Login or Register to Ask a Question and Join Our Community

Sponsored Content
Full Discussion: Solaris 10 + mpxio Path
Operating Systems Solaris Solaris 10 + mpxio Path Post 302562595 by general_lee on Friday 7th of October 2011 11:14:46 AM
Old 10-07-2011
Solaris 10 + mpxio Path
Hi,

I have an Oracle/Sun T3-2 Server with a duel SAS controller Fujitsu DX60 array attached.

I have 2 x SAS HBA's in the T3-2. One HBA cabled to CM0 and one HBA cabled to CM1 of the array.

The array is setup with one RAID GROUP and one VOLUME and one LUN

In Solaris I can see one device under format and mpathadm see the paths to o the LUN.

I want to test failover before I continue building this array. But path failover is failing;


Code:
bash-3.2# mpathadm failover lu /dev/rdsk/c6t600000E00D0000000003142900000000d0s2
mpathadm:  Error: The logical unit doesn't support explicit state change.
mpathadm: Unable to complete operation
bash-3.2#

My paths look like this, note I have an "active not optimized" path, all the docs I have read would suggest the Access State should be active or standby. So what does my Access State of "active not optimized" mean?


Code:
bash-3.2# mpathadm show lu /dev/rdsk/c6t600000E00D0000000003142900000000d0s2
Logical Unit:  /dev/rdsk/c6t600000E00D0000000003142900000000d0s2
        mpath-support:  libmpscsi_vhci.so
        Vendor:  FUJITSU 
        Product:  ETERNUS_DXL     
        Revision:  0000
        Name Type:  unknown type
        Name:  600000e00d0000000003142900000000
        Asymmetric:  yes
        Current Load Balance:  round-robin
        Logical Unit Group ID:  NA
        Auto Failback:  on
        Auto Probing:  NA

        Paths:  
                Initiator Port Name:  500605b0025cead0
                Target Port Name:  500000e0d0d42986
                Override Path:  NA
                Path State:  OK
                Disabled:  no

                Initiator Port Name:  500605b0026858d0
                Target Port Name:  500000e0d0d42906
                Override Path:  NA
                Path State:  OK
                Disabled:  no

        Target Port Groups:  
                ID:  17
                Explicit Failover:  no
                Access State:  active not optimized
                Target Ports:
                        Name:  500000e0d0d42986
                        Relative ID:  65

                ID:  16
                Explicit Failover:  no
                Access State:  active optimized
                Target Ports:
                        Name:  500000e0d0d42906
                        Relative ID:  64


My boot messages look like this;

Code:
bash-3.2# dmesg | grep multipath
Oct  7 10:42:09 perwinldom01 genunix: [ID 483743 kern.info] /scsi_vhci/disk@g600000e00d0000000003142900000000 (sd0) multipath status: degraded: path 1 mpt_sas8/disk@w500000e0d0d42986,0 is online
Oct  7 10:42:09 perwinldom01 genunix: [ID 530209 kern.info] /scsi_vhci/disk@g600000e00d0000000003142900000000 (sd0) multipath status: optimal: path 2 mpt_sas9/disk@w500000e0d0d42906,0 is online: Load balancing: round-robin


I have not edited any /kernel/drv files nor had to stmsboot, this is a new install, thus far everything has been dynamic.

If you guys could help me test my path failover that would be great.

Thanks,
 

10 More Discussions You Might Find Interesting

1. UNIX for Advanced & Expert Users

path name for Solaris 8 ?

hello all, instead of type pwd everytime to find down where I am at. is there a command set prompt to put in to your profile on shell so it invisible every you type cd? thanks in advance... (3 Replies)
Discussion started by: phapvn
3 Replies

2. Solaris

Mpxio

Hi All, I have one question. I found out that my system setting for mpxio-disable=yes. Sun recommend to change it to mpxio-disable=no and run stmsboot -e command. But no one can tell will it affect the system or not and the system is sol-10 and using veritas volume manager and attached... (3 Replies)
Discussion started by: mailbox80
3 Replies

3. Solaris

Mpxio

Hi, I have a quick question. How can I see MPXIO statistics (like errors, how many I/O|throuput per path, ....) on Solaris 10? Does maybe a ready dtrace script exist (I do not have the knowledge to write one myself)? I would be happy if somebody knows some good documentation/links (with... (1 Reply)
Discussion started by: victorinox
1 Replies

4. Filesystems, Disks and Memory

VxVM and MPxIO

:confused: Last week I read that VxVM won't work with MPxIO (i don't recall the link) and that it should be unconfigured when installing VxVM. Today I read that VxVM works in "pass-thru" mode with MPxIO and DMP uses the devices presented by MPxIO. If I create disks with MPxIO and use VxVM to... (1 Reply)
Discussion started by: bluescreen
1 Replies

