Updated an ESXi host from v7.0 U1d (17551050) to v7 U2a (17867351) using Image Mode in Lifecycle Manager. After reboot the iSCSI initiator's name changed breaking iSCSI connectivity.
Whilst I didnt record its previous name (I overwrote the SAN ACL), Altthough I can show the differences from other identical hosts in the cluster. The auto generated number after the hostname completely changed and appended with a ":64" . This is very concerning.
v7U1d = iqn.1998-01.com.vmware:labesx06-4ff17c83
v7U2a = iqn.1998-01.com.vmware:labesx07:38717532:64
It also now shows an iSCSI Alias of "iscsi_vmk".
I couldnt find any references to this change in the release notes. Am I going crazy here?
I decided to try another host, same problem. Undocumented iSCSI name changes. Anyone else getting this?
No, you are not crazy as I am also getting the same issue too, I first noticed it with I ran the non-critical security updates. I assumed it was because I had not used the Dell branded ESXi image so I reverted the patch.
I imported the Dell EMC VMware-VMvisor-Installer-7.0.0.update02-17867351.x86_64-DellEMC_Customized-A03.iso image, ran the upgrade again still no storage. At that point I noticed the iSCSI initiator name had changed from ....-418bdc22 to ...:21404486:64, WTH!
I changed it back to the original name, rescanned storage and adapter, no luck. I then rebooted and now it has all come back up as it was before.
Nice move VMWare, thanks a lot!
Anyway, relatively easy fix for me.
I have a response from VMware support...
This is a known issue with 7.0U2 and here configuration restore is failing because of invalid iqn of the software adapter.
Ofcourse its another "Internal Bug" with no KB private or public.
Unfortunately even we do not have any internal kb's yet for reference. We are right now getting all these information on a different channel from our Engineering team. Sorry again.
Joke
What is the SR number for this?
I found the case internally... will push engineering/support to release a KB.
Thank you depping
Just an update, KB is being worked on, and situation is being investigated by engineering.
Just another indicator that 7.0 U2(a) is a complete mess. Never had so many issues with previous upgrades.
Issue is identified, KB should be published shortly, This should only happen when upgrading from U1 to U2, when you use randomly generated IQNs.
please create a log dump and create a support request!