iSciPrt Errors

[lindy37]

On a Windows 2008 R2 vm, running Exchange 2010, I get the 2-3 errors "iSciPrt" errors (connection to target is lost) in the System Log of the Windows server when I take a synchronized snapshot of Exchange Data with verify turned on.  They dont happen every time, but most times.  They dont seem to happen when verify is off, second snapshot of the day for replication.  The errors all happen in 1 second according to the logs and then they recover (connection to target resumes).  I think it is happening when it is mounting the snapshot to do the verify of the Exchange Logs.

I have all hotfixes applied. (hotfix checker).  I am running the latest NCM.  I am running MPIO on this VM.  I am running in-guest iSCSI and VMware is not involved in the snapping process.  Everything seems to be working fine, I just happened to see the errors when reviewing the logs.

I found this MS KB and these are the errors I get.  https://support.microsoft.com/en-us/kb/981482

It talks about tuning or turning off TCP Delayed ACK for the Windows Server.

My question is this something I need to be worried about?

Thanks for any help!

[jonathan_disley · Accepted solution]

Hi Chris,

If you haven't already, I'd suggest you log a call with Nimble Support regarding this question.  There is error logging functionality within the Nimble DSM (part of NCM that handles iSCSI connections and multi-pathing) so it's possible that support will be able to provide further diagnosis using the DSM logs to ascertain what is happening and advise you regarding whether it is of concern or not.   

Best regards,

Jonathan