iSCSI slow HBA rescans

[Groundbeef79]

We have a 3 host cluster attached to an Equallogic SAN. Everything is configured there following Dell best practices with redundant switching, network connections, etc. No problem there. Things started to turn sour when we introduced a second iSCSI device into the mix. To keep disk costs low, we decided to put less important/less used VMs on a Windows Storage Server device from Dell as well as use it for storage of ISOs, some file storage, etc. Specifically, an NX3000 NAS using Microsoft's iSCSI initiator. Now I'm not even sure if this is a problem, but would like to avoid any in the future! When I rescan the storage adapter, it takes forever. It did not do this until I introduced the Microsoft solution into the mix. I can't really find any best practice documentation on attaching that to VMware. About all I can turn up has to do with HyperV and the like. Does anybody know of any specific SCSI timeouts or anything like that that has to be set? I setup the NAS similar to our SAN, it has fully redundant teamed NICs setup in LAGs on the switches. Any thoughts? Thanks!

[actixsupport]

Hi,

First thing I'd do is check the output of the messages(ESXi) or vmkernel(ESX) when you kick off a rescan. That should show up where slowdown is.

I'm hazarding a guess is that it could be related to the way the MS box handles multiple logons with the same to a target.

If you've set it up optimised for EQL then you'll have multiple initiator connections coming off each host, some iSCSI targets don't like this and while they still work they're not optimal.

If you can put up a detailed description of your setup and the output of you log files we might be able to help out a bit more.

Cheers

Ray

[admin]

Ensure if your using Jumbo Frames its end to end if not set mtu 1500 end to end, enable portfast, disable spanning tree - ensure flow control is enabled (vendor specific)

Have you installed a DSM for MPIO ?

Are you MPIO from ESX to the storage as in vmk binding to vmhba?

[Groundbeef79]

Well, I know this is late, I was on vacation then forgot about the question.  Thanks for the input guys.  I'm going to give up on it for now.  We've got our Dell storage guy coming in the next couple of weeks and they're going to take an in-depth look at it.  I can tell you this though.  When I have a host connected to the storage server all by itself scans are fast.  Likewise when the host is connected to the iSCSI SAN.  When two are connected is where the slow down happens.  We'll see what happens with the storage guy.  I'll post a solution when I have one.

[hostasaurus]

Check the switch(s) the new storage connects to for any errors.  I was having the same slow rescans with our EMC storage and it turns out it was just a dirty fiber connector on one of the four ten gig paths to the storage; it was producing CRC errors that the switch was recording.  The percentage was pretty small but steady.  Pulled it, cleaned it, problem went away.