Hey, Ken.

So this raises some questions. First, what firmware are you running on the array? Second. how many connections do you have from the host to the switch, and then from the switch to the array?

Hey, Ken.

This sounds like it got a little more complicated than a simple cabling issue. I'm going to send you an email, and if you could reply with a support bundle, that would be great!

I'll take a look at it for you. I may even come back with more questions, but hopefully we'll get it solved for you.

Let me know!

Hey, Ken.

So, I've gone over your support bundle. I have a couple suggestions.

It's showing that all the virtual disks are owned by the raid controller in slot 1, YET I can see that you have iSCSI sessions to each port on BOTH raid controllers. Based on that, I can see maybe we have a config issue.

Do all three hosts have the MDSM installed on them? I'm wondering if you're not missing some multipath drivers on a couple of these hosts. In the device manager, you can look at the disks, and the properties should say, "multipath disk". If it doesn't, it will say *dell scsi disk* or something to that affect.

Also, it might be that you have the mpio software installed, but it's set to fixed or something OTHER than Round Robin. (which is what we suggest it's set at)

Another thing to consider is, if you set up the cluster BEFORE you installed the software, then the cluster  shared volumes in the windows cluster manager may be trying to reference only a specific path because that's all it knew when the cluster was created.

If THAT'S the case, you would need to remove, and re-add the CSV's in cluster manager.

Then, you can redistribute the virtual disks in the MDSM, and see that they alternate from RC0 to RC1 back and forth.

Start here, and check these things and let me know how it goes, and what you find.

Hopefully it makes sense. Let me know if you have any questions.

Have a good rest of the week!

Hi Daniel,

The firmware is 08.20.05.60. I have eight connections from the SAN to the switches, and four connections from each server to the switches. The issue hasn't reoccurred since I posted the last message.

Thanks,

Ken

Hello Daniel,

All three hosts have the MDSM software. in device manager, disk drives it shows four "DELL MD32xxi Multi-Path Disk Device" instances (but also shows eight "DELL Universal Xport SCSI Disk Device" instances.) I looked into the iSCSI Initiator Properties, Targets tab, Properties, MCS and it shows round Robin.

How can I find out if the cluster shared volumes is trying to reference only a specific path? The SAN was set up over four years ago, and the systems, operating systems and software have all changed several times since then. The latest change probable was updating the Dell MDS software, in which I remove the software, then reinstall it with the newest version. The new Dell MDS software seems to set everything up for me automatically and correctly after installation - happy about that.

It seems that I have always received an event log errors from VDS Basic Provider, SmProvider and Application Error (SMPROV~1.EXE) which I had googled extensively previously, but never found a solution. Some errors seem to be related to Veeam backup jobs running.

What do you think? Any ideas?

Thanks,

Ken