Support for Solaris MPxIO in conjunction with Veritas Dynamic Multi-Pathing (DMP)

Problem

This article describes Veritas Support of Solaris MPxIO in conjunction with Veritas Dynamic Multi-Pathing (DMP).
 

Solution
 

This article is provided as formal documentation that Solaris MPxIO interoperability with Veritas Dynamic Multi-Pathing (DMP) is supported by the Storage Foundation in relation to specific key requirements being implemented.
 

Veritas provides support if you run Solaris MPxIO in conjunction with Veritas Dynamic Multi-Pathing (DMP) as outlined in the configuration requirements above. 

MPxIO is not supported in Solaris LDOM (Oracle VM for SPARC) environments, I/O domains & LDOM guests
MPxIO is not supported when Veritas DMP_NATIVE_SUPPORT is enabled

To confirm if MPxIO is disabled, type:

# stmsboot -L
stmsboot: MPXIO disabled
 

DMP is included with Storage Foundation & InfoScale and is the preferred multi-pathing solution for device presentation. 

DMP is thoroughly tested with all arrays listed in the Veritas Hardware Compatibility List (HCL).  Check the HCL to confirm the supportability standpoint for a given array.

Things to be considered before deploying MPxIO in relation to Storage Foundation:

6.0.x:

When considering to deploy MPxIO in relation to Storage Foundation (SF) 6.0.1 onwards, please kindly note that Veritas only validated the MPxIO solution in relation to simple coexistence and basic functionality testing with some common array types.   In the event that interoperability issues should arise in relation to the coexistence between DMP and MPxIO,

Veritas will assist Sun/Oracle with troubleshooting such events. If Veritas identifies the issue as being out of its control, it will provide the necessary information to reproduce the error to the appropriate third party. In the situation where interoperability issues arise that require code modification or patches to third party products, it is at the third party's discretion to implement any such recommendations or required changes.

To determine the source of any defect, you might be requested/required to recreate any issue without MPxIO in the stack to confirm that the issue occurs independent of MPxIO interoperability.  Veritas will provide commercially reasonable effort to support such implementations which may be encountered by MPxIO related deployments, but assumes no liability for any user data impacted by the use of MPxIO.

InfoScale 7.1 onwards:
 

NOTE: The DMP tunable "dmp_monitor_osevent" has been removed from InfoScale 7.1 and onwards.
There is no additional tunable to replace "dmp_monitor_osevent".

This statement of support is applicable only to Storage Foundation, Storage Foundation High Availability (SFHA) and InfoScale product releases:
 

Software Limitations:

• SCSI-3 Persistent Reservation fencing is supported only with DMP and the vxfen driver. 
   
• SCSI-3 Persistent Reservation fencing is not supported when MPxIO is operating beneath DMP, and SCSI-3 reservations created outside the vxfen driver are not supported.
   
• Storage Foundation for Oracle RAC is not supported due to fencing limitations.
   
• The DMP dmp_fast_recovery tunable must be disabled with DMP and MPxIO.
   
• The DMP dmp_monitor_osevent tunable must be disabled with DMP and MPxIO.
  Not applicable to InfoScale 7.1 onwards.
   
• Turn off persistent naming when using MPxIO (vxddladm set namingscheme=ebn persistence=no)
   
• The Thin Reclamation feature in Storage Foundation is not supported with MPxIO
   
• Advanced Reporting is only supported with DMP
   
• DMP Array Volume Identifier (AVID) naming is only supported with DMP when no other third party multipathing (TPD) manages the disk.
   
• On Solaris configurations affected by Oracle Bug 15468112 ( Sun Bug ID 6680575) Storage Foundation High Availability on MPxIO devices with Sun/Oracle Active/Passive storage hardware is not supported.

DMP tunables:
 

The DMP dmp_fast_recovery tunable controls whether DMP should attempt to obtain SCSI error information directly from the HBA interface (whereby DMP will bypass the sd/scsi stack and send path inquiry/status CDBs directly from the HBA in order to bypass long SCSI queues and recover paths faster) . Setting the value to "on" can potentially provide faster error recovery, provided that the HBA interface supports the error enquiry feature. If set to off, the HBA interface is not used. The default setting is on.

The recommendation to turn off the DMP dmp_fast_recovery tunable in configurations involving MPxIO was based on the fact that MPxIO is different from EMC PowerPath (PP) as MPxIO integrates with the SUN SCSI driver (ssd) and exposes only a single OS device path for DMP to operate on.

EMC PowerPath (PP) is a path non-suppressing  multipathing driver whereby the OS and DMP will see the special PP (emcpower#) pseudo device and the corresponding OS device paths (handles). DMP operates on the PP managed "pseudo" device which then in turn operates on the OS device paths. When DMP detects the presence of a PP managed device, it automatically turns off the " dmp_fast_recovery " tunable. This is not the case for MPxIO managed devices.

Veritas Volume Manager (VxVM) 5.1 SP1 introduced a new DMP tunable named  " dmp_monitor_osevent " to reduce past interoperability issues with the SF stack and Third Party Drivers (TPD) such as MPxIO and EMC PowerPath. Veritas previously advised customers to disable the eventsource daemon (vxesd) in connection with TPD environments. This is no longer required as long as both DMP tunables are disabled in TPD environments.

The DMP " dmp_monitor_osevent " tunable determines whether the eventsource daemon (vxesd) monitors operating system events such as reconfiguration operations.

- If the DMP tunable is set to on, vxesd monitors operations such as attaching operating system devices.
- If the DMP tunable is set to off, vxesd does not monitor operating system operations.

Veritas strongly recommends disabling the " dmp_monitor_osevent " tunable to avoid known interoperability issues with MPxIO and EMC PowerPath, prior to InfoScale 7.1 onwards.

 

The DMP tunable settings can be displayed by using the vxdmpadm CLI interface:

# vxdmpadm gettune all

To disable the DMP tunable " dmp_fast_recovery ", type:
 
# vxdmpadm settune dmp_fast_recovery=off
 

Prior to InfoScale 7.1, disable the DMP tunable " dmp_monitor_osevent " by typing:

# vxdmpadm settune dmp_monitor_osevent=off

Note: The DMP tunable "dmp_monitor_osevent" has been removed from InfoScale 7.1 onwards:

How to disable persistence naming

MPxIO managed devices will default to enclosure based naming (ebn) as the OSN name exceeds 31 characters.

# vxddladm set namingscheme=ebn persistence=no

# vxddladm get namingscheme
NAMING_SCHEME       PERSISTENCE    LOWERCASE      USE_AVID
============================================================
Enclosure Based     No             Yes            Yes