I'm new to this group, so my apologies if this has been answered
previously, though I can't seem to find the answers I'm looking for. I
also come to NetWare not out of choice, so hopefully you'll all bear
with me if you're willing to help.

My issue is this: I have a two-node cluster for a GroupWise 6.5 email
system. Each node houses a large (+100GB) Post Office. The nodes are
backed-up by a standalone NetWare 6.5 system running ARCserve 11.1,
which is connected to an Ultrium 460 (LTO-2) tape drive. All servers
are connected into an isolated Linksys Gigabyte switch, which is,
however, unmanageable.

Prior to installing and configuring iSCSI we were seeing throughput of
500-600MB/min, which was marginally acceptable (though I'd certainly
like to see better throughput than that).

After iSCSI was installed throughput was massacred, down to
200-250MB/min, which has led to a situation where I can only get a full
backup of the systems on the weekend, and one or the other PO during
the week.

Can anyone help me understand why iSCSI is costing so much degradation
in backup throughput? And can anyone recommend some tuning tips to
increase my throughput past the 600MB/min threshold?

Thanks in advance to any and all for your suggestions, comments, and


Some relevant details:

All servers involved run NetWare 6.5 SP4

2-node cluster is comprised of 2 HP DL380 G4 servers (dual Intel Xeon
3.4GHz processors, hyperthreading disabled, 4GB of memory) connected to

Backup server is an HP DL380 G4 with single Intel Xeon 3.4GHz processor
with 3GB of memory.