Equallogic iSCSI connection issue

Question

I rebooted a Windows 2012 server that had an iSCSI connection to an Equallogic 6100 SAN. Following the reboot I checked File Explorer and the X: drive was not present. I launched the iSCSI initiator app and the target displays as connected. I disconnect and connected. I checked Disk Manager - ran a re-scan - drive does not appear. I checked the SAN monitoring log and have the following message:

Member MOC-MEMBER02
Subsystem MgmtExec
Event ID 7.4.3 7.4.24
Message iSCSI login to target '172.100.0.10:3260, iqn.2001-05.com.equallogic:0-1cb196-a04af794f-42c004f7b2856e9c-xx-xxxxx-xxxxxx' from initiator '10.100.1.4:49152, iqn.1991-05.com.xxxx.xxxxxx.xxxxxx.xxxxx' failed for the following reason: Requested target not found.

I do see the volume but it is named "production" Tvideo" I assume it is contained within that volume. Is there a way to look into the volume to see its contents?

What steps do I take to re-mediate the lost SAN drive connection?

Thanks,

Milty

Milty60 · Accepted Answer

I have created a new volume on the SAN and configured access control for one initiator IP address only. I removed all settings in the initiator. I launched the iSCSI initiator on the server and entered the IP for the SAN. I copied in the IQN of the server.The firewall on the server is off for testing. I do not have any targets listed on the Target tab. I have rebooted the server.

dwilliam62 · Accepted Answer

Hello,

I don't understand why you copied the IQN name from the EQL GUI. There is no need to do that.

When you add in the iSCSI Group IP address, the iSCSI "Discovery" process which is what failed here will create a list of available iSCSI targets to the initiator. At that point from that populated list you can login and connect that server to that EQL volume.

Here's some Equallogic training videos. You might find some of them helpful.

https://www.youtube.com/playlist?list=PLsHMCOyTtKkXvuklj1I0bdYtPC17jqOjl

This is an EQL 101 on connecting a Windows server to EQL storage

https://www.youtube.com/watch?v=BtogepZwde0

Regards,

Don

dwilliam62 · Answer

Hello Milty,

If in the ISCSI initiator app you see the volume connected, I believe that error message is for a volume that is no longer there. In the EQL GUI if you go to that volume, then look at the connections tab I suspect you will see that it is connected.

If that is correct, then you don't have an iSCSI authentication issue.

For that volume how is access restricted? It should be set to only allow that server to access that volume. If someone used something like an IP wildcard "*" then any server on that network can access the volume not just the correct server. This results in corruption over time. When you reboot then is when you tend to discover the problem as the OS re-reads the partition table and filesystem info. it's possible that filesystem recovery tools or a third party company might be able to recover the information.

If the access control is a wildcard then that needs to be changed immediately. Using IQN name from the server is a very common approach.

Another common cause is a switch that either isn't proper configured for Jumbo Frames or does not support Jumbo frames. Setting the server NICs to 1500 MTU might resolve this problem.

Regards,

Don

Milty60 · Answer

I checked the switch - Jumbo frames are enabled and the NICs are set for MTU 1500   Thanks,    Milty

Milty60 · Answer

In the EQL GUI if I go to the volume, when I check the connections tab I don't see the IP address of the initiator connected although it shows connected on the server initiator app.  Thanks,  Milty

dwilliam62 · Answer

Hello,

That sounds rather strange. Are you sure it's not connecting to the "VSS-CONTROL" volume? That's used with Microsoft Volume Shadow Services.

There is a connection tab under the group settings which will show you all connections to this group.

I would also look at the access control for the volume you are trying to connect to. If you renamed the host or joined an AD domain, MS server changes the IQN name automatically. So that volume access list might not longer match the server.

Regards,

Don

Milty60 · Answer

I checked the connection tab under the group settings and I don't see a "VSS-CONTROL" volume.

I checked the access control for the volume I am trying to connect to - its wide open *.*. We have not renamed the host or joined an AD domain.

Is their a way to look into the SAN volumes via a Putty connection to locate the volume I do not see in the GUI?

Thanks,

Milty

dwilliam62 · Answer

Hello Mity,

The GUI and CLI get the info from same place so they should show the samething.

To show connections the syntax is:

GrpName>volume select VOLUMENAME show connections

e.g. volume select datavolume1 show connections

Initiator UpTime RxData TxData
--------------- ----------------- --------------- ---------------
100.85.238.73 24days 15hr 4min 8.50KB 7.44MB 33secs
100.85.238.72 24days 15hr 4min 8.00KB 6.08MB 30secs
100.85.237.215 24days 15hr 4min 16.50KB 12.96MB 29secs
100.85.238.13 24days 15hr 4min 6.57MB 5.02GB 29secs
100.85.239.20 24days 15hr 4min 0.00KB 11.29MB 27secs
100.85.238.25 24days 15hr 4min 6.52MB 5.01GB 25secs
100.85.232.182 24days 15hr 4min 17.00KB 12.47MB 22secs
100.85.239.75 24days 15hr 4min 2.47MB 350.17MB 20secs

To see all the volumes in that group use:

GrpName>show volumes

Re: Access Control You need to fix that immediately. If that volume isn't part of a cluster corruption WILL occur if more than one server connects to it.

Regards,

Don

EqualLogic

Equallogic iSCSI connection issue

Was this post helpful?