Pair condition and recovery: primary storage system failure

Global Active Device for VSP One Block

Version

10.5.x

Audience

anonymous

Part Number

MK-23VSP1B011-10

ft:lastEdition

2026-05-04

When the primary storage system fails, you can correct the issue by recovering the primary storage system, recovering the physical path between primary and secondary storage systems, and recovering the pair.

GAD pairs might be created between the primary and secondary sites that configure GAD as follows:

A P-VOL of a GAD pair is at site 1 and a S-VOL is at site 2.
A P-VOL of a different GAD pair is at site 2 and a S-VOL is at site 1.

In this case, recover from a failure at the secondary storage system. For more information, see Pair condition and recovery: secondary storage system failure.

The following table shows transitions for pair status and I/O mode, the volumes accessible from the server, and location of the latest data when you can no longer use the primary storage system due to failure.

Before failure		After failure
Pair status and I/O mode		Pair status and I/O mode		Volume accessible from the server		Volume with latest data
P-VOL	S-VOL	P-VOL¹	S-VOL	P-VOL	S-VOL	Volume with latest data
PAIR (Mirror (RL))	PAIR (Mirror (RL))	PSUE (Block)	SSWS (Local)²	NG	OK	S-VOL
PAIR (Mirror (RL))	PAIR (Mirror (RL))	PSUE (Block)³	PSUE (Block)³	NG	NG	Both P-VOL and S-VOL
PAIR (Mirror (RL))	PAIR (Block)	PSUE (Block)	PSUE (Block)	NG	NG	Both P-VOL and S-VOL
COPY (Mirror (RL))	COPY (Block)	PSUE (Local)	COPY (Block)	NG	NG	P-VOL
PSUS/PSUE (Local)	SSUS/PSUE (Block)	PSUS/PSUE (Local)	SSUS/PSUE (Block)	NG	NG	P-VOL
PSUS/PSUE (Block)	SSWS (Local)	PSUS/PSUE (Block)	SSWS (Local)	NG	OK	S-VOL
Notes: If shared memory in the primary storage system becomes volatilized, the P-VOL status changes to SMPL, and the GAD reserve attribute is set for the volume, which prevents host access to the volume. If the server does not issue write I/O, the pair status is PAIR (Mirror (RL)). If either of the following conditions is met, the P-VOL might become PSUE (Block) and the S-VOL might become PSUE (Block): No volume is set for the quorum disk. A quorum disk failure occurs.

SIMs

Primary storage system: SIM varies depending on the failure type
Secondary storage system: 2180xx, DD0xyy, DD3xyy

When the primary storage system is powered off, delete a redundant path (logical path) to the P-VOL, and then turn on the power.
1. Using the multi-path software, distinguish the volumes which are not able to be accessed from the server.
2. Using the multi-path software, delete the redundant paths to the P-VOL.
  If you cannot delete the redundant paths, detach all physical paths which are connected to the server at the primary site.
Turn on the primary storage system.
Recover the primary storage system.

For details, contact customer support.
Recover the physical path between the primary storage system and the secondary storage system.
If S-VOL pair status is PAIR, suspend the pair specifying the S-VOL.
Confirm the quorum disk status. If the quorum disk is blocked, recover from the blockade. If you do not set a volume for the quorum disk, you can skip this step.

Resynchronize or re-create a pair using the following procedure whose pair status and I/O mode match your pair's status and I/O mode.

If a failure occurs after the one volume capacity of a GAD pair can be expanded, the creation, resync, swap resync, and horctakeover operations of the GAD pair cannot be performed because the capacity of both the volumes is not the same. Make sure to expand the other volume capacity so that the capacity of both the volumes is the same, and then retry the operation.

If the failed primary site contains a P-VOL, recover from a failure in accordance with the following procedure. If the failed primary site also contains a S-VOL, recover from a failure at the secondary site. For more information, see Pair condition and recovery: secondary storage system failure.

Pair status		I/O mode
P-VOL	S-VOL	P-VOL	S-VOL
PSUE	COPY	Local	Block

Delete the pair forcibly from the S-VOL.
```
pairsplit -g oraHA -d dev1 -RF -IH2
```
When you perform this step, delete the virtual LDEV ID so that the volume cannot be accessed from the server.
Check if the virtual LDEV ID of the P-VOL is not deleted.
```
raidcom get ldev -ldev_id 0x1111 -fx -IH1
(Omitted)
LDEV:1111
(Omitted)
```
The VIR_LDEV information is not displayed if it is same as the LDEV information. If the virtual LDEV ID is deleted, set a correct LDEV ID.
Delete the pair forcibly from the P-VOL.
When you perform this step, make sure not to delete the virtual LDEV ID, which allows the volume to be accessed from the server.
```
pairsplit -g oraHA -d dev1 -SFV -IH1
```
Check if the virtual LDEV ID of the S-VOL indicates GAD reserve.
```
raidcom get ldev -ldev_id 0x2222 -fx -IH2
(Omitted)
LDEV : 2222
VIR_LDEV : ffff
(Omitted)
```
VIR_LDEV : ffff indicates GAD reserve. If it shows another value, set a correct virtual LDEV ID.
Re-create the GAD pair by specifying the P-VOL.
```
paircreate -g oraHA -d dev1 -f never -vl -jq 1 -IH1
```
If a failure occurs after the one volume capacity of a GAD pair can be expanded, the creation, resync, swap resync, and horctakeover operations of the GAD pair cannot be performed because the capacity of both the volumes is not the same. Make sure to expand the other volume capacity so that the capacity of both the volumes is the same, and then retry the operation. If you accidentally delete the virtual LDEV ID of the P-VOL, you cannot create a GAD pair again. See Setting correct virtual LDEV ID values and set the correct value for the virtual LDEV ID.

Pair status		I/O mode
P-VOL	S-VOL	P-VOL	S-VOL
SMPL	COPY	Not applicable	Block

Delete the pair forcibly from the S-VOL.
When you perform this step, delete the virtual LDEV ID so that the volume cannot be accessed from the server.
```
pairsplit -g oraHA -d dev1 -RF -IH2
```
Check if the virtual LDEV ID of the S-VOL indicates GAD reserve.
```
raidcom get ldev -ldev_id 0x2222 -
fx -IH2
(Omitted)
LDEV : 2222
VIR_LDEV : ffff
(Omitted)
```
VIR_LDEV : ffff indicates GAD reserve. If it shows another value, set a correct virtual LDEV ID.
Check if the virtual LDEV ID of the P-VOL is not deleted.
```
raidcom get ldev -ldev_id 0x1111 -
fx -IH1
(Omitted)
LDEV : 1111
(Omitted)
```
The VIR_LDEV information is not displayed if it is same as the LDEV information. If the virtual LDEV ID is deleted, set a correct LDEV ID.

Note: If shared memory in the secondary storage system becomes volatilized, the S-VOL pair status changes to SMPL, and GAD reserve is assigned to the virtual attribute.
Re-create the GAD pair by specifying the P-VOL.
```
paircreate -g oraHA -d dev1 -f
never -vl -jq 1 -IH1
```
If a failure occurs after the one volume capacity of a GAD pair can be expanded, the creation, resync, swap resync, and horctakeover operations of the GAD pair cannot be performed because the capacity of both the volumes is not the same. Make sure to expand the other volume capacity so that the capacity of both the volumes is the same, and then retry the operation. If you accidentally delete the virtual LDEV ID of the P-VOL, you cannot create a GAD pair again. See Setting correct virtual LDEV ID values and set the correct value for the virtual LDEV ID.