5. Filesystems, Disks and Memory

MPxIO question

Greetings Forumers!! I was given a T2000 with one, dual port, HBA card. The card had one fiber connection to the SAN and the former admin enabled MPxIO. Yesterday, I learned that the other port was connected to the same SAN. This system is in another city so I don't have physical access to... (1 Reply)
Discussion started by: bluescreen
1 Replies

6. Solaris

Solaris 10 - QLogic 4060C iSCSI adapter - multipathed (MPxIO)

System: Sun M3000 and M4000 OS: Solaris 10 (fully up to date with latest patches) iSCSI HBA: 2x QLE4060C (QLogic) In each system are 2 QLE4060C iSCSI adapters. I manage to set up both HBA's to connect to the LUNs on the SAN However I don't manage to enable multi-pathing. Made the... (4 Replies)
Discussion started by: sb008
4 Replies

7. Solaris

MPXIO

Hi Kudo, I wanna to know that how can I do mpxio device name to os native name mapping without using stmsboot -L I tried everything still not able to got the solution till I able to get these o/p here is native disk info as well physical wat is the logical OS name of the disk ... (1 Reply)
Discussion started by: tarunn.dubeyy
1 Replies

8. Solaris

Solaris 10 MPXIO doesn't recover path failure

I have been running some tests with solaris mpxio, dual-pathed to HDS SAN storage. When I take a path offline, Solaris recognised this and a path is shown as offline in luxadm display <device>. However, it does not recover on reinstatement of the path until I run cfgadm -al. Any ideas? Phil (3 Replies)
Discussion started by: UKPhil
3 Replies

9. Solaris

solaris turn off multipath mpxio

Im not able to turn the multipath off on the solaris s10 X86...since im on FC boot.. if i disable multipath , from /kernel/drv/iscsi.conf then im not able to boot the machine.:wall: Is there any other go or any suggestions . (2 Replies)
Discussion started by: gowtham.varma
2 Replies

10. Solaris

Reset MPxIO total path count

We recently moved our netapp disks onto new infrastructure by attaching the new controllers to our fibre switch and netapp migrating the disks. Disks kept the same address. We originally had 4 paths, but after attaching the new controllers we had 8. Now I want to remove the old paths. But can't. ... (5 Replies)
Discussion started by: garydeena
5 Replies

LEARN ABOUT LINUX

multipath

MULTIPATH(8)                                               Linux Administrator's Manual                                               MULTIPATH(8)

NAME

       multipath - Device mapper target autoconfig

SYNOPSIS

       multipath [-v verbosity] [-d] [-h|-l|-ll|-f|-t|-F-B] [-p failover|multibus|group_by_serial|group_by_prio|group_by_node_name] [device]

DESCRIPTION

       multipath is used to detect multiple paths to devices for fail-over or performance reasons and coalesces them

OPTIONS

       -v  level
              verbosity, print all paths and multipaths

                   0           no output

                   1           print the created or updated multipath names only, for use to feed other tools like kpartx

                   2 +         print all info : detected paths, coalesced paths (ie multipaths) and device maps

       -h     print usage text

       -d     dry run, do not create or update devmaps

       -l     show the current multipath topology from information fetched in sysfs and the device mapper

       -ll    show the current multipath topology from all available information (sysfs, the device mapper, path checkers ...)

       -f     flush a multipath device map specified as parameter, if unused

       -F     flush all unused multipath device maps

       -t     print internal hardware table to stdout

       -r     force devmap reload

       -B     treat the bindings file as read only

       -p policy
              force new maps to use the specified policy:

                   failover    1 path per priority group

                   multibus    all paths in 1 priority group

                   group_by_serial
                               1 priority group per serial

                   group_by_prio
                               1  priority group per priority value. Priorities are determined by callout programs specified as a global, per-con-
                               troller or per-multipath option in the configuration file

                   group_by_node_name
                               1 priority group per target node name. Target node names are fetched in /sys/class/fc_transport/target*/node_name.

              Existing maps are not modified.

       device update only the devmap the path pointed by device is in.  device is in the /dev/sdb (as shown by udev in the $DEVNAME  variable)  or
              major:minor format.  device may alternatively be a multipath mapname

SEE ALSO

       multipathd(8), multipath.conf(5), kpartx(8), udev(8), dmsetup(8) hotplug(8)

AUTHORS

       multipath was developed by Christophe Varoqui, <christophe.varoqui@opensvc.com> and others.

                                                                     July 2006                                                        MULTIPATH(8)
All times are GMT -4. The time now is 11:34 AM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy