09-08-2010 10:12 AM
Hi,
This is my first setup of an MDS zone and I can't figure out what I'm doing wrong here. Here is my config that matters:
device-alias database
device-alias name TESTST01-B pwwn 20:15:00:a0:b8:11:39:de
device-alias name TESTSVR01-A pwwn 21:00:00:1b:32:9b:35:8f
device-alias commit
fcdomain fcid database
vsan 1 wwn 20:15:00:a0:b8:11:39:de fcid 0x5b0000 dynamic
! [TESTST01-B]
vsan 1 wwn 21:00:00:1b:32:9b:35:8f fcid 0x5b0100 dynamic
! [TESTSVR01-A]
zone name TESTSVR01-001 vsan 1
member pwwn 20:15:00:a0:b8:11:39:de
! [TESTST01-B]
member pwwn 21:00:00:1b:32:9b:35:8f
! [TESTSVR01-A]
zoneset name FABRIC-B vsan 1
member TESTSVR01-001
zoneset activate name FABRIC-B vsan 1
The PWWN's listed in the device-alias database are the ones that connect to each other when I connect the server HBA directly to the storage device. Here are also some show commands:
-----
sh flogi database
--------------------------------------------------------------------------------
INTERFACE VSAN FCID PORT NAME NODE NAME
--------------------------------------------------------------------------------
fc1/2 1 0x5b0000 20:15:00:a0:b8:11:39:de 20:04:00:a0:b8:11:39:de
[TESTST01-B]
fc1/3 1 0x5b0100 21:00:00:1b:32:9b:35:8f 20:00:00:1b:32:9b:35:8f
[TESTSVR01-A]
-----
sh vsan
vsan 1 information
name:VSAN0001 state:active
interoperability mode:default
loadbalancing:src-id/dst-id/oxid
operational state:up
vsan 4079:evfp_isolated_vsan
vsan 4094:isolated_vsan
-----
sh zoneset active
zoneset name FABRIC-B vsan 1
zone name TESTSVR01-001 vsan 1
* fcid 0x5b0000 [pwwn 20:15:00:a0:b8:11:39:de] [TESTST01-B]
* fcid 0x5b0100 [pwwn 21:00:00:1b:32:9b:35:8f] [TESTSVR01-A]
-----
show zone status
VSAN: 1 default-zone: deny distribute: full Interop: default
mode: basic merge-control: allow
session: none
hard-zoning: enabled broadcast: unsupported
Default zone:
qos: none broadcast: unsupported ronly: disabled
Full Zoning Database :
DB size: 128 bytes
Zonesets:1 Zones:1 Aliases: 0
Active Zoning Database :
DB size: 68 bytes
Name: FABRIC-B Zonesets:1 Zones:1
Status: Activation completed at 18:04:21 PDT Sep 8 2010
-----
sh zoneset active
zoneset name FABRIC-B vsan 1
zone name TESTSVR01-001 vsan 1
* fcid 0x5b0000 [pwwn 20:15:00:a0:b8:11:39:de] [TESTST01-B]
* fcid 0x5b0100 [pwwn 21:00:00:1b:32:9b:35:8f] [TESTSVR01-A]
09-08-2010 06:50 PM
Peter,
Can you tell us what symptoms you see or what makes you think there is anything wrong in the first place?
You have successfully created a zone and activated a zoneset. The MDS indicates that both members of the zone are online and can see eachother. I don't see any issues why these two devices would be having a problem.
09-09-2010 12:44 AM
Thanks for clearing out that the config should work. When I connect server and storage directly I see the PWWN of the storage pop up in the qlogic bios and I can connect to the assigned lun. When I use the MDS with the config listed above I don't see the storage PWWN or the lun. Something tells me that I should see the same behaviour when I have the switch in the middle as when I connect the server and storage directly.
09-09-2010 03:29 AM
Peter,
on some vendor arrays you have to set correct topology on the ports that connect into your fabric. For example on HP MSA i can set "point to point" which is only design for direct "server to storage" connection with no fiber switch involved. You also have arbitrated loop (used only when connected to a hub), another topology is "switched", this is where both initiator (server) and target (array) connect to a fiber switch and you zone them together. Not sure if that's your sure but something to verify.
09-09-2010 07:44 AM
I've checked the settings on the HBA (it's a qlogic) and there is no such thing as a "Server to storage" setting.
I figured something else out:
FCSW# fcping pwwn 20:15:00:a0:b8:11:39:de vs 1
28 bytes from 20:15:00:a0:b8:11:39:de time = 2709 usec
28 bytes from 20:15:00:a0:b8:11:39:de time = 759 usec
28 bytes from 20:15:00:a0:b8:11:39:de time = 757 usec
28 bytes from 20:15:00:a0:b8:11:39:de time = 755 usec
28 bytes from 20:15:00:a0:b8:11:39:de time = 757 usec
5 frames sent, 5 frames received, 0 timeouts
Round-trip min/avg/max = 755/1147/2709 usec
FCSW# fcping pwwn 21:00:00:1b:32:9b:35:8f vs 1
28 bytes from 21:00:00:1b:32:9b:35:8f time = 771 usec
28 bytes from 21:00:00:1b:32:9b:35:8f time = 764 usec
28 bytes from 21:00:00:1b:32:9b:35:8f time = 763 usec
28 bytes from 21:00:00:1b:32:9b:35:8f time = 762 usec
28 bytes from 21:00:00:1b:32:9b:35:8f time = 765 usec
09-09-2010 09:27 AM
You should be able to set an option in the HBA for Point to Point or Loop Preferred. Check the Wizards tab, general configuration wizard in san surfer
09-10-2010 04:59 AM
I found the option "Loop only/Point to point/Loop only preferred". Tried all of them but still no connection. Next up I will try to get access to the storage device to see how that's configured. Any other ideas? I'm guessing it's something simple that I'm not seeing...
09-10-2010 06:50 AM
So your saying in SANsurfer you don't see the storage pwwn? In the window on the left you can click on each hba port and expand what it sees.
09-10-2010 07:12 AM
I can't use SANsurfer since I'm booting from SAN. I have no OS to install it on. All I have is the BIOS which is telling me that there is no storage device attached.
09-13-2010 11:43 PM
In the SAN boot section in the HBA Bios (not 100% sure where it is atm and i can't check) can you see any target ports at all?
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide