Topic: [Setup & Troubleshoot] Bitmain AntMiner S1 180GH/S miner - page 106. (Read 452359 times)

Is it usually very difficult to get a reply from sushi in regards to a faulty board?
I bought from Bitmain directly and he told me to contact Sushi during the CNY.
Even sushi himself send me an email:

Solo mining will take you ages to find a block. Even at 180GH won't it take maybe a year or more to find a block on average?

If you don't want to use third party pools, your best best is to run a local copy of p2pool and mine off of that. Then you will reduce your variance since you are sharing work with all of the other p2pool miners.

is there anyone that solo mines using the antminer s1??? if so can you share how?

I've found a sweet-spot on the damaged miner at 387.5 and it keeps the errors under 1%... can't go any higher without it it jumping over 6%. That one shows no X's and has been running at that speed for days now.
The newer miner is running at 400Mhz and is very stable... the X's come and go and there doesn't seem to be a pattern, sometimes they go away without turning the miner off, so I think it's a software problem.

You have to setup an additional port-forwarding for the other Ant, say 5588 pointed to the other Ant's local LAN IP address  (ie 192.168.1.112) on port 80.

For ssh, each Ant will need a port similar to above but pointed to 22 on the local Ants' ip address.  But that also involves your putty/ssh to be configured to connect on a different port, etc. 

Thanks that works great. How do I change the default external port of 3388 to say 3389 ? So I could remote access my second ant as well.

ssh is also open on my Ants on a different port-forward port on the router for each Ant. 

I am thinking there is a way to check the status with ssh...I remember looking at this for my GPU miner setup.

Yes i did never go higher then 387.5 Mhz because i think its not really worth going higher
The one constant showing the xxxx on all cores does seem to make 180 Gh as written by the makers, and so far is not going much faster if i clock it at 387.5 however it seems to be a little faster on 375 Mhz which is in my view still much below the actual speed these antminers can do.
So for some it might be worth the risk goin beyond the 400 Mhz i think the difference will not be worth the small increase in speed gained.
I rather stay well withing normal operations below the 400 Mhz so the miners stay alive for a longer period.
Its just weird that the x's appeared since the box has been placed, but again it does keep running steady around 180 Gh.
Am just wondering if it might be something with the cooling or something, if i let the fan run at full speed the noise is enormous
And even then after some hours mining the x's appear and stay. So i am not sure if it is nothing to worry about or that there is something wrong with this one for the long run.
 

I'm addicted to watching my 2 Ants from anywhere...

• 1.  Port forwarding on your router (facing the internet (ie Port 3388) forwarded to your Ant LAN address 192.168.1.111  port 80
• 2.  Access your WAN router address  http://WANrouter-address:3388
• 3.  Profit

It's not as simple as that, though.  One thing to consider is security.  I have my router configured to allow a single address from the outside (internet) to access.  I set this to my openVPN Amazon EC2 server.   Maybe configure your work IP address on this.  wimi.com can tell you your address.  Also, to make it easy, you may want to use  dyn.com  (if your WAN router supports it).

Does your home/WAN router support port-forwarding?

Good luck 

Hey has anyone figured out how to check the status of these things while away from home?

Haha, nice one... didn't skip a hash... I wonder if they're more stable on the way down due to the extra cooling

I'd expect some sort of replacement as there's not really much you can do wrong by ssh'ing into the miner and un-commenting some (already supplied) settings and rebooting.

EDIT: P.S. I did a hard reset and it didn't help... thanks anyway.

If you hoesntly have a board with half dead memory... you will get a slow clap from me for breaking these more than I have so far (I have a habit of "breaking" new miners in order to better understand how they work), and that includes the 2 I had fall off my rack and were still hashing after they hit the floor.

Thanks for the tip, I'll give it a try, I did need to do a hard reset during the initial setup (obviously noob) but I'll do it again... the kernel log is reporting half the memory, so I thought it would be physical. To be fair, it does say "may void warranty", so I think they will be understanding depending on what exactly the damage to the unit is... it's not exactly melted

Haha good. Many of my customers have just overclocked them, them come crying to me when something breaks.

For your control board- have you tried resetting the miner way back to factory, then OC again? Some of the ants I have gotten have been finiky, and just needed some hard resets to get back working when I thought I had a bad controller.

Yep, I did read the big fat red letters on the top of the page before I bought it 

You do know by overclocking they will void your warranty right?

Hey bronan, I have the same problem with one of my antminers, I think the control board is faulty. I have 28848k/32768k on the broken one and 61336k/65536k on the newer one that runs at 200GH/s with no problems, except the XXX issue which I think is nothing to worry about if your hashing speed is reported correctly and your HW errors are 1% or below. I've just noticed the X's go away when logging off the GUI and logging on again, without turning the miner off and my pool stats also show there is nothing wrong... which is the best 'independent' way to check if it's OK.

I'm not able to overclock the faulty miner either or it produces too many hardware errors and I'm trying to get a PCB replacement in the meantime.

Why are there no new firmwares released with updated cgminer and such ?
Why can't i see in the miner which firmware is installed and which pcb this thing is or do i really have to go the machines to check ??

I assumed if you bought 2 at the same time that i did got 2 exact same machines but i found out both are different
One is a bit cooler then the other and i see different memory numbers
Box 1 shows  15720/29080 i asume this one is 32 Mb mem
Box 2 shows  48336/61586 i asume this one is 64 Mb mem
and is kinda confirmed on the kernel log
The one with the larger memory is constant turning all OOOO into XXXX a some hours mining.
Not sure what to think of this antminer, It does not matter if it run at 350 or 375 it constant ends up showing all X's
I have not tried to set higher speeds because it is already giving enough hw error as it is at 350 or 375
But thats sadly another mess on the antminer there is no speedy weay to see how bad its doing.
I can not use screen and the status page lacks a simple % value

Another nasty habit is that i can't keep the status page open, the freaking overview page does have refresh but the one needed the STATUS page is not able to auto refresh ..... SIGH

I also did not find a solution to keep an eye on the miner by other means

Anyone have a solution for me to see those or know why one runs fine and the other seems to be wacky.
Or maybe i worry to much about the stupid XXXX but then again i paid enough for it, so i want to know if it is having some kinda hidden defect

Has anyone RMA an AntMiner S1?  How long does the process take?  I sent in a RMA per BitMain's PM direction on 2014-01-26 to an address in the US (the address turns out to be an UPS store).  There is no RMA # given or any kind of tracking number on BitMain's website to track the status.