Bitcoin Forum
December 15, 2017, 08:15:03 PM *
News: Latest stable version of Bitcoin Core: 0.15.1  [Torrent].
 
   Home   Help Search Donate Login Register  
Pages: « 1 ... 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 [59] 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 »
  Print  
Author Topic: [Guide] Dogie's Comprehensive ASICMiner Blade Setup  (Read 579615 times)
btc-diggler
Newbie
*
Offline Offline

Activity: 2


View Profile
December 05, 2013, 03:53:23 AM
 #1161

Could anyone please give ideas about troubleshooting my issues? I've been reading these threads for days. Other people seem to have similar issues but aren't reporting if/how they solve.

Rig
1 asicblade (I have 2 but trying to debug with one) v2 new from jones gear
600w power supply, wired correctly
4 fans per blade
Rasp Pi - Debian with stratum proxy
All network connections are static ip and hard wired straight to the router.

Issues
Under performance
Blade GUIs do not like taking the same server ip and pw twice. Some times random settings appear.
Blades regularly want to switch servers.

When I was running both blades, only got 2.5 Gh consistently each blade, on eligius. I have seen each blade peak before near 7 Gh. I'm currently only running one blade now, on eligius with stratum proxy and only getting 1,733 - 2,500 Mh. Same results on slush pool, if not worse.

Checking temp with laser thermometer I get 80f on heat sink and around 100-111f on backside. Any idea how hot these things want to run? My CPU regularly runs hotter than that. The Rpi is not stressed. No IP address conflicts on network.

Is there another place for information on this setup? I'm going crazy reading these threads, and reading solutions and suggestions out of order. I will document this fix in a clear manner with hopes to help others who seem to be having similar issues.

My next try is to switch to BFGminer for a proxy, but will take more time than I have available in the next few days.

Thanks

Have you tried using a PC running the proxy instead of the Rpi? Maybe it's me, but I've always thought using an Rpi for blades is kinda "advanced". I'd try the dumbest config possible just to ensure the blade isn't defective, grab a crossover cable and hook the blade into a laptop with WiFi, bridge the connections and it should just do it. If that works at least you know the blade is good...

Thanks for the suggestion. I came here to keep my word and update what the problem was...

I installed a new router for the entire network, and installed a switch for just the mining rig and solved all issues. Pi and 2 blades on eligius hashing over 20Gs as we speak. I had to hard reset one blade to get settings to stick but it's all good now.

So glad I don't have to make a new vm to test another setup or install from scratch and try bfgminer.

Thinking back I probably should have replace the router when the dog pulled if off of the shelf. I wonder what other issues I might have just fixed.

Thanks all previous posters for the knowledge and thanks forum for letting me vent. Happy hashing.
1513368903
Hero Member
*
Offline Offline

Posts: 1513368903

View Profile Personal Message (Offline)

Ignore
1513368903
Reply with quote  #2

1513368903
Report to moderator
1513368903
Hero Member
*
Offline Offline

Posts: 1513368903

View Profile Personal Message (Offline)

Ignore
1513368903
Reply with quote  #2

1513368903
Report to moderator
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction. Advertise here.
1513368903
Hero Member
*
Offline Offline

Posts: 1513368903

View Profile Personal Message (Offline)

Ignore
1513368903
Reply with quote  #2

1513368903
Report to moderator
Theo1355
Jr. Member
*
Offline Offline

Activity: 48


View Profile
December 07, 2013, 05:19:16 PM
 #1162

Okay, here's the problem.  I got my 2x v2 blade rig going with 4 high performance fans right on the heatsinks.  I've got more psu than I need and everything was working super well for 24 hours.  However, when I restarted it after an hour or so of inactivity, it can't connect to eligius via getwork as before.

All chips are registering okay, all zeros under chips on the config page.
The pool address and my login info are both inputted twice, getwork.mining.eligius.st,getwork.mining.eligius.st,
Port: 8337-- this should be the current one, it's on the eligius webpage.
My local network must be alive cause I can reach the config page.

