Stratum Proxy Issue/Asicminer Erupter Blade Resarts...

[mikewatson3]

Hi all,
I just got an Asicminer Block Erupter Blade.
I setup the Blade according to dogie's Guide , and got the Blade hashing on Slush's pool using the Stratum Proxy. The Blade would hash at the max rate of 10g/hs. After about 6 minutes, the Blade would Idle/Stop and then Restart and start over hashing. The pool meter would only top out at 6.9g/hs. with the constant restarting. I used a different power supply, made sure the Neutrals were bonded together. Provided enough cooling, etc. I sent an email to Asicminer about a RMA for the Blade thinking that it was bad, and they stated that the Blade restarts because there isn't enough work being sent to it. So I changed to a differnt pool that supported the get work protocol. After connecting, the Blade has been hashing at 10g/hs. and has had NO restarting issues. My question is how can I increase the workload on the Stratum Proxy so that the Blade hashes at the max rate? Is there a command that I can enter in order to get the work faster?? I want to use Stratum but it sucks at the cost of almost 4g/hs, its almost worth it to mine on a pool that supports the get work protocol...like btcguild and pay the 7.5% fee...
Thanks!

[briceperdue]

I have just recently changed my blades over from getwork and btcguild to using bfgminer and stratum.  I have 6 of the old model blades all now overclocked and stable at 12gh\s and 3 of the new blades like 10.5\ghs.  When you log into your gui for blade what is the per minute number?

[mikewatson3]

I have just recently changed my blades over from getwork and btcguild to using bfgminer and stratum.  I have 6 of the old model blades all now overclocked and stable at 12gh\s and 3 of the new blades like 10.5\ghs.  When you log into your gui for blade what is the per minute number?

Briceperdue,
I have the v2 new version, when I check the configuration page, it currently says 143.75 per minute, and that's with it connected using the get work protocol. Don't know what the per minute is when I was connected with Stratum.

[dellnull]

Any luck with this? Do you mine with your blade right now and if so, what was the solution?

I got almost the same error. But my blade (that arrived today and is identical to yours,  AM BE rev 2.01) won´t even start to mine and then restart after 2m 15 sec. in an eternal 2.15 loop. (both ping and http dies and for about 3-4 sec followed by flashing led's on the board during reboot, EVERY 2.15 min)

I have plenty of cooling and also plenty of power (I hope... 300w for a single blade). I have to yellow and two blacks, just as a guide I followed showed me. The heatsink gets 32.1 and under the heatsink the chips gets about 31,4 C (with cooling on the heatsink side). With the fans switched of I can feel that the temp. raises but never checked with at thermometer without cooling, to scared to melt something :-)

This is how the config page looks like:

Total MHS:   00000
Received:   0000000000
Accepted:   0000000000
Per Minute:   000.00
Efficiency:   000.00%
Up Time:   0d,00h,01m,25s

Current Server: 192.168.0.221:8332
Chip: OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO

IP   192.168.0.151
Mask   255.255.255.0
Gateway   192.168.0.1
WEB Port   80
Primary DNS   192.168.0.1
Secondary DNS   8.8.8.8
Ports   8332,8332
Server addresses   192.168.0.221,192.168.0.221
user:pass   any.worker4:whatever,any.worker4:whatever

Anyone have an idea what's going on? I have tried stratum mining_proxy.exe from slush both on btcguild and at slush's pool and not a singel MH/s is produced yet, just the annoying restarts every 2.15 min......

Are the blade broken in som way? Not so fun to send it back from Europe and the wait for 2 month to get a new one, the ROI was bad enough if it had worked from today :-)

//DN

[7queue]

How proficient are you with test equipment? can you verify the blades are not dragging the power supply down?

You need sufficient amperage with these blades, if the power supply is weak it will dump and recover, not totally go.

Use a volt meter to read the voltage, if it fluctuates it is oscillating in recovery mode.

P.S. Don't open a PC power supply unless you absolutely know what you are doing!!!!!!!

Did I mention DO NOT OPEN A PC POWER SUPPLY UNLESS YOU ABSOLUTELY KNOW WHAT YOU ARE DOING!!!

Other wise I would check your cables and connectors.

I pulled the github source and compiled so I haven't had any trouble with software.

I use linux most of the time, I could try and build on MS and see what happens, it is a pain!

next thing I would look at is the network, is it isolated to the miners? or is it on the 'open net'?

I like to segment things.

I stand behind these blades, they are easy to use!

Besides, I tell all my Friends to just 'buy' bitcoins cause I hate tech support...

8 )

[Death2142zerg]

I am getting the same problem as dellnull. When i have my blades hooked up they don't hash no matter what place i use (BTCGuild, Slush). I have tried the mining_proxy.exe as well but the blades still wont connect. only with mine it restarts after 2 mins and 23 seconds.

