3PAR InForm® OS 2.3.1 Concepts Guide (320-200112 Rev B, February 2010)

9.5
Precautions and Planning
InForm OS Version 2.3.1 3PAR InForm OS Concepts Guide
9.2.3 Growth Warning
When the size of the volumes that draw from a CPG reach the CPG’s growth warning, the
system generates an alert to notify you of the CPG's increasing size. This safety mechanism
provides the opportunity to take early action that may prevent snapshot volumes associated
with the CPG from experiencing failures, causing host or application write failures, and
exhausting all free space on the system.
When setting growth warnings for CPGs, it is critical to consider the number of CPGs on the
system, the total capacity of the system, and the projected rate of growth for all volumes on
the system.
The storage system does not prevent you from setting growth warnings that exceed the total
capacity of the system. For example, on a 3 TB system you can create two CPGs that each have
a growth warning of 2 TB. However, if both CPGs grow at a similar rate, it is possible for the
volumes that draw from the CPGs to consume all free space on the system before either CPG
reaches the growth warning threshold.
9.2.4 Growth Limit
If the volumes that draw from a CPG are allowed to reach the CPG’s growth limit, the system
prevents them from allocating additional space. This safety mechanism stops a runaway
application or volume from exhausting all free space available to the CPG and causing invalid
(stale) snapshot volumes and/or new application write failures for volumes associated with
that CPG. However, the storage system does not prevent you from setting growth limits that
exceed the total capacity of the system. For example, on a 4 TB system it is possible to create a
CPG with a 5 TB growth limit. Likewise, it is possible to create five CPGs, each with a 2 TB
growth limit, etc.
In addition, volumes that draw from a CPG can only use the space available to that CPG based
on the CPG's logical disk parameters. For example, if you create a CPG that only uses logical
disks that belong to controller node 0, when the virtual volumes that draw from a CPG have
filled up all space available to that CPG based on it's logical disk parameters, the following will
happen:
■ New writes to any Thinly-Provisioned Virtual Volumes (TPVVs) mapped to that CPG will
return write failures.
■ Snapshot volumes mapped to the CPG may become invalid (stale), subject to the virtual
copy policy associated with the base volume.
For base volumes with a no stale snapshots virtual copy policy, new writes to the base
volume will result in write failures.