Start a Conversation

Unsolved

This post is more than 5 years old

1531

April 2nd, 2009 03:00

Number of supported paths with Solaris MPXIO and CLARiiON CX3-80

Does anyone know if Solaris MPXIO supports more than two paths when configured with a CLARiiON CX3-80 (3.26.80.5.014)?

I followed the set-up instructions in the Installation Roadmap for CLARiiON Storage Systems and zoned each of the ports on the Sun OEM Emulex LPe11002 HBA with an SPA and SPB port. However when I enabled MPXIO on the Sun Fire X4150 (x64) running Solaris 10 it seems to have configured a multipathed device c6t6xxxx based on one of the HBA ports that has a PRIMARY associated with SPB and STANDBY associated with SPA, which is okay, but also shows two additional disk devices representing both of the paths on the other HBA port c3t5xxx. I was expecting one device but with four paths!

root@red-sun-001 # luxadm probe

Found Fibre Channel device(s):
Node WWN:50060160c1e03702 Device Type:Disk device
Logical Path:/dev/rdsk/c6t600601608B121E009460DDF7DC1EDE11d0s2
Node WWN:50060160c1e03702 Device Type:Disk device
Logical Path:/dev/rdsk/c3t5006016141E03702d0s2
Node WWN:50060160c1e03702 Device Type:Disk device
Logical Path:/dev/rdsk/c3t5006016941E03702d0s2

[root@red-sun-001 ># luxadm display /dev/rdsk/c6t600601608B121E009460DDF7DC1EDE11d0s2
DEVICE PROPERTIES for disk: /dev/rdsk/c6t600601608B121E009460DDF7DC1EDE11d0s2
Vendor: DGC
Product ID: RAID 5
Revision: 0326
Serial Num: CK200070400044
Unformatted capacity: 953344.000 MBytes
Read Cache: Enabled
Minimum prefetch: 0x0
Maximum prefetch: 0x0
Device Type: Disk device
Path(s):

/dev/rdsk/c6t600601608B121E009460DDF7DC1EDE11d0s2
/devices/scsi_vhci/disk@g600601608b121e009460ddf7dc1ede11:c,raw
Controller /dev/cfg/c2
Device Address 5006016041e03702,0
Host controller port WWN 10000000c984f978
Class secondary
State STANDBY
Controller /dev/cfg/c2
Device Address 5006016841e03702,0
Host controller port WWN 10000000c984f978
Class primary
State ONLINE

4.5K Posts

April 2nd, 2009 09:00

You need to check what the failover setting on the array is for this host. Check on PowerLin for Knowledgebase article emc99467 - this has all the operating system's failover mode settings.

Solaris for MPxIO or STMS

Initiator Type

3 (CLARiiON Open)

Arraycommpath

1 (Enabled)

Failovermode

1 ¿or- 4 - see note 12

UnitSerialNumber

LUN or Array - see note 5

NOTES:

5. The UnitSerialNumber setting is dependent on the configuration. Select the proper setting as follows: For Solaris 2.6, 2.7, 8: a. For DMP or PowerPath, determine if any of the following packages are present:
¿ Solstice DiskSuite (SUNWmdr)
¿ Sun Cluster 3.x (SUNWscr)
¿ VERITAS DBE/AC (VRTSdbac)
¿ SFRAC (VRTSdbac)
b. Perform the applicable sub-step:
¿ If none of the packages listed above are present, the UnitSerialNumber MUST be set to ARRAY.
¿ If any of the packages listed above are present, set the UnitSerialNumber as follows: - SDS, SC 3.x, SFRAC: UnitSerialNumber MUST be set to LUN - VRTSvcs with I/O fencing: UnitSerialNumber MUST be set to LUN - VRTSvcs without I/O fencing: UnitSerialNumber MUST be set to ARRAY The following indicates whether I/O fencing is enabled:
o The /etc/VRTSvcs/conf/config/main.cf file contains a line of "UseFence = SCSI3".
o The /etc/vxfentab file is not empty.
o The /sbin/vxfenadm -g all -f /etc/vxfentab command returns at least one SCSI3 keys.
For Solaris 9 and 10: the unit serial number MUST be set to ARRAY.

12. For Solaris MPxIO or STMS, failovermode settings are as follows:
¿ FLARE 24 or earlier: failovermode MUST be set to 1
¿ FLARE 26 or later: - Solaris 9 with SAN 4.4.7 thru 4.4.11: failovermode MUST be set to 1 - Solaris 9 with SAN 4.4.12 or later: failovermode may be set to 1 for active/passive behavior, or - Solaris 9 with SAN 4.4.12 or later: failovermode may be set to 4 for ALUA behavior - Solaris 10 U2 or earlier: failovermode MUST be set to 1 - Solaris 10 U3 or later: failovermode may be set to 1 for active/passive behavior, or - Solaris 10 U3 or later: failovermode may be set to 4 for ALUA behavior
Note: Use of failovermode = 4 (ALUA) with MPxIO/STMS cannot be supported with clustering environments requiring SCSI2 Reservations such as 2 node Sun Clusters.

glen

11 Posts

April 3rd, 2009 09:00

I checked all that before contacting the forum. However, I have discovered what may be the source of the problem. Despite both HBA ports being successfully logged in and registered on each of the SPs when I run the Verify Storage Environment util it comes back saying the host has only one HBA. I have checked the Connectivity Status and it is OK. I even tried restarting the Management Server on each SP but still no joy. Anybody seen this before?

March 23rd, 2010 08:00

Please modify /etc/kernel/sd.conf  reference cpg

failover mode set=4

11 Posts

August 2nd, 2010 01:00

Turned out host registration had not completed successfully and second HBA port was not really avaiable.

No Events found!

Top