My Oracle Support Banner

Oracle VM Server Node Evictions during IBM 2145 SVC Firmware Controller Upgrades (Doc ID 2409589.1)

Last updated on OCTOBER 10, 2022

Applies to:

Oracle Cloud Infrastructure - Version N/A and later
Oracle VM - Version 3.3.4 and later
Linux x86-64

Symptoms



ACTUAL BEHAVIOR
---------------

While upgrading the firmware on the storage first controller, paths to the LUN's fail and are not restored. Subsequently when performing the firmware upgrade on the second controller, all paths are faulty and node evictions occur.

Paths are restored after nodes have rebooted.

example multipath output 

36005076801810xxxxx00000000001ce3 dm-146 IBM,2145
size=12G features='1 queue_if_no_path' hwhandler='0' wp=rw
|-+- policy='round-robin 0' prio=0 status=enabled
| |- 0:0:7:250 sdfr 130:208 failed faulty running
| '- 1:0:6:250 sdlr 68:400 failed faulty running
'-+- policy='round-robin 0' prio=0 status=enabled
|- #:#:#:# - #:# failed faulty running
'- 1:0:7:250 sdmj 69:432 failed faulty running

 
EXPECTED BEHAVIOR
-----------------------
OVS servers should have the paths restored after the first controller is upgraded successfully.

STEPS
-----------------------
The issue can be reproduced at will with the following steps:
1. Perform firmware upgrade on primary controller (SVC-2145)
2. While performing upgrade on the second controller, Oracle VM servers lose access to the pool filesystem and are rebooted.


Cause

To view full details, sign in with your My Oracle Support account.

Don't have a My Oracle Support account? Click to get started!

In this Document
Symptoms
Cause
Solution
References


My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.