The only thing registered is the run time, which goes up to around 2min15secs before the blade automatically restarts.

I've sifted through most of this thread to find out what might be the issue but this has proved unfruitful.  

Does anyone know what might be the issue?
rekweam
Jr. Member
*
Offline Offline

Activity: 33


View Profile
December 07, 2013, 05:39:58 PM
 #1163

maybe you guys can point me in the right dirrection, i finally got the board hashing on a proxy by

using mining proxy.exe  with the "-o mint.bitminter.com -p 3333" argument according to the gui i am hashing about 8000MHS but bitminter is showing under 1gps so about 700 difference  Huh




do i have the port set right in the GUI? it is the only one i can get to work, i have tried 3333 and 5050 but it just stalls my proxy

Bitminter also outputs an estimation based on number of proofs returned. So what you see on their site is a ruff idea of what your hashing. I have 10 blades running and most of the time bit minter shows me 105-120ghps when i am usually doing a solid 111ghps.
dogie
Legendary
*
Offline Offline

Activity: 1456


dogiecoin.com


View Profile WWW
December 07, 2013, 08:05:54 PM
 #1164

Added pinout info to OP:

My cables are braided or aren't color coded: See below diagram for pin-outs.


bliskars
Newbie
*
Offline Offline

Activity: 17


View Profile
December 07, 2013, 10:42:31 PM
 #1165

Hey dogie wondering if you could help me out.

I have a 6 pin PCI-e to 4 pin molex connecter. Looking at your guid and drawings I figured I need to plut in 2 yellow cables and 4 black cables. If I only have 3 black cables is that going to be a problem since its a 6 pin connector and not an 8?

Whats the best way to plug it in that way?
dogie
Legendary
*
Offline Offline

Activity: 1456


dogiecoin.com


View Profile WWW
December 07, 2013, 11:22:44 PM
 #1166

Hey dogie wondering if you could help me out.

I have a 6 pin PCI-e to 4 pin molex connecter. Looking at your guid and drawings I figured I need to plut in 2 yellow cables and 4 black cables. If I only have 3 black cables is that going to be a problem since its a 6 pin connector and not an 8?

Whats the best way to plug it in that way?
6 pin isn't great, and especially STILL COMING FROM ONE MOLEX. Use your 4 pin ATX, they're designed for that current typically.

bliskars
Newbie
*
Offline Offline

Activity: 17


View Profile
December 07, 2013, 11:29:22 PM
 #1167

4 Pin atx has less cables right? I saw from you picture you did the following config:

1 yellow
1 yellow
2 black
2 black

with 4 atx I can only do

1 yellow
1 yellow
1 black
1 black

and with 6 pin to 4 molex
1 yellow
1 yellow
2 black
1 black

Isn't 6 pin better?

I guess I can just chop the power supply but it has 2 8 pins. Any idea how I should go about it? Heres the power supply I have. I have to power 3 blades

http://www.amazon.com/Corsair-430-Watt-Certified-Compatible-Platforms/dp/B004W2T2TM

matt4054
Legendary
*
Offline Offline

Activity: 1428


BitcoinQueue.com


View Profile WWW
December 07, 2013, 11:32:32 PM
 #1168

Hey guys,

I have a couple of Blades running for the past few months and I've been tweaking them a bit when running into issues with the chips, i.e. when I was getting one or more X in the "Chip: ..." diagnostics line.

One single X (might be XX), or a couple of them scattered among O's:

Probably heat related. Try power down, 5 min cool down and restart.
Your Blade may require more cooling.

XXXX or XXX (might be XX), sticky after power cycle

Probably voltage related. Locate the corresponding power lane, and use a multimeter to adjust the voltage to the required level (1.06 for low clock, 1.2 for high clock)

All XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

If 5 min cool down doesn't help,

- check PSU voltage under load (should be near 12V)
- check power lanes voltage (should be 1.06 or 1.2V)
- if low clock is OK but not high, your power lanes may need 1.2 instead of 1.06 V
sidehack
Legendary
*
Online Online

Activity: 1540

Curmudgeonly hardware guy


View Profile
December 08, 2013, 01:15:28 AM
 #1169

Quote
The only thing registered is the run time, which goes up to around 2min15secs before the blade automatically restarts.

