Link to home
Start Free TrialLog in
Avatar of Brocklv6
Brocklv6

asked on

ESXi iSCSI attached store

Hello all,
 I recently has an issue with a Windows Server 2008 Storage Server Standard Edition that forced me to do a Windows repair. In that process I lost some configuration. This Windows server is the data store for 2 ESXi (ESXi 4.1.0) servers. The interface between the 3 servers is iSCSI over ethernet.  After I reconfigured the interfaces back to the original configs everything was fine until I restarted on of the ESXi servers. When the ESXi server came back up it could not see the data store any longer. The second ESXi server is still communicating with the NAS and it still sees the file store it needs. I did not reboot the second ESXi server. Here are the details.

1. The interfaces on the ESXi and NAS are configured as follows.
   A. Iface 1
       * NAS, local lan subnet
       * ESXi, local LAN subnet (management network)

   B. Iface 2
      *NAS, iSCSi -1  10.10.10.1
      *ESXi, iSCSI on vSwitch 10.10.10.11

   C. Iface 3
       *NAS, iSCSI -2 10.10.20.1
       *ESXi, iSCSI on vSwitch 10.10.20.11
   

2. The ESXi server can ping the NAS on all 3 interfaces that are configured.
3. The NAS server can ping the ESXi server on all 3 interfaces that are configured.
4. No configuration changes have been made on the ESXi server.
5. The ESXi server can not auto discover the data store on the NAS.

How can I get this communication back between these two servers to bring this data store back online?
Avatar of Andrew Hancock (VMware vExpert PRO / EE Fellow/British Beekeeper)
Andrew Hancock (VMware vExpert PRO / EE Fellow/British Beekeeper)
Flag of United Kingdom of Great Britain and Northern Ireland image

does the ESXi server connect to the iSCSI server?
Avatar of Brocklv6
Brocklv6

ASKER

The ESXi server connects to the NAS over 2 iSCSI over Ethernet interfaces.
SOLUTION
Avatar of Aaron Tomosky
Aaron Tomosky
Flag of United States of America image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
so, a single esxi server is missing a datastore?

when you rescan storage devices, is the lun displayed?
This is what I see for devices.

User generated image
User generated image
Would it be something with the iSCSI Initiator on the Windows NAS?
check the properties of the iSCSI hba that it has the IP address of the iSCSI server, and if you are using chap settings check these are also correct.

rescan storage device

also check access is correctly allocated to this ESXi server iqn, permissions are correct on lun
Another thing, t2here has been no changes made to the ESXi server. This has to be an issue with a config Windows NAS.
check iSCSI nas configs between servers, it would seem this server does not have access
Seems that he Windows NAS is not listening on port 3260. But I cant find the service to start that uses that port.
Ok, here is the latest update. I was able to re-install a iSCSI Target application on the NAS server, so now the ESXi host can see the storage on the NAS over iSCSI. For some reason I can not ad it as a datastore. Should the ESXi server see this storage now that auto discovery is working? Or do I need to actually add this as a datastore? Attached is the error I am getting. User generated image
 Note: The VM data on this NAS drive is two virtual hard drive that contain all of the data for this ESXi host,  12 virtual machines. User generated image
that error above is a Windows sharing error.

so you can see the LUN now, but no datastore is visible?
That is correct. When I start to add the datastore is when the error is thrown. I also do not want to format the data store, I want to still attach the existing VMs.
you will need to connect the Host ESX server using SSH, or run these commands at the console

type the following

esxcfg-volume -l

you should get output similar to the following:-

~ # esxcfg-volume -l
VMFS3 UUID/label: 4cc89503-f054d0d4-2865-00221958e471/nas3-vmfs
Can mount: Yes
Can resignature: No (the volume is being actively used)
Extent name: naa.600140588e9bab7d2b7ed4b53dadc1de:1     range: 0 - 1048319 (MB)

and then type

esxcfg-volume -m 4cc89503-f054d0d4-2865-00221958e471

or

esxcfg-volume -m nas3-vmfs

Refresh the displays on vSphere Client
I attached the wrong error screen, this is the correct one. User generated imageAlso, I changed the target connector to attach the partition instead of the physical drive. This is how it was configured before. Would this change the command you posted?User generated imageUser generated image
SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
No it is not blank. User generated imageI do not have physical access to the ESXi host and I can not SSH into it because is is not enabled. How can I enable SSH with the vSphere client? Is it enable  VMkernel.Boot.techSupportMode?
ASKER CERTIFIED SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Yes it does show that the disk is blank but that is after it it errors. I know the disk is not blank. The 4 files are the only ones that have ever been present in the partition .vhd and .cbm
SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
With the help form the people at experts-exchange, a VM specialist here locally, and some persistence and research on my part the problem was solved. Than you to all that helped.