Solaris Volume Manager (SVM) the 'metadevadm -u' Command Shows "New device reloc information" or "Invalid device relocation information detected" But DevID Is Not Updated (Doc ID 2090016.1)

Last updated on SEPTEMBER 30, 2022

Applies to:

Symptoms

The problem described in this article can have multiple symptoms and errors. One of these issues described in this article can occur right after disk replacement, or when disk was not correctly replaced in the past. When this problem happens there will be a mismatch of Solaris Volume Manager (SVM) disk Device ID (DevID) is in the metadb that causes various errors and meta command failures.

In below example, the DevID in metastat output for disk c1t1d0 does not match the DevID in /etc/lvm/mddb.cf and /kernel/drv/md.conf despite metadevadm -u command was done. Beware that depending on your server the name of the disk can also be a WWN disk name like "c0t5000xxxxxxxxxxxxx0s0".

Example:

 You will need to enter maintenance mode and delete the metadbs on the disk that got the metadevadm error.
The following document has the metadb commands to run.
<Document 1010270.1> Solaris Volume Manager (SVM) Recovering From Insufficient Metadevice State Database Replicas

Changes

The following can lead to this issue:

1) patching the system with recommended patch cluster so it is now at a patch level that is now vulnerable to the bug described in this article.

2)  incomplete or incorrect disk replacement

Cause

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