Go to previous page / next page of Tier3 site log

28. 12. 2012 t3fs07,t3fs08 went down

A bit of theory about ACPI states; from that I understand that S5/G2: soft-off => we still have power, for instance to use ILOM like indeed I'm doing.

Today these 2 servers went down:

t3fs08
ID =   f8 : 12/28/2012 : 09:04:43 : System ACPI Power State : ACPI : S5/G2: soft-off
ID =   f9 : 12/28/2012 : 09:04:45 : Power Supply : PS0/PWROK : State Deasserted
ID =   fa : 12/28/2012 : 09:04:47 : Power Supply : PS1/PWROK : State Deasserted
t3fs07
ID =  308 : 12/28/2012 : 10:04:29 : System ACPI Power State : ACPI : S5/G2: soft-off  # 10:04:29 => 09:04:29
ID =  309 : 12/28/2012 : 10:04:31 : Power Supply : PS0/PWROK : State Deasserted
ID =  30a : 12/28/2012 : 10:04:33 : Power Supply : PS1/PWROK : State Deasserted

Notification Type: PROBLEM
Service: Temperatures Celsius t3fs07 [t3fs08]
Host: t3admin01
Address: 192.33.123.21
State: CRITICAL
Date/Time: 12-28-2012 09:25:11
IPMI Status: Critical [P0/T_CORE = N/A, P1/T_CORE = N/A] 

t3fs07 : ipmitool -I lanplus -H rmfs07 -U root -f /root/private/ipmi-pw sel list
 307 | 12/28/2012 | 10:04:29 | System ACPI Power State #0xea | S0/G0: working | Deasserted
 308 | 12/28/2012 | 10:04:29 | System ACPI Power State #0xea | S5/G2: soft-off | Asserted
 309 | 12/28/2012 | 10:04:31 | Power Supply #0xbd | State Deasserted
 30a | 12/28/2012 | 10:04:33 | Power Supply #0xcc | State Deasserted
t3fs08 : ipmitool -I lanplus -H rmfs08 -U root -f /root/private/ipmi-pw sel list
  f7 | 12/28/2012 | 09:04:43 | System ACPI Power State #0xea | S0/G0: working | Deasserted
  f8 | 12/28/2012 | 09:04:43 | System ACPI Power State #0xea | S5/G2: soft-off | Asserted
  f9 | 12/28/2012 | 09:04:45 | Power Supply #0xbd | State Deasserted
  fa | 12/28/2012 | 09:04:47 | Power Supply #0xcc | State Deasserted

[root@t3admin01 ~]# ssh rmfs08 
Sun(TM) Integrated Lights Out Manager
Version 3.0.3.36
Copyright 2009 Sun Microsystems, Inc. All rights reserved.

-> show /SYS/PS0/PWROK
 /SYS/PS0/PWROK
    Targets:

    Properties:
	type = Power Supply
	ipmi_name = PS0/PWROK   <---
	class = Discrete Sensor
	value = State Deasserted   <---
	alarm_status = major   <--- 

Possible causes

-- FabioMartinelli - 2012-12-28

Go to previous page / next page of Tier3 site log