Total MHS:   00000
Received:   0000000000
Accepted:   0000000000
Per Minute:   000.00
Efficiency:   000.00%
Up Time:   0d,00h,00m,22s

Current Server: btcguild.com:3333
Chip: OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO

IP                 192.168.1.200
Mask                   255.255.255.0
Gateway           192.168.1.1
WEB Port           8000
Primary DNS   192.168.1.1
Secondary DNS   211.148.192.141
Ports                   3333,3333
Server addresses   btcguild.com,btcguild.com    (i have tried ip's, stratum.btcguild.com, localhost, and several others i cant think of at the moment)
user:pass           ausername_1:123,ausername_1:123

any and all help in fixing this would be greatly appreciated.

[7queue]

OK I'll pull the windohs/nsa binaries and give them a try. I only use windohs when I absolutely have to for clients.

if you have more than one blade each one has to have a different username and the dual user/pass is for failover which I take advantage of. I have even ip addy's on one bfgminer and odd on another bfgminer just in case one of the laptop dies (which one did but I have a pile of used so meh)

you also have to point the blades at the miner/proxy not the pool, I'll see if I can dig up the docs but I've been reading a lot but always save docs. Slush's explanation on how to setup the blades is good, re-read it while I look.

Once I can I will also document my setups for you all.

I use multiple 100Amp+ 12V power supplies on 220 mains/solar per 8 blades so there is no stressing the supplies, yes it's over kill but why not?

8 )

[7queue]

I don't know what MS just did but the latest Updates just trashed all my Windohs systems! I didn't have a choice it just updated! I want choice!

I pulled slush's latest windohs binary and it faults on my systems, so I can't test anything.

I seriously advise using a linux install for the blades, I will help you with that, it is easy, much easier than Windohs

8 )

[7queue]

I haven't given up, I'm trying to figure out this MS way of running the blades

besides MS is a pain, the binary keeps faulting so I don't know what you are getting.

[Death2142zerg]

I am currently trying something, i would try Linux but i don't have one and have never used one only windows and mac but i only own windows.

Do blades accept: "localhost" as a valid address or does it need to be an ip?

[7queue]

the blades need to point to the ip of the system the proxy is on so no

I've run into a few obstacles with MS, if you haven't tried ubuntu it is the easiest and is at no cost

this perplexes me and is a challange! I except!

I still like these blades over anything else, and I design chips...

8 )

[Death2142zerg]

I got my blades to run they are now running an avg of 10 GH/s.  One of the problems was i had the gate way set to 192.168.1.1 when for me it needed to be 192.168.0.1, but the biggest thing was the IP address thing where it couldnt be local host which is fixed now and it is working i will post what it is now, so that it may hoppfuly help someone else in the future. And thanks 7queue for your help with out you i wouldn't have fixed the problem.


Total MHS:   10856
Received:   0000002336
Accepted:   0000002357
Per Minute:   149.65
Efficiency:   100.89%
Up Time:   0d,00h,15m,45s

Current Server: 192.168.1.14:8332
Chip: OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO

IP   192.168.1.201
Mask   255.255.255.0
Gateway   192.168.0.1
WEB Port   8000
Primary DNS   192.168.1.1
Secondary DNS   211.148.192.141
Ports   8332,8332
Server addresses    192.168.1.14,192.168.1.14
user:pass            username_1:123,username_1:123

the thing i am still trying to do is set the ports to 3333 because the proxy program doesnt agree with 3333 only 8332.

[7queue]

The blade needs(s) to point to the mining software that is on the PC, make sure port 8332 inbound is open in the windows firewall and/or your antivirus, that is the getwork protocol.

Then the mining software will point to the pool outbound on port 3333 that is the stratum protocol.

I got the mining software to work on windohs 7.

if you open a command prompt and issue: ipconfig /all
it will show you your network settings allong with the gateway and dns servers.

[hatelabs]

I had the same issue,.. I'm very hardware savvy, but I bought the backplane and power supply anyway (just to make sure i had no wiring issues)

Plug in, setup proxy and NOTHING. But,.. if I change the port to 3333 the proxy gives me all the appropriate erorrs so I KNOW the blade is trying to talk to the proxy.

Everyone said try slushs's so I did and zero,... Reconfiged for eclipse mining consortium and BINGO, (or so I thought) MH/s would ramp up but only for a few seconds and die... Watching the proxy server it would do a burst of requests, then sit idle (and thus trip the dreaded 2:20 reset timer.

I pointed the blades at BitMinters port 80 and boom they all came up mining at 2-5GH/s each...
Obviously I want them to max out, but at this point I've proven that "the cards are good" and that I can get data in/out from the net.

I'm going to keep working on this and see what I can finally come up with to make these work ok, but my guess is that it's a network/firewall proxy issue, as port 80 is pretty much going to be allowed by everything.

Oddly enough I tried it on my home network (crappy DSL modem) and had the same issues different proxy love and 80 still working so,... that's not a 100% sure answer, but hopefully it will get a few of you to "the next step" (ie where I am now)

Anyway if you're having the 2min 20second, 30 second reset issue on your block erupter blade, I hope this will at least get you one troubleshooting step further.


LTC:LSTjG357aKAwJEMFzNUAm2RY4wyUyvKL59
BTC:19h2PkTKsPcMxtwGspcpnQSEuVCefSBjmp