Intermittent diskgroup failures on upgrading from 5.0.x to either 5.1.x or 6.0.x due to mix of standard and cloned devices in diskgroup

Problem

Diskgroups  will experience intermittent import failures if they contain a mix of standard and cloned devices in the (same) diskgroup

Error Message

VxVM vxdg ERROR V-5-1-10978 Disk group <diskgroupname>: import failed:
Disk group has no valid configuration copies

Sep  9 08:40:31 hostname vxvm:vxconfigd: [ID 702911 daemon.warning] V-5-1-0 Disk Group <diskgroupname> has a mix of standard and cloned disks: Trying to import the disk group from the standard disks: c3t5006048C52A83528d86s2 c3t5006048C52A83528d87s2

Cause

Volume Manager can cause intermittent diskgroup import failures for a diskgroup that has a mix of standard and cloned devices .

- Volume Manager implemented a change in version 5.1  that caused the Lun Serial Number (LSN)  for EMC arrays to be shorter than in previous versions.

-The UDID is a tuple( 4 field) consisting of : VendorId , Product ID, Cabinet Serial number , LSN.

- LSN format change from version 5.0.x to 5.1 leads to UDID mismatch upon upgrade for disks that were initialized in 5.0.x

-UDID mismatch triggers  the clone flag to be set internally and the diskgroup will be flagged/imported as “cloned”.

-If a disk is added to such a diskgroup it will be added as a STANDARD device causing the diskgroup to contain a mix of clone and standard devices.

Solution

Note: The clone/udid related article with a command to determine disks with differing UDIDs  is: https://www.veritas.com/support/en_US/article.100001562

For all the disks that have two different udid’s

Applies To

Storage Foundation Upgrade from : 5.0.x to 5.1.x and 6.0.x

PLATFORM(S):  SOLARIS, HPUX, AIX & LINUX