If the server sees the MSA controller device at LUN address 0, then basic Fabric connectivity is there.
You MUST make sure that the operating system profile on the connection is set properly. It used to be "Linux", but I have not checked for the latest firmware revisions. Finally, you must present the logical disks to the new server has well, if selective storage presentation (SSP) is enabled in the Accary Configuration Utility (ACU) or an Access Control List (ACL) has been used in the Command Line Interface (CLI). Make sure you present them through both controllers/connections.
I used "add connection ESX4 wwpn=xxxxxxxx-xxxxxxxx profile=linux offset=0" on the msa and it wont show it online afterwards.
no acl's are used.
From what I can tell, the server will not see any LUNs from the MSA. I will go into VC, select the server, choose configuration, storage, add storage and run the wizard, but it wont show any luns.
When the MSA gets the add connection command it will show a connection but it uses the wwpn as the wwnn. The add connnecton wont take a wwnn entry. Perhaps this is firmware upgrades im looking at. I just have 20 production servers on the LUNs and rebooting it would take an act of congress to get done. We are using HP Blade servers so there are no cables to connect, other than the ones already connecting the chassis to the switches.
There is no SSP enabled. This system is very small and does not require this yet. I will double check to be sure.
I want to just verify something at this point...If the connectivity between the SAN and Blade is correct, is there any configurations in vmware other than making sure the HPA is working and going to add connection on a "default" install. I dont mess with any of the advanced settings, nor do we put any ACL's on the msa. I checked the acls and it says there are none.
Drop the 'offset=0' option. There is no need for that when adding a connection.
Also, do a 'show acl' perhaps something is set and you do not realize it.
Also, verify the WWPN is the correct one between the systems. If you have multiple ports, make sure all are setup properly.
When an ESX V3 server can not see a LUN there are two causes: 1) The number of the LUN or the number of LUNs exceeds 128. 2) THere is something wrong with the LUN presentation.
Edward L. Haletky, author of the forthcoming 'VMWare ESX Server in the Enterprise: Planning and Securing Virtualization Servers', publishing January 2008, (c) 2008 Pearson Education.
show acl says there are no acl's configured.
I added the offset=0 because I was trying to get this to work.
There are 2 LUNs total. 0 and 1.
My Virtual Center is in the same blade chassis and the HP diag util is saying 'loop failed' on the diag of the hba. At this point, Im going through the hp side of the coin since that shows a no connect too.
With that error it sounds like either a HBA, GBIC, Cable, or Switch issue. Check to see if there is light on both sides of the cable. That is a good way of telling if there are issues. My GBICs light up when there is light through the cable. I have one crushed cable and only one of the two possible lights are lit on my switch for the GBIC.
Edward L. Haletky, author of the forthcoming 'VMWare ESX Server in the Enterprise: Planning and Securing Virtualization Servers', publishing January 2008, (c) 2008 Pearson Education
I located the problem a couple of hours ago. Had local admin take photos of rear of blade chassis and switch fabric. I found that the uplinks from the chassis to the FC switches are independent of each other and need a fiber connect for each blade server to the FC switch, this was something I did not realize. I assumed the switch chassis operated like the switch chassis for networking where the ports were uplinks for the whole system. We use the p-Class blade system and only 3 of the 8 uplink ports were connected on each side (redundancy). As soon as we got them linked the ESX saw the LUNs after I did the 'add connection ESX4 wwpn=xxxxxxxx-xxxxxxxx profile=linux' command.