Agent faulty reported dead, minerfarm restart on rigg restore connection.

[agkbill]

Hi,

have for some time experienced problem with that agents are reported dead.

In picture below, agen "RX-480 Clare P2.5" is reported dead, but if I SSH to the rigg and look in minerfarm all is OK. This is confirmed if I look at the mining pool that Clare is connected to.

Restart of minerfarm (first press 'x' and then 'o') agent start to report OK again.

Any ideas how to troubleshoot or solve this would be fantastic.

All the best you you all!
Christer

[lily]

Please ensure the time on the rig is current. Offset by more than a minute will cause agent reporting issues.

[agkbill]

On one of my riggs I have.

[/root]:# timedatectl status
Local time: Mon 2018-03-12 16:47:14 EDT
Universal time: Mon 2018-03-12 20:47:14 UTC
RTC time: Mon 2018-03-12 20:47:14
Time zone: America/New_York (EDT, -0400)
Network time on: yes
NTP synchronized: yes
RTC in local TZ: no
[Agent:

Looks quite OK, but timezone is all off, I'm in Stockholm/Europe Zone. will change that and se what happen.

[lily]

In general, you should not need to change any of the time settings on your pimp rig. However, sometimes the time drifts on a rig. ntpd is a time daemon that should keep it in sync but if the gap is too large it will sometimes refuse to update things. In the event that your rig is at the wrong time, you can do:

'sytemctl stop ntp; ntpdate time.nist.gov; systemctl start ntp'

This series of three commands stops the ntp service, updates your clock to time.nist.gov then restarts the ntp service.