UCS blades logging into VNX5300

i am not seeing the WWN's logged in.    There is no OS on the Blade we powered up.  I was told we need to manually add the WWNS to the VNX...        hostwwn:wwn  

force it to login

With Palo, once you enable a boot FNIC and reboot the blade, the Option ROM for that FNIC gets enabled and tells Palo FW to log into

the fabric to discover the boot LUNs. If the SAN switches are configured correctly, LUNs mapped, and boot targets configured correctly -

Palo Option ROM reports the boot LUN to the System BIOS and you will see it in the boot menu.

 Access Palo ASIC interface directly (this card has 2 vHBAs configured and is configured to boot from a NetApp FAS)

1. Connect to the adapter chassis/slot/card:

phx2-dc-ucs-A# connect adapter 2/2/1

2. Open rlogin session to Palo interface:

adapter 2/2/1 # connect

3. Attach to the FC port services:

adapter 2/2/1 (top):1# attach-fls

4. Query vNIC interfaces to get numbering:

adapter 2/2/1 (fls):2# vnic

---- ---- ---- ------- -------

vnic ecpu type state lif

---- ---- ---- ------- -------

7 1 fc active 5

8 2 fc active 6

5. Get port login information for each interface - target PWWN and FCID:

adapter 2/2/1 (fls):3# login 7

lifid: 5

ID PORTNAME NODENAME FID

0: 50:0a:09:83:99:fa:e1:e3 00:00:00:00:00:00:00:00 0xe60000

adapter 2/2/1 (fls):4# login 8

lifid: 6

ID PORTNAME NODENAME FID

0: 50:0a:09:83:89:fa:e1:e3 00:00:00:00:00:00:00:00 0xef0001

6. Have the interfaces logged into the target?

adapter 2/2/1 (fls):5# lunmap 7

lunmapid: 0 port_cnt: 1

lif_id: 5

PORTNAME NODENAME LUN PLOGI

50:0a:09:83:99:fa:e1:e3 00:00:00:00:00:00:00:00 0000000000000000 Y

adapter 2/2/1 (fls):6# lunmap 8

lunmapid: 0 port_cnt: 1

lif_id: 6

PORTNAME NODENAME LUN PLOGI

50:0a:09:83:89:fa:e1:e3 00:00:00:00:00:00:00:00 0000000000000000 Y

Palo does not have a traditional "BIOS utility", you use UCSM CLI commands and the blade navigator "Boot Order Details" dialogue to

monitor, and troubleshoot SAN booting issues. The traditional utilities you are used to using are not required.

 MDS FCNS Database - verify target FCID and initiator login:

PHX2-DC-MDS9509-1# sh fcns database vsan 42

VSAN 42:

--------------------------------------------------------------------------

FCID TYPE PWWN (VENDOR) FC4-TYPE:FEATURE

--------------------------------------------------------------------------

0xef0000 N 20:47:00:0d:ec:e2:06:40 (Cisco) npv

0xef0001 N 50:0a:09:83:89:fa:e1:e3 (NetApp) scsi-fcp:target

0xef0008 N 20:00:04:00:0a:00:00:07 scsi-fcp:init fc-gs

0xef000a N 20:00:04:00:0a:00:00:06

0xef0010 N 20:00:04:00:0b:00:00:07 scsi-fcp:init

0xef0011 N 20:00:04:00:0b:00:00:06 scsi-fcp:init

0xef0022 N 20:45:00:0d:ec:e2:06:40 (Cisco) npv

Total number of entries = 7

Using the native Palo and UCSM tools to troubleshoot and investigate SAN booting situations.

Storage Page 1

PHX2-DC-MDS9509-2# sh fcns database vsan 43

VSAN 43:

--------------------------------------------------------------------------

FCID TYPE PWWN (VENDOR) FC4-TYPE:FEATURE

--------------------------------------------------------------------------

0xe60000 N 50:0a:09:83:99:fa:e1:e3 (NetApp) scsi-fcp:target

0xe60001 N 20:47:00:0d:ec:e2:85:80 (Cisco) npv

0xe60002 N 20:45:00:0d:ec:e2:85:80 (Cisco) npv

0xe60010 N 20:00:08:00:0a:01:00:06

0xe60011 N 20:00:08:00:0a:01:00:07 scsi-fcp:init fc-gs

0xe60017 N 20:00:08:00:0b:01:00:07 scsi-fcp:init

0xe60018 N 20:00:08:00:0b:01:00:06 scsi-fcp:init

Total number of entries = 7

 Use Boot Order Details to verify the current BIOS boot order configuration

"Boot Order Details" is a real time view into the current status of the BIOS boot order configuration, rather than have to stop the

boot and enter a separate interface one can view the BIOS live. If you observe it as a blade cycles you will see it transition several

times. For example, during the PNUOS boot phase the "Actual Boot Order" will be a PXE boot from an internal XGA interface.

