Topic: [Guide] Dogie's Comprehensive Bitmain AntMiner S1 Setup [HD] - page 21. (Read 112382 times)

I just noticed that the one that isn't working is running cgminer 3.12.0-1 while the ones that are working are running 3.8.5-1 that has to be the problem right?  Do you know where I can download the older version and do you know of anything I need to be aware of when doing it?  I see the spot in software to download and install new packages, do I need to do anything else?

Thank you SO much for all of your help so far!

I am trying to flash the firmware from there but it's giving me this error:

The uploaded image file does not contain a supported format. Make sure that you choose the generic image format for your platform

I imagine I'm missing something obvious.

I tried both antMiner_openwrt20131226.bin and antMiner_openwrt20140207.bin and got the same error.

To me that looks like what happens when machine code is just derp'ed. Try resetting once more to factory defaults and entering stuff. If not, flash the latest firmware and try again. https://github.com/AntMiner/AntGen1/tree/master/firmware

Thank you so much for the reply!

I checked all of the WAN/LAN settings and they're identical to all of my other miners.  I went through just about all of the tabs to compare everything and it all looks the same to me.  Outside of MAC and ip address of course.  All of the other miners have different ip addys so it doesn't appear to be an ip conflict.

The only other thing that is different is that the local time on the bad miner reads:

Local Time   Mon Dec 2 11:45:25 2013

While on the good ones they read current Chinese time.  I don't know if that points to something?

The part that makes no sense to me is that I can mine just fine on the factory default pools but as soon as I change any of them to the pools I want to mine, it says they're dead.

To me that sounds like the machine at least works but I can't figure out where to go next for trouble shooting.

Do you have any other ideas?

Thank you again!

Edit:

I should add the system log has tons of these errors:

Mon Dec  2 11:48:20 2013 kern.err kernel: [ 1233.530000] SQUASHFS error: Unable to read page, block 206cc2, size f458
Mon Dec  2 11:48:25 2013 kern.err kernel: [ 1238.530000] SQUASHFS error: Unable to read fragment cache entry [206cc2]
Mon Dec  2 11:48:25 2013 kern.err kernel: [ 1238.530000] SQUASHFS error: Unable to read page, block 206cc2, size f458
Mon Dec  2 11:48:30 2013 kern.err kernel: [ 1243.560000] SQUASHFS error: Unable to read fragment cache entry [206cc2]
Mon Dec  2 11:48:30 2013 kern.err kernel: [ 1243.560000] SQUASHFS error: Unable to read page, block 206cc2, size f458
Mon Dec  2 11:48:35 2013 kern.err kernel: [ 1248.530000] SQUASHFS error: Unable to read fragment cache entry [206cc2]
Mon Dec  2 11:48:35 2013 kern.err kernel: [ 1248.540000] SQUASHFS error: Unable to read page, block 206cc2, size f458

Check the LAN/WAN connection settings, something might be freaky there. Could also be an IP conflict.

I'm having a bit of an odd problem.  I have setup 11 of them without any issue but I have one that is giving me all kinds of problems.  It connects to the factory default pool fine but when I try to connect it to my pools it won't connect to any of them.  If I leave a factory pool as pool 0 and fill mine in at pool 1 and 2 it says 1 and 2 are dead.  I have all of my other miners mining just fine on those pools so I don't believe they're down.  Anyone have any suggestions?

Thanks!

ah what can you do as long as small companies are building the gear it will continue to be this way.

 And to see my total hash at 650 not 830 ( i have some sticks plus the 3 good ants ) is frustrating. At least the other three seem good.  It could have been a lot worse.

Hopefully the other seller ships out the replacement this mon and I get it by weds.  

I totally understand, your situation is unfortunate, I was lucky enough to keep my machine hashing at a lower rate until the replacement arrived. I just figured as long as I was getting the btc back, I didn't mind the deposit, but I'll admit that it also made me triple-check that I was right, maybe that's part of their tactic. Seeing how skew (and loose) some of your soldering is, is also disappointing, the quality of the machines usually seems to be very good. I hope the rest goes well and you don't lose much more time.

yeah I suppose I understand your viewpoint but to ask 1.35 btc for the controller/ethernet card when the entire machine now sells for 1.37 btc is wrong.   once he asked that much to cross ship for a small part I considered my options and I fully checked the entire machine. after doing the very through check and finding the other issues it was obvious I need a complete replacement.  .    

 If he said to me I will send you a full machine do not bother with parts and I will cross ship it if you leave me a 1.37 btc deposit I would be fine with that.    Basically he failed by asking for 98 percent of an entire machine to replace a small part not very complicated.  BTW  I sent this to another s-1 dealer and he will ship a full machine out to me on monday.  Frankly buying these from sushi review turned out to be at least a 200 usd mistake due to the dead machine.
   remember I trusted them to deliver good gear and they failed to do that.  they asked for 98 percent deposit on a part that is worth less then 10 percent of the machine.  that is a fail.  If you offer cross ship of the whole machine fine you can ask for 98 percent of the cost. if you offer cross ship for 1 low cost part asking for 98 percent like they did it  was not fine with me.

  I would have not complained but now I will complain. Sad as the gear is nice.  the 3 working machines are nice.
Now I am really nervous since I feel if any of them fail under warranty they will want to screw me to the wall with crazy high cost RMA fees.  

So before you buy from them   consider all the risks:

 and remember what they asked me  to  deposit to send a small part up front. 1.35btc for the controller ethernet card.  

I suspect Bitmain/Sushi has a strict returns policy because they 'trusted' people in the past and got burned... I had no problem getting my btc back the day the card arrived with them.

A refund works for 1.45 btc.   and give me a place to send the dead gear to.

what kills me is this one looked pretty new.

 well it is a brick

 the loose cap  this is a problem





 crooked solder job   but most likely not a problem




 crooked solder job  this looks like it may be a problem

I looked at this one more closely has 2 other issues a cap is not soldered correctly and finally another part is really crooked i.e. compared to a mirror image on the other board.   so basically I now have a dead card a bad cap and a mystery part soldered in very off angle. 

 I need a new machine asap.   It would be nice if sushi just sent it out  and trusted me to send the return. I will post other broken parts.

I've read comments regarding problems with the card a few times. If it's the card shouldn't there be some in inventory somewhere where they could be acquired ASAP? It's three plug in connectors, an antenna base and some plastic standoffs to make the swap. Sending the whole thing in for that seems excessive.

Any fixes for the on-board wifi yet. Mine only connect if I use the wired LAN first then disconnect. The GUI only works through the wired LAN. I think that's where we left off with these discussions AFAIK.

I use 2 RM-750 PSU to power 3 OC Ant's.

Two units full powered, and the third, each board on a PSU using the modular cables.
Works like a charm!

The only thing I can suggest is to reset it to factory.  Just in case the configuration on it has the Ethernet as disabled.  If the physical LAN port is defective, I don't know how you're gonna be able to set it up.  Too bad we just can't buy a controller for the Antminer S1...

Good luck.

I got my third and fourth machines today one ethernet card seems to not light up is there a reset do I have a dead ethernet port? help please.

Its very hard to power two from one power supply, as you'll need a huge unit like this: http://goo.gl/NwGTyW
It actually is more economical, flexible and generally easier to use a PSU per unit. Most people use these, so flexible, strong yet cheap: http://goo.gl/BSzTte

I am new to mining and ordered 2 units and like to know if I could run both units with one power supply, looks like some people here are using it with one PS but no details on how to do it, which PS should I get and would it have enough connectors for 2 units.

Would appreciate any input.

I was also interested in the cgminer screen and been playing around with it.  The way the Antminer S1 is setup is quite complex.  Convoluted is a word that comes to my mind, but it's because it's too difficult for me to follow.  I had to face up to it, as I really wanted to play around with the idea of a multi-pool strategy and luckily for me giletto already started a thread listing the files that were needed to be modified and in doing so, I had to deal with the logic ("flow") of how the Antminer is setup.  I now have a working load-balance strategy with quota on both of my Antminer S1's (all 3 pools are active with percentage settings for each pool).

Back to the post subject.  Antminer S1 "daemonizes" the cgminer application via busybox tool start-stop-daemon during the boot  (/etc/init.d/cgminer) and sending it to the background.  If we modify that, the cgminer may "freeze" the boot up process as it continues to stay in the foreground.  We need screen to replace the start-stop-daemon as we don't wanna have a risk of  bricking the device if it doesn't finish the boot up.

The Antminer S1 has a recovery mechanism by way of a cron job that checks the status of the cgminer and start/stop as necessary and restarts the process, through a series of checks of application logs (shares completed, time stamps, etc.).  Nice.

So, for fun as I wanted to see the cgminer active screen; I ssh'ed to the Ant, killed the crond application and the cgminer application.  Started cgminer with the --bitmain options and 1 pool setting, and the screen shot is below.  My password has a $ in it which messed up the pool login, so my hash didn't show up on the pool.  The miner status Antminer web page showed empty everything, but I can clearly see the miner is very active.  I brought up my Anubis, but it couldn't connect to this particular Antminer, only to the other one.

TL; DR