First I'd like to say thank you for providing a free high availability option for vSphere users with no size restrictions
![Smile :)](./images/smilies/icon_smile.gif)
Now onto my questions... I'm following your PDF on configuring my two host ESXi cluster (http://www.starwindsoftware.com/starwin ... re-vsphere). I'm having difficulties configuring the loopback interface so I can mount LUNS on the host that is also hosting the SAN VM. I was able to successfully create the vSwitch as per Step 3 on page 14 as well as the vmkernel to enable iSCSI on that vSwitch. I added a network interface to the Starwind VM which is attached to the same vSwitch. I was able to ping the SAN target but vSphere would never see any of the devices I created on the SAN VM. The only way I was able to mount them was to attach the Starwind VM network adapter and the ISCSI vmkernel to a vSwitch that had a vmnic attached to it. Once I did that I was able to see the LUNs and created datastores, etc. Following is an excerpt of the Starwind logs that I saw every time I would kick off a rescan.
10/26 8:45:48.624 474 C[11], XPT_UP: Event - LOGIN.
10/26 8:45:48.624 474 C[11], IN_LOGIN: T4.
10/26 8:45:48.624 474 Params: <<< String param 'InitiatorName': received 'iqn.2012-06.com.cedarwoodtechsolutions:esx2', accepted 'iqn.2012-06.com.cedarwoodtechsolutions:esx2'
10/26 8:45:48.624 474 Params: <<< Enum param 'SessionType': received 'Discovery', accepted 'Discovery'
10/26 8:45:48.624 474 Params: <<< Enum param 'HeaderDigest': received 'None', accepted 'None'
10/26 8:45:48.624 474 Params: <<< Enum param 'DataDigest': received 'None', accepted 'None'
10/26 8:45:48.624 474 Params: <<< Numeric param 'DefaultTime2Wait': received 0, accepted 0
10/26 8:45:48.624 474 Params: <<< Numeric param 'DefaultTime2Retain': received 0, accepted 0
10/26 8:45:48.624 474 Params: <<< Boolean param 'IFMarker': received No, accepted 0
10/26 8:45:48.624 474 Params: <<< Boolean param 'OFMarker': received No, accepted 0
10/26 8:45:48.624 474 Params: <<< Numeric param 'ErrorRecoveryLevel': received 0, accepted 0
10/26 8:45:48.624 474 Params: <<< Numeric param 'MaxRecvDataSegmentLength': received 32768, accepted 32768
10/26 8:45:48.624 474 Params: >>> ErrorRecoveryLevel=0.
10/26 8:45:48.624 474 Params: >>> HeaderDigest=None.
10/26 8:45:48.624 474 Params: >>> DataDigest=None.
10/26 8:45:48.624 474 Params: >>> OFMarker=No.
10/26 8:45:48.624 474 Params: >>> IFMarker=No.
10/26 8:45:48.624 474 Params: >>> InitialR2T=No.
10/26 8:45:48.624 474 Params: >>> ImmediateData=Yes.
10/26 8:45:48.624 474 Params: >>> MaxRecvDataSegmentLength=262144.
10/26 8:45:48.624 474 Params: >>> MaxBurstLength=262144.
10/26 8:45:48.624 474 Params: >>> FirstBurstLength=262144.
10/26 8:45:48.624 474 Params: >>> DefaultTime2Wait=0.
10/26 8:45:48.624 474 Params: >>> DefaultTime2Retain=0.
10/26 8:45:48.624 474 Params: >>> MaxOutstandingR2T=1.
10/26 8:45:48.624 474 Params: >>> DataPDUInOrder=Yes.
10/26 8:45:48.624 474 Params: >>> DataSequenceInOrder=Yes.
10/26 8:45:48.624 12c T[11,1]: session 0x11, connection 0x11 : end of stage 1, next stage 3.
10/26 8:45:48.624 12c C[11], IN_LOGIN: Event - LOGIN_ACCEPT.
10/26 8:45:48.624 12c C[11], LIN: T5.
10/26 8:45:48.624 474 error: assertion failed: c:\_starwind\_release_8_0_r2\starwind\service\server\task.cpp:3462 (m_readData.size() == 0)
10/26 8:45:48.624 474 error: assertion failed: c:\_starwind\_release_8_0_r2\starwind\service\server\task.cpp:3470 (m_readData.size() == 0)
10/26 8:45:48.624 474 Srv: <no target>: 'SendTargets=All' for a Discovery session...
10/26 8:45:48.624 12c C[11], LIN: Event - LOGOUT.
10/26 8:45:48.624 12c C[11], IN_LOGOUT: T9/10.
10/26 8:45:48.624 404 C[11], IN_LOGOUT: Event - LOGOUT_ACCEPT.
10/26 8:45:48.624 404 C[11], FREE: T13.
10/26 8:45:48.624 474 C[11], FREE: recvData returned 10054
Onto the second issue: I created datastores and cloned two VMs to the new datastore. That went fine. I then started to copy files to the new VM. After about half way through, the file copy froze along with the VM. I was never able to recover it. Eventually my VM got corrupted. It is no longer visible in the datastore browser. The name of the VM became "/vmfs/volumes/544cf7d2-cfe144ad-9398-00151763a4bc/Files/Files.vmx" instead of simply, Files. Following is an excerpt of the Starwind logs when this corruption happened. These logs appear quite a lot in the file at the time the file copy was happening.
10/26 12:36:22.895 4c C[4], LIN: T5.
10/26 12:36:59.113 53c T[4,8]: Management command: abort task (CmdSN 72025, ITT 0x1a1a0100) not found.
10/26 12:37:00.004 53c T[4,b]: Management command: abort task (CmdSN 72026, ITT 0x1b1a0100) not found.
10/26 12:37:00.004 a74 IMG: *** ImageFile_IoCompleted: Error (1117) returned to IO completion!
10/26 12:37:00.004 a74 IMG: *** ImageFile_ReadWriteSectorsCompleted: Error occured (ScsiStatus = 2, DataTransferLength = 0)!
10/26 12:37:00.004 a7c IMG: *** ImageFile_IoCompleted: Error (1117) returned to IO completion!
10/26 12:37:00.004 a7c IMG: *** ImageFile_ReadWriteSectorsCompleted: Error occured (ScsiStatus = 2, DataTransferLength = 0)!
10/26 12:37:00.004 a7c IMG: *** ImageFile_IoCompleted: Error (1117) returned to IO completion!
10/26 12:37:00.004 a7c IMG: *** ImageFile_ReadWriteSectorsCompleted: Error occured (ScsiStatus = 2, DataTransferLength = 0)!
10/26 12:37:00.004 a74 IMG: *** ImageFile_IoCompleted: Error (1117) returned to IO completion!
10/26 12:37:00.004 a74 IMG: *** ImageFile_ReadWriteSectorsCompleted: Error occured (ScsiStatus = 2, DataTransferLength = 0)!
10/26 12:37:01.566 a7c IMG: *** ImageFile_ReadWriteSectorsCompleted: Error occured (ScsiStatus = 2, DataTransferLength = 0)!
10/26 12:37:01.566 a7c IMG: *** ImageFile_ReadWriteWithCacheCompleted: Error (0xC0000001) returned to cache request completion!
10/26 12:37:01.566 a7c IMG: *** ImageFile_IoCompleted: Error (1117) returned to IO completion!
10/26 13:07:13.301 a74 IMG: *** ImageFile_IoCompleted: Error (1117) returned to IO completion!
10/26 13:07:13.301 a74 IMG: *** ImageFile_ReadWriteSectorsCompleted: Error occured (ScsiStatus = 2, DataTransferLength = 0)!
10/26 13:07:13.301 a74 IMG: *** ImageFile_ReadWriteWithCacheCompleted: Error (0xC0000001) returned to cache request completion!
Some additional notes to keep in mind. I have not configured high availability as I need to migrate my VMs off of disks so I can repurpose them and add them to my other host. So for right now, this is just a hyperconverged single node iSCSI VM. I have write back cache configured on the LUN as well since I wanted to be prepared for when I bring the second node online.
I am running the latest version of Starwind (8.0.7145).
I would really appreciate any help you can provide.
Thanks
![Smile :)](./images/smilies/icon_smile.gif)