rssLink RSS for all categories
 
icon_red
icon_green
icon_red
icon_red
icon_blue
icon_green
icon_green
icon_red
icon_red
icon_red
icon_orange
icon_green
icon_green
icon_green
icon_green
icon_blue
icon_green
icon_orange
icon_red
icon_green
icon_red
icon_red
icon_green
icon_red
icon_red
icon_red
icon_red
icon_orange
icon_green
 

FS#10913 — FS#14813 — bhs-103-n5

Attached to Project— Network
Incident
Beauharnois
CLOSED
100%
This n5 crashed.
We had to restart it electrically.

Reason: Kernel Panic

Everything is back up.


-
20 Slave impacted by T01A43.
The service is switched to master, no interruption of service.
Date:  Monday, 28 September 2015, 17:35PM
Reason for closing:  Done
Comment by OVH - Monday, 28 September 2015, 17:30PM

This n5 just crashed for the 2nd time.

A warning has been logged! Warning Code = 0x13, Minor Warning Code = 0x0, Data = 0xFF
Socket = 0 Channel = 0 DIMM = 0


A warning has been logged! Warning Code = 0xB, Minor Warning Code = 0x0, Data = 0xFF
Socket = 0 Channel = 0 DIMM = 0

RDIMM population
Command phase 0
Re-center RdDqs
RdDqs re-training with loop count = 4/8
Re-center WrDq
WrDqs re-training with loop count = 3/7
Re-run Rd Vref
Read Vref training with loop count = 10
Round Trip Latency Fix-up
Round trip training with loop count = 7

Checking margins for all ranks with loop count = 10...

RxDqLeft RxDqRight RxVLow RxVCenter RxVHigh TxDqLo TxDqHi
---------------------------------------------------------------------------------------------------------------------
Fatal Error! All channels are disabled!


Comment by OVH - Monday, 28 September 2015, 17:31PM

We are copying the images in order to update.


Comment by OVH - Monday, 28 September 2015, 17:32PM

We have not had time to copy the image where n5 crashed for a 3rd time. We have replaced it with a spare.


Comment by OVH - Monday, 28 September 2015, 17:35PM

The new n5 is updated. We are copying the backup configuration.

We will go over it with robots to check that the configuration compiles on storage ports.


Comment by OVH - Monday, 28 September 2015, 17:35PM

Everything is back to normal.