Pair status		I/O mode
P-VOL	S-VOL	P-VOL	S-VOL
PSUS/PSUE	SSWS	Block	Local

Resynchronize the pair specifying the S-VOL.
```
pairresync -g oraHA -d dev1 -swaps -IH2
```

Pair status		I/O mode
P-VOL	S-VOL	P-VOL	S-VOL
SMPL	SSWS	Not applicable	Local

Check if the virtual LDEV ID of the P-VOL indicates GAD reserve.
```
raidcom get ldev -ldev_id 0x1111 -fx -IH1
(Omitted)
LDEV : 1111
VIR_LDEV : ffff
(Omitted)
```
VIR_LDEV : ffff indicates reserve. If it shows another value, set a correct virtual LDEV ID.
Delete the pair forcibly from the S-VOL.
When you perform this step, make sure not to delete the virtual LDEV ID, which allows the volume to be accessed from the server.
```
pairsplit -g oraHA -d dev1 -RFV -IH2
```

Check if the virtual LDEV ID of the S-VOL is not deleted.

raidcom get ldev -ldev_id 0x2222 -
fx -IH2
(Omitted)
LDEV : 2222
VIR_LDEV : 1111
(Omitted)

Re-create the pair specifying the S-VOL.
```
paircreate -g oraHA -d dev1 -f never -vl -jq 1 -IH2
```
If a failure occurs after the one volume capacity of a GAD pair can be expanded, the creation, resync, swap resync, and horctakeover operations of the GAD pair cannot be performed because the capacity of both the volumes is not the same. Make sure to expand the other volume capacity so that the capacity of both the volumes is the same, and then retry the operation. If you accidentally delete the virtual LDEV ID of the S-VOL, you cannot create a GAD pair again. See Setting correct virtual LDEV ID values and set the correct value for the virtual LDEV ID.

Pair status		I/O mode
P-VOL	S-VOL	P-VOL	S-VOL
PSUS/PSUE	SSUS/PSUE	Local	Block

Resynchronize the pair specifying the P-VOL.
```
pairresync -g oraHA -d dev1 -IH1
```

Pair status		I/O mode
P-VOL	S-VOL	P-VOL	S-VOL
SMPL	SSUS/PSUE	Not applicable	Block

Delete the pair forcibly from the S-VOL.
When you perform this step, make sure you delete the virtual LDEV ID so that the volume cannot be accessed from the server.
```
pairsplit -g oraHA -d dev1 -RF -IH2
```
Check if the virtual LDEV ID of the S-VOL indicates GAD reserve.
```
raidcom get ldev -ldev_id 0x2222 -
fx -IH2
(Omitted)
LDEV : 2222
VIR_LDEV : ffff
(Omitted)
```
VIR_LDEV : ffff indicates GAD reserve. If it shows another value, set a correct virtual LDEV ID.
Check if the virtual LDEV ID of the P-VOL is not deleted.
```
raidcom get ldev -ldev_id 0x1111 -
fx -IH1
(Omitted)
LDEV : 1111
(Omitted)
```
The VIR_LDEV information is not displayed if it is same as the LDEV information. If you cannot find the virtual LDEV ID, set a correct virtual LDEV ID.

Note: If shared memory in the secondary storage system becomes volatilized, the S-VOL pair status changes to SMPL, and GAD reserve is assigned to the virtual attribute.
Re-create the pair specifying the P-VOL.
```
paircreate -g oraHA -d dev1 -f never -vl -jq 1 -IH1
```
If a failure occurs after the one volume capacity of a GAD pair can be expanded, the creation, resync, swap resync, and horctakeover operations of the GAD pair cannot be performed because the capacity of both the volumes is not the same. Make sure to expand the other volume capacity so that the capacity of both the volumes is the same, and then retry the operation. If you accidentally delete the virtual LDEV ID of the P-VOL, you cannot create a GAD pair again. See Setting correct virtual LDEV ID values and set the correct value for the virtual LDEV ID.

Pair status		I/O mode
P-VOL	S-VOL	P-VOL	S-VOL
PSUE	PSUE	Block	Block

See Forcibly changing the I/O mode of GAD pairs for recovery.

Add a redundant path if a redundant path to the P-VOL is deleted.
1. If the physical paths to the server at the primary site are removed, restore all physical paths to the server at the primary site, and then add a redundant path.
2. Use the multi-path software to add the redundant path to the P-VOL, which is deleted in step 1.