HP XP24000 HP Replication Adapter for VMware vCenter Site Recovery Manager Adm - Page 17

HP P9000 RAID Manager User Guide, In Windows, the RAID Manager provides

Get HP XP24000 PDF manuals and user guides View all HP XP24000 manuals
Add to My Manuals
Save this manual to your list of manuals

Page 17 highlights

This testFailover configuration is required the conditions below. - To be able to communicate between HORCM of primary site and HORCM of recovery site. - In NON stopping test of APP, CA Sync CTG/CAJ volumes must be used. Snapshot volume for testFailover The snapshot volume for testFailover must be created as a volume using Point-in-time-Split, and must be set as a write disable volume after testFailover is completed. If datastore is configured as a single LUN, it does not need to be created using Point-in-time-Split. If you are using BC, the PAIR/COPY state is write disable after testFailover. For example: C:\HORCM\etc> paircreate -g -m grp ... If you are using CoW, the PAIR/COPY state is read/write disable after testFailover. For example: C:\HORCM\etc> paircreate -g -m grp ... If you are switching to CA/CAJ_SVOL using = true, the PAIR/COPY state is write disable after testFailover. For example: C:\HORCM\etc> paircreate -g ... If read/write disable after testFailover is required, set it up as follows: • Set the variable RMSRA_USE_NOREAD=1. • Install DRU (Open LDEV Guard) The testFailover command sets the read/write disable state for CA/CAJ_SVOL using raidvchkset -vg rdb. This state must be reset by the testfailover or the failover or the raidvchkset -g -vg command. Required LUN ID SRM recognizes the LUN WWN returned from the array using the SCSI information page 0x83. The Device Identification must include "NAA version 6 information". Protection groups using CA Sync/CA Sync CTG If the PVOLs for protection groups use CA Sync, HP recommends that you use data as the fence level. If the fence level used is never, internal horctakeover fails with EX_VOLCUR, because fence level never cannot completely guarantee data consistency. According to SRM/VMware, the goal of Failover/testFailover is to boot the VM's. It can then accept fence level never and RMSRA20 succeeds in failing over, even though horctakeover returns EX_VOLCUR. If you want to use Never as the fence level, consider the recovery ability of applications such as SQL, Exchange, and Oracle. They could be impacted if you use fence level NEVER. The PVOL and SVOL does not go into write protect if the link fails during failover when NEVER is used. CAUTION: Caution: Fence NEVER does not place the PVOL and SVOL into write protect if the link fails during failover. Protection groups and CT groups using CA Async/CAJ If you are using CT groups, make sure your SRM protection group lUNs match your CT group lUNs. CT groups cannot be used in a multiple array (MxN configuration). Starting HORCM service In Windows, the RAID Manager provides the svcexe.exe command and a sample script file (HORCM0_run.txt) so that the RAID Manager (HORCM) can be started automatically from the service. For details, see the HP P9000 RAID Manager User Guide. RMSRA20 configurations 17

  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31
  • 32
  • 33
  • 34
  • 35
  • 36
  • 37
  • 38
  • 39
  • 40
  • 41
  • 42
  • 43
  • 44
  • 45
  • 46
  • 47
  • 48
  • 49
  • 50
  • 51
  • 52
  • 53
This testFailover configuration is required the conditions below. - To be able to communicate
between HORCM of primary site and HORCM of recovery site. - In NON stopping test of APP,
CA Sync CTG/CAJ volumes must be used.
Snapshot volume for testFailover
The snapshot volume for
testFailover
must be created as a volume using Point-in-time-Split,
and must be set as a
write disable
volume after
testFailover
is completed. If datastore is
configured as a single LUN, it does not need to be created using Point-in-time-Split.
If you are using BC, the PAIR/COPY state is
write disable
after
testFailover<stop>
. For
example:
C:\HORCM\etc> paircreate
–
g <group>
–
m grp
…
If you are using CoW, the PAIR/COPY state is
read/write disable
after
testFailover<stop>
.
For example:
C:\HORCM\etc> paircreate
–
g <group> -m grp
…
If you are switching to CA/CAJ_SVOL using
<SplitReplication> = true
, the PAIR/COPY
state is
write disable
after
testFailover<stop>
. For example:
C:\HORCM\etc> paircreate
–
g <group>
…
If read/write disable after
testFailover<stop>
is required, set it up as follows:
•
Set the variable
RMSRA_USE_NOREAD=1
.
•
Install DRU (Open LDEV Guard)
The
testFailover<stop>
command sets the read/write disable state for CA/CAJ_SVOL using
raidvchkset -vg rdb
. This state must be reset by the
testfailover<start>
or the
failover
or the
raidvchkset
–
g <group> -vg
command.
Required LUN ID
SRM recognizes the LUN WWN returned from the array using the SCSI information page 0x83.
The Device Identification must include "NAA version 6 information".
Protection groups using CA Sync/CA Sync CTG
If the PVOLs for protection groups use CA Sync, HP recommends that you use
data
as the fence
level. If the fence level used is
never
, internal
horctakeover
fails with EX_VOLCUR, because
fence level
never
cannot completely guarantee data consistency. According to SRM/VMware,
the goal of
Failover/testFailover
is to boot the VM’s. It can then accept fence level
never
and RMSRA20 succeeds in failing over, even though
horctakeover
returns EX_VOLCUR.
If you want to use Never as the fence level, consider the recovery ability of applications such as
SQL, Exchange, and Oracle. They could be impacted if you use fence level NEVER. The PVOL
and SVOL does not go into write protect if the link fails during failover when NEVER is used.
CAUTION:
Caution: Fence NEVER does not place the PVOL and SVOL into write protect if the
link fails during failover.
Protection groups and CT groups using CA Async/CAJ
If you are using CT groups, make sure your SRM protection group lUNs match your CT group
lUNs. CT groups cannot be used in a multiple array (MxN configuration).
Starting HORCM service
In Windows, the RAID Manager provides the
svcexe.exe
command and a sample script file
(
HORCM0_run.txt
) so that the RAID Manager (HORCM) can be started automatically from the
service. For details, see the
HP P9000 RAID Manager User Guide
.
RMSRA20 configurations
17