Volume groups in inactive state after AIX 6.1 Migration
meyawi 270002D6NU Visits (6435)
Even if support ended for AIX 5.3 since April 2012, some companies still have old AIX releases like 5.2 or even older. I had some migrations to do for one of my customers on standalone P5 and blades, everything went well but I saw filesystems mount error messages at boot. Once logged in, I saw all my volume groups but few of them were inactive, and a varyonvg gave error messages.
root@meyawi:/> lspv hdisk0 000954960b2bd9b0 rootvg active hdisk1 00095496a7e39adc rootvg active hdisk4 0057408a369d6fbd vg_stg_tsm hdisk9 0009549602bff66e vg_stg_archive active root@meyawi:/> varyonvg vg_stg_tsm 0516-1273 varyonvg: LTG size must be less than or equal to the max_transfer size of all disks in the volume group.
Let’s find hdisk’s max_transfer value (hexadecimal, bytes) and the volume group’s LTG size (kilobytes)
root@meyawi:/> lsattr -El hdisk4 -a max_transfer max_transfer 0x40000 Maximum TRANSFER Size root@meyawi:/> lqueryvg -Atp hdisk4 | grep LTG LTG size: 1024
To get max_transfer, you could also use : lquerypv –M hdisk4
Available values for max_transfer are found with this command : lsattr -Rl hdisk4 -a max_transfer
which give us :
All we have to do is to set up a proper max_transfer value, that is 0x100000 :
root@meyawi:/> chdev -l hdisk4 -a max_
Now filesystems are correctly mounted.
When we compare LTG parameters of volume group vg_stg_tsm, it seems that LTG determination is not dynamic even if we didn’t force LTG size. It is because this volume group was created prior to AIX 5.3.
root@meyawi:/> lsvg vg_stg_tsm |grep LTG LTG size: 1024 kilobyte(s) AUTO SYNC: no root@meyawi:/> lsvg vg_stg1_archive |grep LTG LTG size (Dynamic): 256 kilobyte(s) AUTO SYNC: no
Here is an excellent post that explains this behavior and from where I quoted the above part :