I've sifted through most of this thread to find out what might be the issue but this has proved unfruitful. 

Does anyone know what might be the issue?

Did you try hitting "Switch server"? Sometimes these guys get locked and won't connect until you tell it to look at the secondary. Happened to me several times.

DKO
Newbie
*
Offline Offline

Activity: 13


View Profile
December 08, 2013, 05:09:54 AM
 #1170

Hello I wanted to thank dogie and many other members on this site, dogie's Blade Setup Guide helped me out especially.

I'll show off my rig if I may.

I have three ASICminer Blades setup at the moment powered by a Thermaltake 550w PSU.

Four 120mm fans. Two on the front sucking air in and two on the back blowing air out.

Fan controller with thermistats.

I used an old PC case with conduit to mount the blades. The side panels stay on the case usually.

https://dl.dropboxusercontent.com/u/90286356/2013-12-07%2020.41.39.jpg

https://dl.dropboxusercontent.com/u/90286356/2013-12-03%2017.38.20.jpg

https://dl.dropboxusercontent.com/u/90286356/2013-11-30%2021.11.06.jpg

This photo was taken before I used the conduit mount.

https://dl.dropboxusercontent.com/u/90286356/2013-12-03%2017.36.57.jpg

I have two more 120mm fans on order which i'm going to mount on the side panels.

My Raspberry Pi arrived today, might use that for the proxy. Also thinking about buying three more blades.

Overall I've had no hardware problems at all. One blade has been running 24/7 for about five weeks now and the other two for three weeks.
pluMmet
Sr. Member
****
Offline Offline

Activity: 279



View Profile
December 08, 2013, 11:35:05 AM
 #1171

My Blades (2 of them) stopped working 2 days ago and I have no idea why.

On the Config screen I have no "x"'s and it's been great until now.

My whole system is on a UPS system even the cooling so I know it's never gotten hot and I'm stuck on why this has happened.

I've been using them with getwork which I keep seeing people trash but that is why I bought them and the rest of my equipment is utilized anyway.

I had an issue a week or so ago due to eligius going down and me scrambling to get the blades on any getwork system. eventually I got it back on eligius and it had been fine until 2 days ago.

I now have to divert a cpu and finally got them working but only using proxy on Slushs pool and that was 2 days ago (would not work on my preferred pool or the 2 others I tried).. I just found out that after initially reporting my blades as generating shares slushes pool though accepting hashes from me the whole time is not reporting them as credit to me and though my Blades are working fine right now Slushes pool says 0/mh. They were reporting fine when I final got them running 2 days ago and it looks like within 2 or 3 hours the pool just started not noticing that my workers have been doing their work.

checking on my workers on eligius is a great process on slushes pool its a nightmare just logging in.

I'm at my wits end with these Blades... They only work on 1 pool now and that pool in my mind is just horrible.

Right now I can put them to hash on a getwork pool and they do nothing..I put them back to mine proxy on Slushes pool and they are hashing away... but are not showing up on slushes pool at all.

Resetting my proxy terminal might get slushes pool to recognize but this is the reason I left that pool to begin with.

The proxy for what ever reason will not accept my doing stratum on another pool it just ignores my "-"flags. I have no choice in the matter.

WTF do I do?



