IBM Support

Partition keeps booting to SMS

Troubleshooting


Problem

Partition keeps booting to SMS

Symptom

A partition is rebooted and instead of booting to the operating system, AIX or PowerVM Virtual I/O Server (VIOS), it stops in SMS and requires manual intervention to complete the boot process.

Resolving The Problem

If the partition boot process stops at SMS, ensure you do not get any LED codes that indicate that no boot device was found. If the bootlist is not properly set or no boot device is found, the partition stops at SMS.

If you are able to select X to exit SMS and the boot process continues normally without having to select the boot device, there are some more things you can check.

1. Check that the partition's Boot Mode is set to Normal.

On the HMC Classic GUI (V7 and V8)
> Select the partition and go to Properties
> In the settings tab, check that Boot Mode is not set to SMS. If it is set to SMS, you need to bring down the partition and reactivate it with a profile that has "Boot mode" set to Normal.



2. Check that the Multiboot Option is set to OFF

If the partition is already booted to AIX, run the following as root.  For VIOS, first run 'oem_setup_env' to drop to the AIX root shell:

# /usr/lib/boot/bin/dmpdt_chrp -i  | sed -n '/auto-boot/{;N;p;}'
If auto-boot is enabled, the output shows:
  auto-boot?
        74727565 00                         [true............]
Alternatively, open a vterm console to the partition and activate it to SMS.
  1. In the SMS Main Menu, select #5 "Select Boot Options"
  2. Check #3 "Multiboot Startup". It should be set to OFF



3. Check that keylock position is not set to Manual

Select the managed system that has the partition you are working on and select Properties.
In the Power-On Parameters tab, make sure the Current Keylock position is not Manual. If so, set the Next Keylock position to Normal. Unfortunately, this requires the entire managed system to be power cycled for the change to take effect.



4. For p5 systems, check the parent adapter of the boot disk and its adapter microcode

The following Ultra320 SCSI Adapter microcode levels have selection time-out problems w/ HUS10K DASD. This might cause a boot drive not to be seen during the boot process which would cause the system to boot to SMS.
 
CCIN/Feature Code Microcode version Adapter description
5702/1974 05080064 Dual Channel Ultra320 SCSI Adapter
5703/1975 05090066 Dual Channel Ultra320 SCSI RAID Adptr
5709/1976 050D0067 SCSI RAID Enablement Card
2780 050E0064 Quad Channel Ultra320 SCSI RAID Adptr
573D/1908 05110048 SCSI RAID Enablement Card
570A/570B 050A0064 Dual Channel Ultra320 SCSI Integrated
573E 05100047 Dual Channel SCSI Integrated Cntrllr
NOTE:CCIN 573E includes the following part numbers which were built with this problem SCSI microcode:
SYSTEM FRUs
9111-285 39J4245
9131-52A 39J4240
9115-505 10N6679, 10N6688, 10N6696

If you have the Ultra320 SCSI Adapter at the microcode level listed above, you will need to update to the latest available microcode level for your given adapter or integrated controller. This will prevent occurrence of the issue.


http://techsupport.services.ibm.com/server/mdownload/download.html

Since the adapters contain the boot disks, a Diagnostic CD and an ISO image CD containing the microcode image can be used to download microcode to an adapter in a system that will not boot to AIX/Linux. Instructions are contained in the various microcode readme files.

[{"Type":"MASTER","Line of Business":{"code":"LOB08","label":"Cognitive Systems"},"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Product":{"code":"SWG10","label":"AIX"},"ARM Category":[{"code":"a8m50000000L0FmAAK","label":"Boot"}],"ARM Case Number":"","Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"All Versions"}]

Document Information

Modified date:
29 September 2022

UID

isg3T1011504