HP 3PAR F-Class, T-Class, StoreServ 10000 Storage Troubleshooting Guide (QL226-96953, May 2013)

Format of Possible LD Exception Messages
LD ld:<ldname> "LD is not mapped to a volume"
LD ld:<ldname> "LD is in write-through mode"
LD ld:<ldname> "LD has <X> preserved RAID sets and <Y> preserved chunklets"
LD ld:<ldname> "LD has reduced availability. Current: <cavail>, Configured: <avail>"
LD ld:<ldname> "LD does not have a backup"
LD ld:<ldname> "LD does not have owner and backup"
LD ld:<ldname> "Logical Ddisk is owned by <owner>, but preferred owner is <powner>"
LD ld:<ldname> "Logical Disk is backed by <backup>, but preferred backup is <pbackup>"
LD ld:<ldname> "A logging LD is smaller than 20G in size"
LD ld:<ldname> "Detailed State:<ldstate>" (degraded or failed)
LD -- "Number of logging LD's does not match number of nodes in the cluster"
LD -- "Preserved data storage space does not equal total node's Data memory"
LD Example 1
Component -------Description-------- Qty
LD LDs not mapped to a volume 10
Component -Identifier-- --------Description---------
LD ld:Ten.usr.0 LD is not mapped to a volume
LD Suggested Action 1
Examine the identified LD(s) using CLI commands such as showld, showld –d, showldmap,
showvvmap, etc.
LDs are normally mapped to (used by) VVs but they can be disassociated with a VV if a VV is
deleted without the underlying LDs being deleted, or by an aborted tune operation. Normally, you
would remove the unmapped LD to return its chunklets to the free pool.
cli% showld Ten.usr.0
Id Name RAID -Detailed_State- Own SizeMB UsedMB Use Lgct LgId WThru
MapV
88 Ten.usr.0 0 normal 0/1/2/3 8704 0 V 0 --- N
N
cli% showldmap Ten.usr.0
Ld space not used by any vv
LD Example 2
Component -------Description-------- Qty
LD LDs in write through mode 3
Component -Identifier-- --------Description---------
LD ld:Ten.usr.12 LD is in write-through mode
LD Suggested Action 2
Examine the identified LD(s) using CLI commands such as showld, showld –d, showldch, and
showpd for any failed/missing disks. Write-through mode (WThru) indicates that host I/O operations
must be written through to the disk before the host I/O command will be acknowledged. This is
108 Troubleshooting