Use prtconf to find the reason of unexpected reboots
In some cases, “prtconf -vp” can help determine the cause of an unexpected crash/reboot just by knowing what the “reset-reason:” is.
# /usr/sbin/prtconf -pv
System Configuration: Sun Microsystems sun4u
Memory size: 4352 Megabyte
System Peripherals (PROM Nodes):
Node 0xf0029c1c
clock-frequency: 05f5e10
previous-reset-reason: ‘POR’
banner-name: ‘8-slot Sun Enterprise E4500/E5500’
idprom: 01800800.20a0f6d3.00000000.a0f6d3a9.00000000.00000000.00000000.00000000
reset-reason: ‘S-POR’
fatal-reset-info: 00006000
breakpoint-trap: 0000007f
#size-cells: 00000002
name: ‘SUNW,Ultra-Enterprise’
We see that the “reset-reason” is a ‘S-POR’. Use the following UltraSPARC(R) II chart to see what might have caused this reset.
reset-reason
Description
Could be caused by
POR
Power On Reset
HW Fatal Reset
S-POR
Soft Power On Reset
Panic, Regular reboot
B-POR
Button Power On Reset
Power-cycle, Power loss
S-XIR
Soft eXternaly Initiated Reset
Watchdog Reset
B-XIR
Button eXternaly Initiated Reset
XIR Button pressed
For the UltraSPARC(R) III systems only the V880 and V480 with OBP 4.10.7+ records the reset-reason. These are the possible codes:
POR Non-specific Trap Type 1
HPOR Sys Power On Normal Power On Reset
APOR Alarm Power On Alarm Power On Reset
(03) Undefined Undefined, should not happen
BPOR Button Reset Hardware “Button” reset
RPOR R/SC Reset RSC/RMC/SC/etc. reset
JPOR JTAG Reset JTAG/Scan-induced reset
WPOR Sys Watchdog System-wide external Watchdog
WDOG CPU Watchdog CPU-specific watchdog reset
XIR External/NMI System/CPU XIR
SIR CPU Instr CPU-specific SIR
ED CPU RED-State CPU-specific RED-State Expection
FATAL Sys Hardware System-wide hardware error/reset
ROMT ROM-Trap ROM Trap — should not happen
SPOR Software/User General command reset

Note: The system should not have been power cycled or rebooted manually otherwise the “reset-reason” won’t correspond to the reason of the unexpected reboot.


4 Comments

Erin · 08/06/2018 at 17:59

It works very well for me

Hassan · 11/06/2018 at 16:01

I enjoy the report

Colette · 16/08/2018 at 12:35

Thanks for the wonderful post

Lauren · 25/08/2018 at 19:53

I enjoy the report

Leave a Reply

Your email address will not be published. Required fields are marked *

five × 3 =

Related Posts

Oracle

Solaris Snoop for packet sniffing

The snoop command can come very handy to monitor the network traffic to troubleshoot any network related issues like packet drops, high network latency etc. Snoop command can be run to see the real time Read more…

Oracle

Verifying HBA card connectivity in Solaris

1. Use the “luxadm -e port” command to verify HBA has established communication with a node. # luxadm -e port /devices/[email protected]/[email protected]/[email protected]/[email protected]/SUNW,[email protected]/[email protected],0:devctl CONNECTED /devices/[email protected]/[email protected]/[email protected]/[email protected]/SUNW,[email protected],1/[email protected],0:devctl CONNECTED CONNECTED : HBA has established connection with either target or initiator. Read more…

Oracle

Solaris File System Structure

Hi Guys, here’s a little explanation about the structure of the file system on solaris; 1. / – Root Every single file and directory starts from the root directory. Only root user has write privilege under Read more…