Do your S1's stop hashing? Reboot it Automatically.

[rotediablo]

Every once in awhile my S1's will stop hashing on a random basis.
I have to power cycle to reboot or login through the web interface and reboot.

This becomes a big hassle. I have to babysit my miners.
Sometimes I don't catch it right away (I have to sleep sometime) and the miners might go for hours without hashing.

Does this happen to anyone else?

I have solved this problem by making the S1 reboot every 30 minutes.

1.) Login to the web interface.
2.) Go to the system tab.
3.) Go to the Scheduled Tasks tab
4.) You should see the following line of code in the text box --> */3 *  *   *   *     /usr/bin/cgminer-monitor
5.) Go to the next line and paste the following code --> */30 *  *   *   *     reboot
6.) Click the green Submit button in the bottom right hand corner.


Now your miner will reboot automatically every 30 minutes.
You can change the 30 to any other number of minutes you desire.
60 would cause the reboot to happen every hour.


No more babysitting for me.

Hopefully this helps some of you too.

I have never been able to find out why the miners stop hashing.
In the meantime this solves the problem.

[WinterParker]

Thank you for posting this.  I have the same problem.  I was starting to hook all mine on a variable power timer to reset every hour!  This is much more efficient, Thanks again!

[MistuhSoftee]

extremely helpful, Thanks!

[rotediablo]

Glad it is helping! 

[nuff]

This helps, until someone can figure out how to automatically reboot when it stops hashing, this is the only workaround I think. Would rebooting affects anything as oppose to keeping it constantly hashing for extended periods of time? Like if I mine with a pool that disconnection might reset my share count or anything like that?

[bbxx]

for me ant stop hashing when error -71 appear in logs (lack usb response from device)
easiest way to check if ant is hashing is to measure cpu load
if it is below 1 it is time to reboot

maybe lets try simple script

cat /proc/loadavg | grep . | cut -c 1

gives 0 when dead, 2 when hashing

lets assume it is x

if x<1 reboot

any tips how to finish this script ?

[rotediablo]

I wrote a small script to use the above info to see if the miner has stopped hashing and reboot automatically when it does.

1. SSH into your miner. I use putty (http://www.chiark.greenend.org.uk/~sgtatham/putty/download.html)
2. Login to the miner
3. type cd /bin
4. type vi hashcheck
5. paste the following code (in bold) into the vi editor

     #!/bin/ash
      var=$(cat /proc/loadavg | grep . | cut -c 1)
      if [ $var -lt 1 ]
      then
      /sbin/reboot
      else
      fi

6. Press the Esc key
7. Type :x and then press enter
8. Close putty
9. Now access your miner via the web interface
10. Click the system tab
11. Click Scheduled Tasks
12. Under */3 *  *   *   *     /usr/bin/cgminer-monitor add the following on the next line */1 *  *   *   *    /usr/bin/hashcheck
13. Click Submit in the bottom right hand corner

That's it. Now your miner will run this script every minute. If the miner has stopped hashing the script will return a value of 0 causing the miner to reboot automatically.
As long as the miner is hashing the script will return the number 2 and nothing will happen.

[tzortz]

Nice! Have you tried it ?

[Bogart]

I wonder why your S1s stop hashing.  I have 8 S1s myself and I never have that problem with them.  They're currently showing 44 days uptime.

I have mine overclocked to 400MHz, running the 20131226 firmware, and receiving their IP address assignments using DHCP.

If you only have one pool configured, I would suggest adding a second pool in case of downtime on the primary.

[aurel57]

I wonder why your S1s stop hashing.  I have 8 S1s myself and I never have that problem with them.  They're currently showing 44 days uptime.

I have mine overclocked to 400MHz, running the 20131226 firmware, and receiving their IP address assignments using DHCP.

If you only have one pool configured, I would suggest adding a second pool in case of downtime on the primary.

Same here with my 6 S1's.No down time and  I have them mining on GHash pool and use their UK address and then BTC guild for back up.

[Soros Shorts]

I wonder why your S1s stop hashing.  I have 8 S1s myself and I never have that problem with them.  They're currently showing 44 days uptime.

I have mine overclocked to 400MHz, running the 20131226 firmware, and receiving their IP address assignments using DHCP.

If you only have one pool configured, I would suggest adding a second pool in case of downtime on the primary.

Same here with my 6 S1's.No down time and  I have them mining on GHash pool and use their UK address and then BTC guild for back up.

I have one out of a dozen that does this every couple of weeks. When the pool stats reports it dead, I have an external monitoring program that power cycles the S1 through an iBoot switch. This is probably overkill but I had already developed the system for my Batch 1 Avalons and had a bunch of iBoots lying around.