Page 1 of 1

ESXi boot to SAN

Posted: Thu Jul 28, 2022 3:30 pm
by MN99GT
We just installed our new Alletra 6030. We had a 3Par that we booted from SAN with all our ESXi hosts. I created a boot LUN on the Alletra and was able to install ESXi with no problems. However, every time I boot the server, the Alletra sends a warning:

Object: Alletra6030-01-A
Status: Open
Acknowledged By: N/A
Acknowledged time: N/A
Reminder repeat period: 1 day(s)
Type: 10023
ID: 3
Message: Array Alletra6030-01 has Fibre Channel Initiator WWPNs 50:06:0b:00:00:c3:4a:0c connected only to controller A. If the condition persists, ensure your Fibre Channel zoning includes both controllers to prevent disruption.

I have both HBA configured in the BIOS, although one is obviously primary. I'm just wondering if this is the way it is? Or if there is some way around this?

Thanks!
Mike

Re: ESXi boot to SAN

Posted: Thu Jul 28, 2022 10:48 pm
by Richard Siemers
Glad to see that alert!

It's possible that you don't have both WWNS in the same initiator group, so the nimble things the 2 WWNS are different hosts, OR just when the server boots, it is only logging into the SAN with one card, until the OS loads and initializes MPIO, then brings the redundant paths. In that case, I would expect the alert is auto-cleared after that happens.

Nimble/Alletra 6K support is great, they answer the phone quick. Give the a call to get to the root issue, and make sure zoning best practice is followed. If it is an expected alarm from booting from SAN (that short period where the boot process is not using MPIO), see what they recommend for mitigating the false/expected alarm. Please share what you learn! If you are in the USA the number is 877-364-6253

Re: ESXi boot to SAN

Posted: Fri Jul 29, 2022 12:29 am
by MammaGutt
Yeah..

Remember that the Alletra 6000 is based on Nimble with Active/Passive controllers. From the looks of your error message you've only zoned your host to the currently active controller. If that controller fails (or reboots), I would expect that your host loses connectivity.

So as Richard says, check your zoning and/or reach out to Nimble support.