topic Re: fibre channel differences in Operating System - HP-UX

fibre channel differences

Louber — Tue, 06 Sep 2005 14:41:10 GMT

I have a couple questions with regards to the setup on our Fiber Channels. A few differences have been discovered and I am unsure as to whether they are relevant or not. The differences have been picked up via the esmg site, and I have been parcelled with some work to follow up on ensuring the clusters are exact. Here goes....

Below shows that 2 fibre channels have different XL2 Chip Revisions. Does this pose an issue.?

The other 2 are actually TL Chip Revision No is = 2.3 and are identical. However the Local N_Port_id is differs. Is this how they should be setup.?

Reference System MBNAHP92

Class I H/W Path Driver S/W State H/W Type Description
=================================================================
fc 2 0/1/0/0 td CLAIMED INTERFACE HP Tachyon XL2 Fibre Channel Mass Storage Adapter
/dev/td2
fc 3 0/1/1/0 td CLAIMED INTERFACE HP Tachyon XL2 Fibre Channel Mass Storage Adapter
/dev/td3
fc 0 0/3/0/0 td CLAIMED INTERFACE HP Tachyon TL/TS Fibre Channel Mass Storage Adapter
/dev/td0
fc 1 0/6/0/0 td CLAIMED INTERFACE HP Tachyon TL/TS Fibre Channel Mass Storage Adapter
/dev/td1
root@mbnahp92:/ # fcmsutil /dev/td2

Vendor ID is = 0x00103c
Device ID is = 0x001029
XL2 Chip Revision No is = 2.2
PCI Sub-system Vendor ID is = 0x00103c
PCI Sub-system ID is = 0x00128c
Previous Topology = UNINITIALIZED
Link Speed = UNINITIALIZED
Local N_Port_id is = 0x000000
Local Loop_id is = 126
N_Port Node World Wide Name = 0x50060b0000112c8d
N_Port Port World Wide Name = 0x50060b0000112c8c
Driver state = AWAITING_LINK_UP
Hardware Path is = 0/1/0/0
Number of Assisted IOs = 0
Number of Active Login Sessions = 0
Dino Present on Card = NO
Maximum Frame Size = 960
Driver Version = @(#) PATCH_11.00: libtd.a : Jul 15 2002, 11:34:12, PHSS_26798

root@mbnahp92:/ # fcmsutil /dev/td1

Vendor ID is = 0x00103c
Device ID is = 0x001028
TL Chip Revision No is = 2.3
PCI Sub-system Vendor ID is = 0x00103c
PCI Sub-system ID is = 0x000006
Topology = PRIVATE_LOOP
Local N_Port_id is = 0x000002

Target System MBNAHP93

Class I H/W Path Driver S/W State H/W Type Description
=================================================================
fc 2 0/1/0/0 td CLAIMED INTERFACE HP Tachyon XL2 Fibre Channel Mass Storage Adapter
/dev/td2
fc 3 0/1/1/0 td CLAIMED INTERFACE HP Tachyon XL2 Fibre Channel Mass Storage Adapter
/dev/td3
fc 0 0/3/0/0 td CLAIMED INTERFACE HP Tachyon TL/TS Fibre Channel Mass Storage Adapter
/dev/td0
fc 1 0/6/0/0 td CLAIMED INTERFACE HP Tachyon TL/TS Fibre Channel Mass Storage Adapter
/dev/td1
root@mbnahp93:/usr/conf/master.d # fcmsutil /dev/td2

Vendor ID is = 0x00103c
Device ID is = 0x001029
XL2 Chip Revision No is = 2.3
PCI Sub-system Vendor ID is = 0x00103c
PCI Sub-system ID is = 0x00128c
Previous Topology = UNINITIALIZED
Link Speed = UNINITIALIZED
Local N_Port_id is = 0x000000
Local Loop_id is = 126
N_Port Node World Wide Name = 0x50060b000021e229
N_Port Port World Wide Name = 0x50060b000021e228
Driver state = AWAITING_LINK_UP
Hardware Path is = 0/1/0/0
Number of Assisted IOs = 0
Number of Active Login Sessions = 0
Dino Present on Card = NO
Maximum Frame Size = 960
Driver Version = @(#) PATCH_11.00: libtd.a : Jul 15 2002, 11:34:12, PHSS_26798

root@mbnahp93:/usr/conf/master.d # fcmsutil /dev/td0

Vendor ID is = 0x00103c
Device ID is = 0x001028
TL Chip Revision No is = 2.3
PCI Sub-system Vendor ID is = 0x00103c
PCI Sub-system ID is = 0x000006
Topology = PRIVATE_LOOP
Local N_Port_id is = 0x000004

Re: fibre channel differences

Ranjith_5 — Tue, 06 Sep 2005 14:47:16 GMT

Hi Louber,

there shouldn't be any problem if at all the version is different. On My machine all the versions are 2.3. Can you compare the outputs of

#fcmsutil /dev/td1 vpd

with

#fcmsutil /dev/td2 vpd

will get some more idea about the cards.

Regards,
Syam

Re: fibre channel differences

Colin Topliss — Tue, 06 Sep 2005 16:39:08 GMT

You shouldn't have a problem with this. We've run systems with different firmware revs.

The only issue we ever came across was with using OLAR, when the firmware revs between the old card and the new card were different. As far as I am aware, OLAR is not available for 11.00 (and is supported on certain types of hardware only).

FYI, OLAR = Online Addition And Replacement.

If you want to be 100% sure, you could ask your HP rep if the older rev had any problems that were fixed with the later rev which may cause you a problem.

Col

Re: fibre channel differences

marie-noelle jeanson_1 — Wed, 07 Sep 2005 09:04:20 GMT

Louber,

You can work with either version of the chip.

In XL2 chip rev 2.2, there was a problem with the support for a frame payload of 2048 bytes. The FC HBAs with the Tachyon XL2 Rev 2.2 controllers would not always work correctly with 2K frame payloads.

A workaround has been implemented in the FC driver with chip rev 2.2, therefore it is OK to work with this chip. Chip rev 2.3 fixed later the issue but the FC driver still requires a patch PHSS_26798 or later (11.00) and PHSS_26799 or later (11.11) to enable support of 2K frame payload with the XL2 Rev 2.3 controller when the target device also supports 2K frame payload sizes
(see patch information for SR:8606256538 or CR:JAGae20853.

Regards,

Marie.

Re: fibre channel differences

Chris Loggans — Thu, 01 Dec 2005 09:57:29 GMT

Marie,

What, if any, are the performance impacts of using a 1k vs 2k frame size?

Thanks,
-Chris

Re: fibre channel differences

marie-noelle jeanson_1 — Tue, 06 Dec 2005 12:32:58 GMT

Louber,

there could be an improved payload efficiency of ~3% at the maximum payload size (2K). This is not very significant.

Marie.