After the OS boots and the host drivers are loaded, Palo FW  logs out of the fabric and the host drivers take over.

If you run the "lunmap" command at this point, Palo FW does not have any LUNs to report, since it has logged out of the fabric. All

FC information will need to be derived from the host drivers

adapter 2/2/1 (fls):100# lunmap 7

lunmapid: 0 port_cnt: 1

lif_id: 5

PORTNAME NODENAME LUN PLOGI

Storage Page 2

50:0a:09:83:99:fa:e1:e3 00:00:00:00:00:00:00:00 0000000000000000 N

adapter 2/2/1 (fls):101# lunmap 8

lunmapid: 0 port_cnt: 1

lif_id: 6

PORTNAME NODENAME LUN PLOGI

50:0a:09:83:89:fa:e1:e3 00:00:00:00:00:00:00:00 0000000000000000 N

adapter 2/2/1 (fls):101# login 7

lifid: 5

ID PORTNAME NODENAME FID

0: 50:0a:09:83:99:fa:e1:e3 00:00:00:00:00:00:00:00 0x000000

adapter 2/2/1 (fls):101# login 8

lifid: 6

ID PORTNAME NODENAME FID

0: 50:0a:09:83:89:fa:e1:e3 00:00:00:00:00:00:00:00 0x000000

Were initiators logged in to the switch when you zonned them?

Sent from my Verizon Wireless Droid

OS doesn't need to be installed for you to see the initiators on the VNX. Reboot the server one more time and see if the initiators log in. I assume this is a boot from SAN server, correct?

Yes they were

Didnt login and i have verified zoning is configured.

NLAALFABA01-A# connect adapter 2/2/1

Error: Adapter 2/2/1 not found

NLAALFABA01-A# connect adapter 1/4/1

adapter 1/4/1 # connect

adapter 1/4/1 (top):1# attach-fls

adapter 1/4/1 (fls):1# vnic

---- ---- ---- ------- -------

vnic ecpu type state   lif

---- ---- ---- ------- -------

9    1    fc   active  6

10   2    fc   active  7

adapter 1/4/1 (fls):2# login 9

lifid: 6

  ID   PORTNAME                 NODENAME                 FID

    1: 50:06:01:6a:47:20:48:7c  00:00:00:00:00:00:00:00  0x000000

    0: 50:06:01:62:47:20:48:7c  00:00:00:00:00:00:00:00  0x000000

adapter 1/4/1 (fls):3# lunmap 9

lunmapid: 0  port_cnt: 1

  lif_id: 6

    PORTNAME                 NODENAME                 LUN               PLOGI

    50:06:01:62:47:20:48:7c  00:00:00:00:00:00:00:00  0000000000000000  N

lunmapid: 1  port_cnt: 1

  lif_id: 6

    PORTNAME                 NODENAME                 LUN               PLOGI

    50:06:01:6a:47:20:48:7c  00:00:00:00:00:00:00:00  0000000000000000  N

adapter 1/4/1 (fls):4# login 10

lifid: 7

  ID   PORTNAME                 NODENAME                 FID

    1: 50:06:01:6b:47:20:48:7c  00:00:00:00:00:00:00:00  0x000000

   0: 50:06:01:63:47:20:48:7c  00:00:00:00:00:00:00:00  0x000000

adapter 1/4/1 (fls):5# lunmap 10

lunmapid: 0  port_cnt: 1

  lif_id: 7

    PORTNAME                 NODENAME                 LUN               PLOGI

    50:06:01:63:47:20:48:7c  00:00:00:00:00:00:00:00  0000000000000000  N

lunmapid: 1  port_cnt: 1

  lif_id: 7

    PORTNAME                 NODENAME                 LUN               PLOGI

    50:06:01:6b:47:20:48:7c  00:00:00:00:00:00:00:00  0000000000000000  N

no sure, I configured a couple of boot from SAN boxes a few weeks ago, WWNs showed up in Unisphere as soon as i zoned them. Might have to create WWN initiators manually.

Post screen shots... i have UCS blades B200M2 and B230M2 going thru nexus 5548s L3 switches licenses with 8 ports FibreChannel (8 ports of it poses as my MDX)... and VNX5300 as well... booting from SAN as well... all 8 host show up , each blade profile has 2 vHBAs and i am seeing all the paths from each one... i did my zoning on the 5548s as you problably guessed...

haha ..how did that happen ?

All,.

This problem has been resolved.   All of your answer were very helpful.  They didnt work because someone plug the cellerra ports into the MDS instead of the HOST PORTS on the SP.