best I can get is about 7-8 days now with 2.0.2 before it starts reboot looping any time it sees 1 get failure.
Just caught it after running fine for 8 days. Unfortunately for all day today since last night, its been rebooting every 5 minutes.
Hopefully there is a software update SOON that fixes this issue
Long time no see, Eyedol.
It's hard for us to catch a "1 Get Failure" in our lab. We've tested for weeks but no luck to catch it.
To exclude the possibility of IP conflict, could you please reset the Pi then move the miner to another place to see if problem is fixed.
Till now we only received your feedback about this issue like this. Really sorry about that. Could you please try?
I understand you've been unable to duplicate this issue and I know you haven't been very active here so I have outlined everything for you below including posts from Mjgraham experiencing and reporting the same issue, in the same timeframe.
I have yet to find someone that can produce a screen capture illustrating B3 A4 running 2.0.2 that can show any number of Get Failures other than 0 because if 1 Get Failure is observed, the entire Raspberry Pi reboots itself. I challenge you to find me someone that can show you a screen capture that shows more than 1 Get Failure on a B3 A4 running 2.0.2. So far I can't find any evidence of that but I can show you screen captures below of my A2's running for more than 89 Days without a reboot and have multiple Get Failures. Get Failures are not always network issues on the client side and the entire system shouldn't reboot itself every time one happens. This doesn't happen in the A2 but it does happen in the A4 on multiple networks.
Mjgraham has reported the same issue
1st report here - January 18th, 2017 >
https://bitcointalksearch.org/topic/m.17545969"I did verify that it runs /sbin/shutdown when there is a stratum connection problem, renamed the bin file to test but it shuts down the board and just quits working just tons of comm errors so no way to stop it."
3rd report here with details of what he's done to address this issue - February 9th, 2017 >
https://bitcointalksearch.org/topic/m.17796776"Well batch 3 is better however I have made some modifications, I replaced /sbin/shutdown with another a file that does a hardware MCU reset , had to wire that in along with a sudo pkill cgminer to kill cgminer off, all of the stuff in ran in a do loop do it instantly restarts, well nearly for some reason the cgminer binary takes 10 seconds to start now for some reason."
4th report directly above this post
Another thing is the reboot on slightest problem, I changed the /sbin/shutdown binary to just do a sudo pkill cgminer, the only thing is you have to run cgminer command line and in a while loop but it restarts, but normally when it dies like this you have one more more hash boards that are going to act up ... so the next thing was to solder a wire to the MCU reset and tie them to a pin on the rPi that hardware resets them as well, seems to help some but still testing..
My reports of this issue
January 18th, 2017 - 1st report -
https://bitcointalksearch.org/topic/m.17544909January 19th, 2017 - 2nd report with details -
https://bitcointalksearch.org/topic/m.17553436January 29th, 2017 - 3rd report -
https://bitcointalksearch.org/topic/m.17660448And a few e-mails exchanged with you all directly.
To date here is what I have done:
- I have replaced the Pi
- I have replaced the SD Card with a new one containing a fresh 2.0.2 image
- I have had this same thing occur on a totally different network/location with a 3rd A4 I purchased.
What I have not done:
- I have not yet made a video and uploaded it to youtube that demonstrates this issue in real time while simultaneously logged into the Pi to see the reboot as it occurs -- This video is my next thing to do after I get done with some international business travel coming up. I will be able to provide this video in a few weeks when I have time to make it. Hopefully this issue will be identified and fixed by the time I get back. I've been dealing with it for well over a month now
Here is an example of an A4 with 1 Get Failure on Prohash - January 19th, 2017 -- Immediately after the Pi Rebooted itself. This was with my original Pi and original SD Card running 2.0.2.
Here is an example of an A4 with 1 Get Failure on LitecoinPool - February 19th, 2017 -- Immediately after the Pi Rebooted itself. This is with my new Pi and new SD Card running 2.0.2 on a different network.
Here is an example of an A4 with 1 Get Failure on Nicehash - February 19th, 2017 -- Immediately after the Pi Rebooted itself. This is with my new Pi and new SD Card running 2.0.2 on a different network.
Here is one of my A2's running for 89 days on Prohash on a different network,
60 Get Failures but not a single reboot
Here is one of my A2's running for 109 days on Prohash on a different network,
82 Get Failures but not a single reboot