Link to home
Start Free TrialLog in
Avatar of cssunetadmin
cssunetadmin

asked on

iSCSI Connection

We have an EMC San device connected to a Windows 2008 R2 server for user folders.  When a large amount of traffic is sent over this connection- for example to backup the files or when transferring large files such as video files- the connection will be dropped regularly.  Users often have to reboot to re-establish a connection and the back up job fails.  Weh ave talked to tech support of the San device and they indicate that their device is functioning properly.  We have seen some issues related to Win2008 in forums but have the applicable patches applied.  Typical error messages in the event viewer for the server are attached.

Can anyone help us out?  Thanks!
errormessages.txt
Avatar of heperez
heperez
Flag of United States of America image

Try enabling Jumbo Frames between the SAN and your NIC card.
Are you going through switches or direct connection?
Avatar of cssunetadmin
cssunetadmin

ASKER

We implemented jumbo frames and while the backup ran about 2 hours longer, it still failed with the same errors.  We are going through a switch- both the San and the server are plugged into the same switch.  One thing I didn't note originally, the server a virtualized, running on VSphere v4.
Is it on the same VLAN?
ASKER CERTIFIED SOLUTION
Avatar of heperez
heperez
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
We do have jumbo frames configured on the NIC accessing the San.  Since implementing jumbo frames, we can backup from the server for three - four hours before having any problems.  The error message has changed a little.  It now reads, "An error was detected on device \Device\Harddisk1\DR1 during a paging operation."

Searching the Microsoft KB, we found one article about increasing the receive buffers.  We did this, changing them to the recommended 2048.  Last night the backup ran for 6 hours before crashing.