Determining Volume Group Types in AIX
Editor’s note: This is the second article in a two-part series on addressing Logical Volume Manager (LVM) issues. Part one focused on “Dealing With Duplicate PVIDs for rootvg.”
This LVM piece was inspired by a colleague’s query about identifying different volume group (VG) types.
“Hi Chris,
I wonder if you could help. I've got a normal VG (I thought) with 7 [physical volumes] PVs, so I tried to add another two. It only let me add one, but when I did an extendvg for the second PV, I got told to convert it to a big vg.
Haven't used big vg before, but I can't get downtime to convert it to scalable, so... did the migratelp to allow 1 free [physical partition] PP on each PV, then ran:
chvg -B datavg
0516-1213 chvg: Cannot get the current volume
group descriptor area size or volume group is already in
big volume group format.
0516-732 chvg: Unable to change volume group datavg.
How do I see whether it's a big VG? Isn't it the MAX PVs value? The MAX PVs for my volume group is 8, instead of 128 for a big VG! What’s going on?”
So, how do we check whether a volume group is a “big VG?” First, let’s take a quick look at the available AIX VG types and configuration limits:
Table 1
| VG type |
Maximum PVs |
Maximum LVs |
Maximum PPs per VG |
Maximum PP size |
| Normal VG |
32 |
256 |
32,512 (1016 * 32) |
1 GB |
| Big VG |
128 |
512 |
130,048 (1016 * 128) |
1 GB |
| Scalable VG |
1024 |
4096 |
2,097,152 |
128 GB |
As you can see from Table 1, the value for maximum PVs can indicate a volume group’s type. For example, a big VG has a default of 128 for maximum PVs (instead of eight in my friend’s case). You can also use the readvgda command to check a VG’s type:
# readvgda hdisk5 | grep type
..... readvgda_type: bigvg
Note, however, that the output shown from the readvgda command may differ depending on the version of AIX and the technology level (TL) and service package (SP) installed on the system.
A VGs factor value can also impact the number for MAX PVs. Changing a VG’s factor value allows administrators to manipulate the limit on the number of PPs per PV in a given VG. Administrators would typically need to do this if they’re adding disks of a larger PP size to an existing VG that contained disks of much smaller PP sizes.
The trade off is that by changing a volume groups factor, and thus allowing a greater number of PPs per disk, you reduce the maximum number of disks allowed in the volume group. For example if you were to run the command chvg -t 4 datavg (changing the VG factor to 4), this would increase the number of PPs per disk and reduce the number of disks allowed in datavg.
The other “give away” to a group’s type is the maximum logical volume’s (LV’s) value. If it’s 512, this could indicate a big VG (as shown in Table 1). In my colleague’s case, the VG factor had been changed at some point, which changed the expected value for MAX PVs for his big VG.
Chris Gibson is an AIX and PowerVM specialist located in Melbourne, Australia. He is an IBM Champion for Power Systems, IBM CATE (Power Systems and AIX), and a co-author of several IBM Redbooks publications.
More Articles From Chris Gibson
Search our new 2013 Buyer's Guide.
Maximize your IT investment with monthly information from THE source...IBM Systems Magazine EXTRA eNewsletter. SUBSCRIBE NOW.
View past AIX EXTRAs here
Related Articles
Web Exclusive | Understand your options for 12X PCIe I/O drawers
E-Newsletter | PowerHA SystemMirror 7.1 introduces a robust CLI utility
Web Exclusive | AEM controls power flow and cooling distribution