make sure your phone etc are not connected to the wireless feature of the router. Then try again.
I find they like a rest after trying to connect with no success. Then try in 30-60 min.

[stex2009]

My miner runs perfectly fine on 'low'. As soon as I change to 'high', I get
ASIC_01-16: x x x x x x x x x x x x x x x x
ASIC_17-32: x x x x x x x x x x x x x x x x
ASIC_33-48: x x x x x x x x x x x x x x x x
ASIC_49-64: x x x x x x x x x x x x x x x x
ASIC_65-80: x x x x x x x x x x x x x x x x
ASIC_81-96: x x x x x x x x x x x x x x x x
Long Poll: inactive

suggestions? (I have enough power, get 420+ watts)

[Damnsammit]

My miner runs perfectly fine on 'low'. As soon as I change to 'high', I get
ASIC_01-16: x x x x x x x x x x x x x x x x
ASIC_17-32: x x x x x x x x x x x x x x x x
ASIC_33-48: x x x x x x x x x x x x x x x x
ASIC_49-64: x x x x x x x x x x x x x x x x
ASIC_65-80: x x x x x x x x x x x x x x x x
ASIC_81-96: x x x x x x x x x x x x x x x x
Long Poll: inactive

suggestions? (I have enough power, get 420+ watts)

As noted earlier in this thread, the wattage isn't nearly as important as the amps for the Cube.

How many amps does you 12V rail(s) get?

What kind of PSU are you using?

[stex2009]

My miner runs perfectly fine on 'low'. As soon as I change to 'high', I get
ASIC_01-16: x x x x x x x x x x x x x x x x
ASIC_17-32: x x x x x x x x x x x x x x x x
ASIC_33-48: x x x x x x x x x x x x x x x x
ASIC_49-64: x x x x x x x x x x x x x x x x
ASIC_65-80: x x x x x x x x x x x x x x x x
ASIC_81-96: x x x x x x x x x x x x x x x x
Long Poll: inactive

suggestions? (I have enough power, get 420+ watts)

As noted earlier in this thread, the wattage isn't nearly as important as the amps for the Cube.
How many amps does you 12V rail(s) get?
What kind of PSU are you using?

34 amp on 1 X 12V rail.
http://www.frys.com/product/6893426
Got this from my local frys and works perfect on low. (So it's not 420+ watts, its 408 watt max on 12v)

[mchu168]

My miner runs perfectly fine on 'low'. As soon as I change to 'high', I get
ASIC_01-16: x x x x x x x x x x x x x x x x
ASIC_17-32: x x x x x x x x x x x x x x x x
ASIC_33-48: x x x x x x x x x x x x x x x x
ASIC_49-64: x x x x x x x x x x x x x x x x
ASIC_65-80: x x x x x x x x x x x x x x x x
ASIC_81-96: x x x x x x x x x x x x x x x x
Long Poll: inactive

suggestions? (I have enough power, get 420+ watts)

As noted earlier in this thread, the wattage isn't nearly as important as the amps for the Cube.
How many amps does you 12V rail(s) get?
What kind of PSU are you using?

34 amp on 1 X 12V rail.
http://www.frys.com/product/6893426
Got this from my local frys and works perfect on low. (So it's not 420+ watts, its 408 watt max on 12v)

Take all the boards out, add them back one by one and find the bad board. I think this problem exists with 10% or more of all cubes shipped. At least that has been my experience running 35 Cubes. I think Asicminer ought to supply its resellers with ample stock of replacement boards to minimize the hassle and frustration this issue has surely caused.