Here are my notes of the Shared Storage Pools commands that you need to remember. 

SSP is large on concepts and system admin time-saving but simple to operate.

Updated Oct 2017

Creating the pool (once only) by example:

• Assumption: 1 GB Repository disk = hdisk15 and Pool data disks = hdisk7 to hdisk14, cluster and pool both called "alpha"
• Assumption: IBM V7000 A has LUNs: hdisk7 hdisk8 hisk9 hdisk10
• Assumption: IBM V7000 B has LUNs: hdisk11 hdisk12 hisk13 hdisk14

Examples

cluster -create -clustername alpha  -spname alpha  -repopvs hdisk15   -sppvs hdisk7 hdisk8   -hostname orangevios1.domain.com
cluster -addnode -clustername alpha  -hostname silvervios1.domain.com
failgrp -modify -fg Default -attr FGName=V7000A
failgrp -create -fg V7000B: hdisk11 hdisk12
pv -add -fg V7000A: hdisk9 hdisk10 V7000B: hdisk13 hdisk14

Shared Storage Pools - Daily work

Create a LU (Logical Unit - virtual disk in the pool)

 lu -remove | -map | -unmap | -list  [-lu name]

 lu -create -lu name -size nnG -vadapter vhostXX [-thick]

Examples

lu -create -lu lu42 -size 32G -vadapter vhost66 -thick

lu -map    -lu lu42           -vadapter vhost22

lu -list

lu -list -verbose

lu -remove -lu lu42

Snapshots - stop the VM for a safe consistent disk image but you could (if confident) take a live snapshot and rely on filesystem logs and appliapplication-based recovery like RDBMS transaction logs

snapshot [-create -delete -rollback -list]  name [-lu <list-LU-names>]     -clustername x -spname z

 Examples

snapshot -create   bkup1 -lu lu42 lu43   -clustername alpha -spname alpha

snapshot -rollback bkup1    -clustername alpha -spname alpha

snapshot -delete    bkup1    -clustername alpha -spname alpha

Shared Storage Pool – Weekly Configuration & Monitoring

Configuration Details

• cluster -list                  <-- yields cluster name
  
  cluster -status -clustername alpha
  
  cluster -status -clustername alpha -verbose <-- shows you the poolname
  
  lscluster -d <-- yields all the hdisks with local names for each VIOS

Monitor pool use

lssp -clustername alpha

lssp -clustername alpha -sp alpha -bd

• Note that this command uses "-sp" and not "-spname" like many others.

Monitor for issues

alert -set -type {threshold | overcommit} -value N

alert -list

• Ongoing monitoring of VIOS logs for warnings

• Note - Pool Alert Events are logged to your HMC, which you can get emailed to people.
  Look for Resource-Name=VIOD_POOL  Description=Informational Message

Shared Storage Pool – Quarterly or  Yearly Maintenance

Pool mirroring check

failgrp -create       <-- once only when creating the pool
 
failgrp -list [-verbose]

Growing the pool size and monitoring

 pv -list [-verbose]
 pv -list -capable    <-- check new LUN ready
 pv -add -fg a: hdisk100 b: hdisk101

Moving the pool to a different disk subsystem

pv -oldpv hdisk100 -newpv hdisk200

Shared Storage Pool – Tiers feature added in phase 5 Nov 2015 NEW

• Tiers allow LUNs to be grouped and you decide where to place your LU or it goes to the default tier.
• You have up to 9 tiers
• One tier includes the SSP metadata and is called the SYSTEM tier

Good reasons for Tiers

1. Fast new disk Units, slower last years disk, and slow older disk units  
2. Flash systems, regular disks, and cheap'n'cheerful disks
3. Production, in-house, and test 
4. Computer room A, B, and C so the tier reminds you where the disks are

You can mirror (have a second failgrp) at the tier level

Tier commands

tier -create -tier blue: <list of LUNs>
lu -create -lu fred -size 32G -tier blue

•  Without the -tier, the LU is created in the tier marked as default
• SSP LU moves a LU between tiers with:  lu -move -lu vm42 -dsttier blue
• SSP LU Resize (grow only, saves admin time) also added with tiers:  lu -resize-lu vm42 -size 128G
• SSP + Tier support in HMC Enhanced+ view
• VIOS alerts escalated to HMC for Tiers

Shared Storage Pool – Phase 6 November 2016 NEW

New features for this release but many are internal with no new commands

1. Increased SSP VIOS nodes from 16 to 24 - normally this means 12 Servers with dual VIOS
2. Fully support the DeveloperWorks SSP Disaster Recovery software see the viosbr command options for details
3. RAS (Reliable, Available & Serviceable)
   • 3a Cluster-Wide Snap Automation - simpler collecting diag details across the whole cluster

     clffdc -c FULL -p 2

     Saves to /home/ios/logs/ssp_ffdc/

   • 3b Asymmetric Network Handling - let VIOS eviction from the cluster is 1 node has an issue
   • 3c Lease by Clock Tick - no more requiring removing the SSP from the SSP cluster (clstatstop) before changing the time 
4. Find out whether an LU is mapped to an LPAR:  lu -list -attr provisioned=<true/false> 
5. HMC further GUI support for SSP? Arrives with HMC 860 and more with HMC 870
6. HMC Performance & Capacity Metrics (SSP Performance stats) See other AIXpert Blogs for details.

• Nigel's YouTube Channel
• AIXpert Blog

Worldwide