I have the following Enviroment ... 2 x HP DL380G7 , HP 10G NIC with new Firmware ... Lefthand P4300 StarterKit 16TB, 10G Upgrade Kit actually Patches and H3C 10G Switche.
On the Hosts running ESX 4.1 upd1 with last VUM Patches. The iSCSI Configuration on the Host´s are Standard.
1 x vSwitch with 2x 10GB NIC´s, 2 x vmkernel Port Groups -> use iSCSI Software Initiator -> vmkernel Ports bind to vmnic, in the config from vmkernel Ports use the second vmnic as unusable Adapter.
In the Lefthand CMC add die IQN from my Hosts. Create Volumes 500GB and 1,99 TB and assign to the first ESX-Host. I make rescan on the vmhba37 and i see the volumes ... create VMFS Datestores ..change Path Policy to "Round Robin" and all works fine .....
When i assign the volumes to the second ESX-Host´s ... not see the volumes.
Now come the MAGIC Failure -> when i reboot the Second Host , the ost is online back ... the First Host loose the VMFS Volumes.
In the Lefthand CMC see the Connection from the First Host is disconnect and the Second Host is NOW connect ... an i can use the VMFA Volumes on the Second Host. When i reboot the First Host .. the Game beginn new ...
HAVE ANYBODY IDEAS ???
One question: on both hosts you can see the luns under Storage Adapters?
- if yes, you are running into an lun identification problem. each host is seeing the luns with a different identificator. are both hosts on the same storage group?
- if no; maybe you need to activate some sort of "sharing" option for the luns, as they seem not to be shared between more than one host.
marcelo
Hello back,
sometimes when too many work on the Enviroment ... are blind for simple Failure ... the Problem was ..same IP-Adresses on the vmkernel Ports on the ESX-Hosts !