VMware Cloud Community
Jesse1973
Enthusiast
Enthusiast

storage issue NetApp/iscsi/ESX3.5

I get a timeout on iSCSI on different ESXs on different times. It is puzzling, as both the ESX hosts as windows and other guests are experiencing timeouts.

all networks are divided, NFS runs over a different network than iSCSI, which is on a dedicated lan.

an excerpt of vmkiscsid.log of last night:

2010-02-10-12:30:38: iscsid: an InitiatorAlias= is required, but was not found in /etc/vmware/vmkiscsid/initiatorname.iscsi

2010-02-10-12:30:38: iscsid: LogLevel = 0

2010-02-10-12:30:38: iscsid: LogSync = 0

2010-02-10-12:30:42: iscsid: Login Success: iqn.1992-08.com.netapp:sn.101197719,default,192.168.73.2,3260,2001, 0x1

2010-02-10-12:30:42: iscsid: connection1:0 is operational now

2010-02-16-02:03:35: iscsid: Kernel reported iSCSI connection 1:0 error (1008) state (3)

2010-02-16-02:03:39: iscsid: connection1:0 is operational after recovery (2 attempts)

2010-02-16-04:02:27: iscsid: Kernel reported iSCSI connection 1:0 error (1008) state (3)

2010-02-16-04:02:32: iscsid: connection1:0 is operational after recovery (2 attempts)

should i edit the timeout value on the esxhost for iscsi? This may work around the problem but will not solve it.

Anyone?

0 Kudos
8 Replies
marcelo_soares
Champion
Champion

Take a look at the /var/log/vmkernel at the same time this error happened.

Marcelo Soares

VMWare Certified Professional 310/410

Technical Support Engineer

Globant Argentina

Marcelo Soares
0 Kudos
Jesse1973
Enthusiast
Enthusiast

i did, cat /var/log/vmkernel grep|vmhba1

output is the same, sadly

requested a vmkernel output from the client, awaiting response

0 Kudos
AnatolyVilchins

Could you tell me what is your SAN vendor?

iSCSI Software Support Department

Kind Regards, Anatoly Vilchinsky
0 Kudos
Jesse1973
Enthusiast
Enthusiast

Hi Anatoly

Its NetApp.

0 Kudos
DaemonLord
Contributor
Contributor

Hi all,

Do you use snapmirror?

I have the same scsi timeout errors on a FAS2020 with snapmirror enabled.

I'm still trying to figure out what the problem could be..

grtz

Andy

0 Kudos
Igor_The_Great
Enthusiast
Enthusiast

Question: How to configure ESX for optimal availability with NFS Datastores?

Answer: If you keep loosing NFS shares, NetApp recommends making the following changes on each ESX host:

1. Using VI/vSpere Client connect to ESX host. 
2. Select the Configuration tab.
3. In the Software box, select Advanced Configuration.
4. In the pop-up window, left pane, select NFS.
5. Change the value of NFS.HeartbeatFrequency to 12.
6. Change the value of NFS.HeartbeatMaxFailures to 10.
-Igor If you found this or any other answer useful please consider the use of the Helpful or correct buttons to award points.
0 Kudos
Igor_The_Great
Enthusiast
Enthusiast

Also, please make sure that your NFS and iSCSI traffic is going over 1 or 10GbE nic/hba.

-Igor If you found this or any other answer useful please consider the use of the Helpful or correct buttons to award points.
0 Kudos
bulletprooffool
Champion
Champion

There is a best pratcice guide available for this.

I have pretty much followed this each time I set up NFS storage and had no problems:

http://blogs.netapp.com/virtualstorageguy/2009/07/vsphere-on-netapp-best-practices-released.html

The are several little settings to make (like thos menioned above for heapsize etc)

Good luck

One day I will virtualise myself . . .
0 Kudos