MPXIO not working on a machine with both SAN and Tape drives.

UNIX Solaris
1

Hi,

I have a machine, and mpzio fails everytime i reboot the server. The machine has tape drives and SAN storage, I'm not sure if that is my issue or not. I was told to not enable mpxio globally as that would do something bad to the tape drives. So, I set it to only make the two SAN connected ports failover. It just doesnt work, can anyone help?

My fp.conf and scsi_vhci.conf and error messages:

name="fp" class="fibre-channel" port=0;
name="fp" class="fibre-channel" port=1;
scsi-binding-set="fcp";
load-ulp-list="1","fcp";
ddi-forceattach=1;
mpxio-disable="yes";
name="fp" parent="/pci@400/pci@1/pci@0/pci@8/SUNW,emlxs@0" port=0 mpxio-disable="no";
name="fp" parent="/pci@400/pci@2/pci@0/pci@8/SUNW,emlxs@0" port=0 mpxio-disable="no";

name="scsi_vhci" class="root";
load-balance="none";
auto-failback="enable";
device-type-mpxio-options-list =
"device-type=HITACHI OPEN-V -SUN", "load-balance-options=logical-block-options",
"device-type=HITACHI OPEN-V      -SUN", "load-balance-options=logical-block-options";
logical-block-options="load-balance=logical-block", "region-size=18";

/var/adm/messages.0:Jun 12 09:48:44 servername genunix: [ID 834635 kern.info]
 /scsi_vhci/disk@g5000c50039bddb13 (sd6) multipath status: degraded, path 
 /pci@400/pci@1/pci@0/pci@4/scsi@0/iport@1 (mpt_sas5) to target address: w5000c50039bddb11,0 is online Load balancing: none

/var/adm/messages.0:Jun 12 09:48:52 servername genunix: [ID 834635 kern.info] 
/scsi_vhci/disk@g5000c50039bed82b (sd4) multipath status: degraded, path 
/pci@400/pci@2/pci@0/pci@4/scsi@0/iport@1 (mpt_sas2) to target address: w5000c50039bed829,0 is online Load balancing: none
2

We had problems multipathing probalmes like this. Sun will not help, they expect the SAN vendor to open a ticket with Sun, and will not respond to any userland requests.
So Sun support told us.

So: Open a ticket with HITACHI.

The problem was scsi_vhci.conf settings - for us.

3

scsci_vhci.conf is really sensitive to spaces. Did you get the number of trailing spaces correct?

Whatever parses the file is also VERY particular. Make sure you have it entirely correct.

If that doesn't help, simplify the file. Don't disable mpxio globally and the reenable it for two HBAs.

4

You can try if enable mxpio like that stmsboot -e -D fp This will create pesudo driver for fibre device only. If just using stmsboot -e, will have chance to change the device path of your system disks also, which made the system can not be booted until you fix the paths.

5
  1. What HDS array are you sporting (9985, 9990v, etc.)? Version of microcode installed? what type of library do you have and what version of firmware? What Solaris flavor are you running and what patch level? What SAN switch and OS ?
  2. post the output of the following commands:
fcinfo hba-port 
luxadm -e port
grep -i appear /var/adm/messages.*
  1. check if all the host HBAs are logged in properly in the SAN fabric (all the lights green on the ports, all the SFPs firmly plugged in, etc.)