Defects in nbdecommission will result in removal of more credentials than intended when the command is used to remove a media server with MSDP credentials. Storage units may also be removed. EEBs are available.

  • Article ID:100000462
  • Modified Date:
  • Product(s):

Severity

Data Loss

Description

Document history

Date Description
18 December 2015 Initial publication
2 February 2016 Updated to include fix availability in a released version
24 March 2017 Updated to clarify that the fix is included in 8.0 and patches are available for earlier versions

 

Affected versions

These defects (Etracks 3843887, 3862454, 3878426, 3882313) potentially affect the following versions of NetBackup and NetBackup Appliance:

NetBackup NetBackup Appliance
7.6.1.2 2.6.1.2
7.7 N/A
7.7.1 2.7.1
7.7.2 2.7.2
7.7.3 2.7.3


When using nbdecommission to remove a media server with credentials for a Media Server Deduplication Pool (MSDP), ALL credentials for ALL storage servers are removed.

Depending on configuration, storage units for other OST (OpenStorage Technology) vendors will also be removed when the media server being decommissioned had credentials to that storage server as well.

 

Example of this issue

Suppose an enterprise has two media servers in use:
  • knothead is a media server which is not being decommissioned
  • heisenberg is a media server which is being decommissioned
  • both servers have Media Server Deduplication
  • each server is the fingerprint server to the other
  • each has credentials to a DataDomain storage server
 
# nbdecommission –oldserver heisenberg

Storage Server: knothead
Server Type: PureDisk (9)
Disk Pool: dp_k_msdp
Media Server: knothead
Media Server: heisenberg
Disk pool dp_k_msdp (PureDisk) is referenced by the following storage units:
    dp_k_msdp-stu



Storage Server: heisenberg
Server Type: PureDisk (9)
Disk Pool: dp_h_msdp
Media Server: knothead
Media Server: heisenberg

Processing OST host credentials
<- first defect
Deleting OST specific creds for STS host knothead, Media Server knothead
Deleting OST specific creds for STS host heisenberg, Media Server knothead
Deleting OST specific creds for STS host datadomain670a.example.com, Media Server knothead


The resulting action should not have removed knothead from knothead (MSDP storage server) and also should not have removed knothead from datadomain670a (DD storage servr).

Disk pool dp_h_msdp (PureDisk) is referenced by the following storage units:
    dp_h_msdp-stu


Storage unit dp_h_msdp-stu: host heisenberg
De-activating policies using storage unit dp_h_msdp-stu
Deleting storage unit dp_h_msdp-stu on host _STU_NO_DEV_HOST_
Deleted storage unit dp_h_msdp-stu
Deleted PureDisk disk pool dp_h_msdp
Deleted PureDisk storage server heisenberg

Storage Server: datadomain670a.example.com
Server Type: DataDomain (9)
Disk Pool: dp_dd670a
Media Server: knothead
Media Server: heisenberg
Disk pool dp_dd670a (DataDomain) is referenced by the following storage units:  
    dp_datad670a-stu


Storage unit dp_datad670a-stu: host heisenberg
<- second defect
De-activating policies using storage unit dp_datad670a-stu                      
Deleting storage unit dp_datad670a-stu on host _STU_NO_DEV_HOST_ <==            
Deleted storage unit dp_datad670a-stu

Action Required

A fix for this issue is included in NetBackup 8.0. Patches are available for earlier versions.

Workaround

If updating to a resolved version is unfeasible, before using nbdecommission, please contact Veritas technical support, referencing this document and Etrack 3862454 (7.6.1.2), 3843887 (7.7.1), 3878426 (7.7.2) or 3882313 (7.7.3) to obtain an Emergency Engineering Binary (EEB) containing a fix for this issue.

If nbdecommission has already been run and credentials were unintentionally removed, please contact Veritas technical support, referencing this document ID, for further instructions on restoring lost credentials.


Best Practices

Veritas recommends the following best practices:
1. Always perform a full DR backup prior to making any changes to your environment.
2. Always make sure that your environment is running the latest version and patch level.
3. Perform periodic "test" restores.
4. Subscribe to technical articles.
 

Was this content helpful?

Get Support