Oh and also they hashed @ 10-11/mh on getwork and now 8/mh with proxy so boooo to you getwork haters...

center]
▄█████
▄█████████████▄
▄█████████████████████▄
▄█████████████████████████████▄
███████████████████████████████████
█████████████▀       ▀██████████████
██████████▀   ▄▄▄▄▄▄▄  ▀███████████
██████████    █████████    ██████████
█████████    ███▀▀▀▀▀▀▀     █████████
████████     ███▄▄▄▄▄▄       ███████
███████      █████████      ███████
███████       ▀▀▀▀▀▀███     ████████
█████████     ▄▄▄▄▄▄▄███    █████████
██████████    ████████    ██████████
███████████▄  ▀▀▀▀▀▀▀▀   ██████████
██████████████▄       ▄█████████████
███████████████████████████████████
▀█████████████████████████████▀
▀█████████████████████▀
▀█████████████▀
█████▀
   
 
.The Issuance of Tokenized Equity.
▬▬▬ ●  Twitter  Facebook  |  Telegram  ● ▬▬▬
 
 
[/center]
sidehack
Legendary
*
Online Online

Activity: 1540

Curmudgeonly hardware guy


View Profile
December 08, 2013, 03:22:25 PM
 #1172

What's your command line look like for the proxy?

pluMmet
Sr. Member
****
Offline Offline

Activity: 279



View Profile
December 08, 2013, 04:11:35 PM
 #1173

What's your command line look like for the proxy?


I wanted it to use Eligius so I used (Dogies suggestion): mining_proxy.py -o mining.eligius.st -p 3334 -cu 1yourSweetCoinsComeToThisAddress -cp x

Not only did it not work the terminal ignored that and was trying to connect to Slushs pool.

I ended up just doing: mining_proxy.py which had the slushs stuff already going and the blades synced up right away.

It looks like the proxy.py has been nurfed to just use slushes pool. I d/l'd two days ago for the slushes pool site and it's not .tar anymore but .zip. Still linux of course.. well py anyway.

I have them running now even though slushs pool is not awarding me anything.

Beyond that this is ridiculous. I had been using getwork just fine for months.. I obviously know how to do that but the blades just won't do it.

v2 blades btw.

center]
▄█████
▄█████████████▄
▄█████████████████████▄
▄█████████████████████████████▄
███████████████████████████████████
█████████████▀       ▀██████████████
██████████▀   ▄▄▄▄▄▄▄  ▀███████████
██████████    █████████    ██████████
█████████    ███▀▀▀▀▀▀▀     █████████
████████     ███▄▄▄▄▄▄       ███████
███████      █████████      ███████
███████       ▀▀▀▀▀▀███     ████████
█████████     ▄▄▄▄▄▄▄███    █████████
██████████    ████████    ██████████
███████████▄  ▀▀▀▀▀▀▀▀   ██████████
██████████████▄       ▄█████████████
███████████████████████████████████
▀█████████████████████████████▀
▀█████████████████████▀
▀█████████████▀
█████▀
   
 
.The Issuance of Tokenized Equity.
▬▬▬ ●  Twitter  Facebook  |  Telegram  ● ▬▬▬
 
 
[/center]
Theo1355
Jr. Member
*
Offline Offline

Activity: 48


View Profile
December 08, 2013, 08:50:36 PM
 #1174

What's your command line look like for the proxy?


I wanted it to use Eligius so I used (Dogies suggestion): mining_proxy.py -o mining.eligius.st -p 3334 -cu 1yourSweetCoinsComeToThisAddress -cp x

Not only did it not work the terminal ignored that and was trying to connect to Slushs pool.

I ended up just doing: mining_proxy.py which had the slushs stuff already going and the blades synced up right away.

It looks like the proxy.py has been nurfed to just use slushes pool. I d/l'd two days ago for the slushes pool site and it's not .tar anymore but .zip. Still linux of course.. well py anyway.

I have them running now even though slushs pool is not awarding me anything.

Beyond that this is ridiculous. I had been using getwork just fine for months.. I obviously know how to do that but the blades just won't do it.

v2 blades btw.

Dude, I'm having the same problem with getwork-- it's so frustrating, especially when I didn't change anything, the blades just stopped connecting.

