Scribd
Upload
106 views4 pages

EMC Symm Data Devs Primus

The document discusses deleting SAV (log) devices or TDAT (thin) devices from a Symmetrix VMAX array that results in a "red box" error message stating the devices are in use. It provides the steps needed to correctly delete these devices, including draining and removing them from their pool, and checking for allocations before deletion to avoid data loss or corruption.

Uploaded by

Shashi Kanth
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
106 views4 pages

EMC Symm Data Devs Primus

The document discusses deleting SAV (log) devices or TDAT (thin) devices from a Symmetrix VMAX array that results in a "red box" error message stating the devices are in use. It provides the steps needed to correctly delete these devices, including draining and removing them from their pool, and checking for allocations before deletion to avoid data loss or corruption.

Uploaded by

Shashi Kanth
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 4

"Symmetrix VMAX: Red box: "You are about to affect devices that are currently in use" when deleting

SAV or TDAT devices."

Question: How can I delete SAV devices or TDAT's (Thin devices)?


Environment: Product: Symmetrix VMAX Series
Environment: Enginuity: 5875
Environment: Enginuity: 5874
Environment: Configuration: SAV Devices
Environment: Configuration: DATA device (TDAT)
Environment: EMC SW: TimeFinder/SNAP
Red box: "You are about to affect devices that are currently in use" when deleting
Problem:
SAV or TDAT devices.
Red box (step VolRemove_Avoid_SAV_Corruption):
Script 0000070C: Message 0x00000AE0.0x118B:
You are about to affect devices that are currently in use.
Problem: Log Devices: [YYY..ZZZ].
Thin DATA devices: [AAA..BBB].
This will cause DATA LOSS OR DATA CORRUPTION.
The active volumes are listed in the file stderr.log
Red box occurs during an online deletion of SAV/TDAT devices with error
Problem: message "You are about to affect devices that are currently in use". However, the
devices had never been used.
Change: Online configuration change
Change: Volume Deletion
This step in the script performs a check on the log/DATA devices in the pool.
Script checks to determine if there are Allocations on the Log/DATA devices,
checks their current pool status and pool offset. There should be no data on the
device(s) being deleted. The current pool state should be 'INAC' and device
Root Cause: should be in the Default Data Device Pool at offset 0x100. These three steps are
needed to delete DATA devices: 1. Drain the device, 2. disable device, 3. remove
the device and if all DATA devices are being removed, then THIN devices should
be Unbound from the pool. If any of these checks fail, the script will redbox with
the above message.
Fix: Confirm that devices are not in use. The user must perform the following using Config
Manager on the affected devices:
1. Deactivate.
2. Start the Drain.
3. Wait until the drain completes.
4. Remove the device from the pool.
5. Delete the device.

The command syntax to achieve this is as follows (note the disable command below will
perform the deactivate and start the drain process). These commands should be run by
the customer:

For SAV devices:

1. symconfigure -sid XXXX -cmd "disable dev YYY:ZZZ in pool


Pool_Name, type=SAVEDEV;" commit (this command will deactivate
and start to drain the device).
2. symconfigure -sid XXXX -cmd "remove dev YYY:ZZZ in pool
Pool_Name, type=SNAP;" commit (This command will remove the
device from the current pool and move it to the default pool).
3. The customer should run a query command to make sure their are no
allocated tracks against the device.
4. The status of the device to be removed needs to be inactive/disabled then
the customer can issue the command to delete the device from the SAV
pool.

For TDAT devices:

1. symconfigure -sid XXXX -cmd "disable dev aaa:bbb in pool


Pool_Name, type=thin;" commit (this command will deactivate and
start to drain the device).
2. symconfigure -sid XXXX -cmd "remove dev aaa:bbb from
Pool_Name, type=thin;" (This command will remove the device from
the current pool and move it to the default pool).
3. Then the customer can delete the device.

4. Once done, you can restart the script and resume from failed step, it will
now pass.
PSE/SAC: You can run these commands to confirm that there are no THIN devices
bound to the pool and the DATA devices have no allocations:

1. 8D,,,THIN,POOL,,<pool_offset> - display output will be blank


2. 8D,,,DATA,,data_device - keep hitting enter and verify there is no tracks under
the USED column.

Fix: 3. 98,POOL,SHWD,data_device - keep hitting enter and verify all your devices
are in RED and showing state of INAC

If the above 3 steps are done and the script is still failing, then check and see if the
devices being removed are in the DF_DDEV_POOL pool, you can see this from step 3
above. If they are NOT showing at offset 0100, then have the customer run the
symconfigure command to 'remove dev' from their current pool. Once done, resume
from failed step and the step should be run and pass.
The Inlines command 98,'pool','SHWD',dev can be used to display whether a device is
associated with a pool. Devices should have a status of inactive (i.e., STAT =INAC). If
Notes:
the status is active (i.e., ACTV) then it can be deactivated as shown above. Refer to
solution emc170275 for additional information regarding determining the device status.

"You're about to delete 1 Thin DATA devices. This will cause DATA LOSS OR DATA CORRUPTION."

Environment
Product: Symmetrix DMX-3
:
Environment
Product: Symmetrix DMX-4
:
Environment
Enginuity: 5773
:
Environment
Configuration: Thin Device
:
Silconfig.log file shows: "You're about to delete 1 Thin DATA
Problem:
devices. This will cause DATA LOSS OR DATA CORRUPTION."
Red box: "You're about to delete 1 Thin DATA devices. This
Problem:
will cause DATA LOSS OR DATA CORRUPTION."
Red box: Delete_vols
(sub_Prepare_Delete_Vols.VolRemove_Avoid_SAV_Corruption
):
Problem: Message 0x00000AE0.0x118B:
You''re about to delete 1 Thin DATA devices.
This will cause DATA LOSS OR DATA CORRUPTION.
The active volumes are listed in the file stderr.log
Change: Volume Deletion
If any Thin device or Thin Data devices are included in the online deletion plan,
Root Cause:
prior to the deletion, the customer has to follow a few standard procedures.
When deleting a Thin or a Thin Data device the following must be done prior to
deleting the devices:
1. THIN Device: A THIN device can be deleted only after it is unbound from
the respective pool.
Fix:
2. THIN DATA Device: A THIN DATA device can be deleted once it is
disabled and removed from the pool. Upon disabling the device, 'draining'
will be automatically started and the number of allocated track count will
become zero.

EMC Support:

Download the onlupg.log to confirm the script failure and the failed step. Also make a note
of the devices from stderr.log
 Confirm the device is not in use and disabled in the pool. Issue
98,'POOL','SHWD',DEV in inlines, the device listed in stderr.log should be
inactive . The TDAT Device will be in red and 'STAT = INAC' . If the device is
still active,have the customer disable the devices and then resubmit the change.
 If the device is already disabled and 8D,,,THIN,,DEV shows non zero number of
allocated tracks. You may need to wait and if the track count reduces to zero, the
script can be resumed and it should not report DU/DL.
Fix:
 If either of the below scenario is matching, escalate to a thin provisioning expert to
fix the issue before retrying the commit.

o If the device is already disabled and allocated track count in


8D,,,THIN,,DEV is '0' but still the script reports data loss.

o 8D,DEV,1,BIND shows pool offset '0000' for all the pools.

If at any point you did not understand this solution engage a SME/PSE3.

You might also like