打印

One STMS issue?

One STMS issue?

Hi Oh, Great ones!

E420R with Solaris 9 installed with SFS 4.4 and 2 single Solaris JNI
PCI-X SG-XPCI1FC-JF2 Model:FCX-6562-L HBA's
!!Please see attatched diagram.!!

This is the way I have it set up and the problem is, I cant get both
paths to work together. The optimal path with the disk on the fastT700
works fine, but the degraded path I have problems reading it.


cfgadm -al
c4                             fc-fabric    connected    unconfigured
unknown
c4::200400a0b80fe6c2           disk         connected    unconfigured
unknown
c5                             fc-fabric    connected    configured   
unknown
c5::200500a0b80fe6c2           disk         connected    unconfigured   
unknown

cfgadm -c configure c5::200500a0b80fe6c2

May  2 23:41:27 zeus scsi: [ID 799468 kern.info] ssd2 at scsi_vhci0:
name g600a0b80000fe6c10000000a4062baa0, bus address
g600a0b80000fe6c10000000a4062baa0
May  2 23:41:27 zeus genunix: [ID 936769 kern.info] ssd2 is
/scsi_vhci/ssd at g600a0b80000fe6c10000000a4062baa0
May  2 23:41:27 zeus mpxio: [ID 669396 kern.info]
/scsi_vhci/ssd at g600a0b80000fe6c10000000a4062baa0 (ssd2) multipath
status: degraded, path /pci at 1f,2000/SUNW,jfca at 1/fp at 0,0 (fp1) to target
address: 200500a0b80fe6c2,0 is online. Load balancing: round-robin
May  2 23:41:27 zeus genunix: [ID 408114 kern.info]
/scsi_vhci/ssd at g600a0b80000fe6c10000000a4062baa0 (ssd2) online

cfgadm -c configure c4::200400a0b80fe6c2

May  3 10:48:50 zeus mpxio: [ID 669396 kern.info]
/scsi_vhci/ssd at g600a0b80000fe6c10000000a4062baa0 (ssd2) multipath
status: optimal, path /pci at 1f,4000/SUNW,jfca at 2/fp at 0,0 (fp0) to target
address: 200400a0b80fe6c2,0 is online. Load balancing: round-robin


format
Searching for disks...
Mode sense page(3) reports nsect value as 64, adjusting it to 64
done

c6t600A0B80000FE6C10000000A4062BAA0d0: configured with capacity of 820.12GB


AVAILABLE DISK SELECTIONS:
       0. c0t0d0 <SUN36G cyl 24620 alt 2 hd 27 sec 107>
          /pci at 1f,4000/scsi at 3/sd at 0,0
       1. c0t1d0 <SUN36G cyl 24620 alt 2 hd 27 sec 107>
          /pci at 1f,4000/scsi at 3/sd at 1,0
       2. c6t600A0B80000FE6C10000000A4062BAA0d0 <IBM-1742-0520 cyl 52488
alt 2 hd 512 sec 64>
          /scsi_vhci/ssd at g600a0b80000fe6c10000000a4062baa0
Specify disk (enter its number): 2
selecting c6t600A0B80000FE6C10000000A4062BAA0d0
[disk formatted]
Disk not labeled.  Label it now? yes
Warning: error writing VTOC.
Illegal request during read
ASC: 0x94   ASCQ: 0x1
Warning: error reading backup label.
Illegal request during read
ASC: 0x94   ASCQ: 0x1
Warning: error reading backup label.
Warning: some backup labels incorrect
Write label failed

But without the other configured:

root at zeus: cfgadm -c unconfigure c4::200400a0b80fe6c2
root at zeus: format
Searching for disks...done


AVAILABLE DISK SELECTIONS:
       0. c0t0d0 <SUN36G cyl 24620 alt 2 hd 27 sec 107>
          /pci at 1f,4000/scsi at 3/sd at 0,0
       1. c0t1d0 <SUN36G cyl 24620 alt 2 hd 27 sec 107>
          /pci at 1f,4000/scsi at 3/sd at 1,0
       2. c6t600A0B80000FE6C10000000A4062BAA0d0 <IBM-1742-0520 cyl 52488
alt 2 hd 512 sec 64>
          /scsi_vhci/ssd at g600a0b80000fe6c10000000a4062baa0
Specify disk (enter its number): 2
selecting c6t600A0B80000FE6C10000000A4062BAA0d0
[disk formatted]

No problems!

But If I turn load balacing off and configure the degrade path first
then the optimal path last, it works also! BUT the failover wont work,
as the drgraded path cant be read.


I can mount-read-write it when it is hasnt got the other degraded path
disk configured or when load balacing is off and both disks are
configured, in a certain order. But off course pulling out the optimal
path fibre cable whilst copying to san, I/O error..............

The disk on the FastT700 is owned by one controller at a time, thus
given the optimal path! But why on eath cant it read it on the non-owned
controller path?.....


root at zeus: more scsi_vhci.conf
#
# Copyright (c) 2001 by Sun Microsystems, Inc.
# All rights reserved.
#
#pragma ident   "@(#)scsi_vhci.conf     1.3     01/08/22 SMI"
#
name="scsi_vhci" class="root";
#
# mpxio global enable/disable switch: setting mpxio-disable="no" will
activate
# I/O multipathing; setting mpxio-disable="yes" disables this feature (do
# not remove this property).
#
mpxio-disable="no";
#
# Load balancing global configuration: setting load-balance="none" will
cause
# all I/O to a given device (which supports multipath I/O) to occur via one
# path.  Setting load-balance="round-robin" will cause each path to the
device
# to be used in turn.
#
load-balance="round-robin";
#
# Force load driver to support hotplug activity (do not remove this
property).
#
ddi-forceattach=1;
# Automatic failback configuration
# possible values are auto-failback="enable" or auto-failback="disable"
#
# This functionality requires patch 113039-02 (or higher).
#
auto-failback="enable";
#
# For enabling MPxIO support for 3rd party symmetric device need an
# entry similar to following in this file. Just replace the "SUN     SENA"
# part with the Vendor ID/Product ID for the device, exactly as reported by
# Inquiry cmd.
#
# This functionality requires patch 113039-02 (or higher).
#
device-type-scsi-options-list =
"IBM     1742            ", "symmetric-option";
symmetric-option = 0x1000000;

TOP

Please double check if FastT700 support auto-failover.   If it doesn't,  you have to
switch it manually, and MPXIO doesn't work as you expected.

IBM DS400 has similar issue.

TOP


感谢一直以来您对我们的支持!
当前时区 GMT+8, 现在时间是 2008-9-7 17:25 京ICP证060528 号

Designed By 17DST