I still have issues with these and the Raspberry PI.
They are detected, and they mine for a while, on the 7-port d-link usb 2.0 powered hub. Then some of them die. (go ZOMBIE on cgminer) and disappear from lsusb.
Is this a power issue? They work fine on an anker usb 3.0 hub connected to my macbook pro.
I Had two bi*fury connected to the pi, working for some 20 minutes fine... then I connected a third. This happened:
I've seen similar behaviour on my setup (powered USB hub connected to a Notebook) and not limited to the BiFury device (happened with my BlueFury v2 's as well).
IMHO it appears to be some form of small power surge on the hub when inserting an USB miner to a setup of other already running USB miners. Enough to interrupt the normal mining process and send one or more devices into some form of confused AWOL.
Solutions I found :
- remove the failed (Zombie) miner and re-insert
slowly (some USB devices seem to prefer to be inserted rather slow in order to register)
- if that doesn't work (inserting any USB miner causes other running ones to fail), seek another distribution across the USB hub if feasible, sometimes this helps
- if all else fails - with all USB miners installed - unpower the entire hub and turn it back on... normally that gets all USB devices to register normally and start working again
PS.
All that happened to me despite the hub's power supply delivering more than sufficient energy; however despite the
quantity of electrical power being sufficient, I sure don't know about the
quality of the power supplied to the hub. Therefor I'd suspect the power supplies for many hubs may not be top notch in terms of voltage/ampere stability under conditions of sudden load changes.
Happens for me too, but mine dosnt go zombie. The HW errors just keeps going up, from 1% to 100% and then it stops hashing with it.
I just restart the raspberry after that, and it runs fine again for a couple of days.
BTW. i use bfgminer instead of cgminer, but it seems to be the same problem.
Show Posts
