by jjavier69 » Sun Jan 29, 2012 9:50 pm
Has this issue been resolved? I experienced the same issue, and my workaround was more complex:
Everything was the same(everything working great,reboot Starwind device,...), up to the point where I had to re-scan the software HBA more than a couple of times per host - after deleting the LUN - in order for the LUN to be reconnected.
After it saw that the LUN had a pre-configured VMFS datastore, it asked if I wanted to keep the existing signature, which is what I want since I had registered VMs on it. When it tried to re-connect, a cryptic error showed up(sorry, didn't record it), and the only way to get it back without losing the VMs was to create a new signature for the LUN, connect the datastore and re-register the VMs.
What the Starwinds log showed was my LUN file, DDDisk1, failed to create at some point, but could not find where it resurrected itself afterwards. My device info below.
Any assistance is appreciated.
J Javier
*installed on VM running on ESXi 4.1*
AMD Opteron CPU (2vCPU), 4GB RAM
Windows Server 2003, 64-bit
Starwind iSCSI SAN, v5.8 build 20120124
Paravirtual SCSI disk, VMXNET3 NICs, local VMDK disks