Error from Kernel

[Newbie_From_Hell]

can someone translate this for me please

2009/01/15 17:21:40 kernel:data_out_start(1032) unable to find scsi task 727ad82 c860ad

2009/01/15 17:21:40 kernel:data_out_start(1032) unable to find scsi task 727ad83 c860ae

2009/01/15 17:21:40 kernel:data_out_start(1032) unable to find scsi task 727ad84 c860af

2009/01/15 17:21:40 kernel:data_out_start(1032) unable to find scsi task 727ad85 c860b0

2009/01/15 17:21:45 kernel:cmnd_abort(1143) 727ad88 1 0 42 4096 0 0

2009/01/15 17:21:45 kernel:cmnd_abort(1143) 727ada4 1 0 42 512 0 0

Thanks

[To-M]

Check to match the initiator settings from your system and our iSCSI daemon settings, this error can mean a mismatch.

From the DSS Console perform the following:

Enter CTRL + ALT + W
Then select Tuning options
Then select iSCSI daemon options
Then select Target options
Select the Target of choice and match the settings.

Also you can try the following to improve performance.

maxRecvDataSegmentLen=262144
MaxBurstLength=16776192
Maxxmitdatasegment=262144
maxoutstandingr2t=8
InitialR2T=No
ImmediateData=Yes

[wss]

Hello,

We are having the same issue.

We are using the last DSS build in an iscsi fail-over setup. Since yesterday there are 3 VM's (VMWare ESX) running on this storage (servers are not in use, and thus not producing a lot load).

Since the VM's are running (and we are using the setup) we are getting alot of the following errors:

cmnd_abort(1143) 3a2fd 1 0 42 4096 0 0

cmnd_rx_start(1636) 2 ffffffff -4
cmnd_rx_start(1636) 2 ffffffff -4

data_out_start(1032) unable to find scsi task 8d8fae 47552f
data_out_start(1032) unable to find scsi task 8d8fae 47552f

I already changed the suggested values but this doesn't change a thing. Any help is apprectiated!

[To-M]

Typically when I see data_out_start errors I find a miss match setting from the initiator side in comparison with the iSCSI daemon settings on our side, this is where you made the modifications.

But in the case of a VM environment I found this link below that talks about this error and why it happens. See if this will help and it does not matter if you are using our Volume Replication or not that causes the errors.

They state that it can be caused by the guest Linux kernel and to change the kernel or at least patch its mptscsih driver module.

So what you see on the target side is the errors you have disclosed.
From what they state, it is not a problem "per se, just the target logging warnings about
unexpected actions received from the initiator...


http://www.nabble.com/VMware-ESX-3-%2B-iSCSI-Enterprise-Target-DRBD-gone-terribly-wrong---help!-td11356879.html

[wss]

Hi,

Maybe some additional info, which seems to be more serious. When I got this errors the storage also becomes unavailable in my ESX setup.

Pings from ESX SC are also stopping during this event. Pings to virtual interface as well the ip of the bonding on which the virtual interface is are both stopping.

Since the errors are coming up at open-e I think it is not a network issue ...

Thanks!

[To-M]

Try to ping from the DSS side:

Go to the Console then enter CTRL + ALT + T then Ping.

Also you can check the logs for the dmesg, test (look at the RX side of the Ethx to see any packet errors), critical and error logs to see if there are more errors that might help.

Logs can be found in the GUI from Status > hardware > logs - then click on the download button.

You can also try to change the MaxRecvDataSegmentLenght and
MaxXmitDataSegmentLenght to 65536 to see if that helps if there are not other reasons.