Let me know if you get it figured out; I'll reciprocate if I discover what's happening.
pluMmet
Sr. Member
****
Offline Offline

Activity: 279



View Profile
December 08, 2013, 11:06:23 PM
 #1175

What's your command line look like for the proxy?


I wanted it to use Eligius so I used (Dogies suggestion): mining_proxy.py -o mining.eligius.st -p 3334 -cu 1yourSweetCoinsComeToThisAddress -cp x

Not only did it not work the terminal ignored that and was trying to connect to Slushs pool.

I ended up just doing: mining_proxy.py which had the slushs stuff already going and the blades synced up right away.

It looks like the proxy.py has been nurfed to just use slushes pool. I d/l'd two days ago for the slushes pool site and it's not .tar anymore but .zip. Still linux of course.. well py anyway.

I have them running now even though slushs pool is not awarding me anything.

Beyond that this is ridiculous. I had been using getwork just fine for months.. I obviously know how to do that but the blades just won't do it.

v2 blades btw.

Dude, I'm having the same problem with getwork-- it's so frustrating, especially when I didn't change anything, the blades just stopped connecting.

Let me know if you get it figured out; I'll reciprocate if I discover what's happening.

Will do, thanks.

It's a serious issue as I've tried lots of getwork pools and the blades don't work on any of them.

center]
▄█████
▄█████████████▄
▄█████████████████████▄
▄█████████████████████████████▄
███████████████████████████████████
█████████████▀       ▀██████████████
██████████▀   ▄▄▄▄▄▄▄  ▀███████████
██████████    █████████    ██████████
█████████    ███▀▀▀▀▀▀▀     █████████
████████     ███▄▄▄▄▄▄       ███████
███████      █████████      ███████
███████       ▀▀▀▀▀▀███     ████████
█████████     ▄▄▄▄▄▄▄███    █████████
██████████    ████████    ██████████
███████████▄  ▀▀▀▀▀▀▀▀   ██████████
██████████████▄       ▄█████████████
███████████████████████████████████
▀█████████████████████████████▀
▀█████████████████████▀
▀█████████████▀
█████▀
   
 
.The Issuance of Tokenized Equity.
▬▬▬ ●  Twitter  Facebook  |  Telegram  ● ▬▬▬
 
 
[/center]
Theo1355
Jr. Member
*
Offline Offline

Activity: 48


View Profile
December 08, 2013, 11:16:32 PM
 #1176



Will do, thanks.

It's a serious issue as I've tried lots of getwork pools and the blades don't work on any of them.

Totally-- the weird thing is they were designed for getwork.  I even did a factory reset, reconfigged and it still doesn't work.  I'll maybe restart my router, and then I'll go to stratum though I'm not optimistic.
sidehack
Legendary
*
Online Online

Activity: 1540

Curmudgeonly hardware guy


View Profile
December 08, 2013, 11:26:11 PM
 #1177

Have you tried configuring worker/pass in the blade config and not using -cu -cp? Have you tried changing the getwork port -gp in case something on your local machine is using it?

Theo1355
Jr. Member
*
Offline Offline

Activity: 48


View Profile
December 08, 2013, 11:31:21 PM
 #1178

Have you tried changing the getwork port -gp in case something on your local machine is using it?

The 'Web Port' that's usually 8000?
sidehack
Legendary
*
Online Online

Activity: 1540

Curmudgeonly hardware guy


View Profile
December 08, 2013, 11:45:24 PM
 #1179

No, the getwork port that is usually 8332. The -gp option in mining_proxy changes this, and it needs to match the blade's "Pool ports" setting.

Theo1355
Jr. Member
*
Offline Offline

Activity: 48


View Profile
December 08, 2013, 11:47:31 PM
 #1180

No, the getwork port that is usually 8332. The -gp option in mining_proxy changes this, and it needs to match the blade's "Pool ports" setting.

Okay, do you still have to input twice (I assume so?)
Pages: « 1 ... 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 [59] 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 »
  Print  
 
Jump to:  

Sponsored by , a Bitcoin-accepting VPN.
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!