Unsolved
This post is more than 5 years old
1 Rookie
•
7 Posts
1
3568
September 24th, 2018 15:00
IPv6-only iDRAC network
Hi,
I'm wondering whether anyone has been brave enough to run an IPv6-only LOM network.
I've tried this with iDRAC 9 (Dell R740, latest version) in an IPv6-only network with autoconfig (SLAAC) and OpenManage Enterprise 3.0 on a dual stack machine in another network. It works pretty well (even with ActiveDirectory logins), except for Firmware updates.
When OME tries to push a firmware to the server it initiates a download, then fails with
Unable to transfer a file, FOLDER04936438M_1_Drivers-for-OS-Deployment_Application_RHM02_WN64_18.04.07_A00.EXE, because of one or more of the following: 1) remote host cannot be reached. 2) remote share name is not present. 3) incorrect credentials are entered.
Unfortunately there is nothing decent in either the logs of the iDRAC nor the OME. Especially the host/path is missing. It works fine with a dualstack host.
Anyone ever tried something like that? :-)
Bernhard
No Events found!
DELL-Chris H
7 Practitioner
•
9.7K Posts
•
48K Points
0
September 25th, 2018 08:00
Bernilrz,
You can certainly do that, as seen on page 45 here, you can configure the iDrac for IPv6. The only issues, as you stated, are that with the iDrac configured as IPv6 it doesn't support auto-discovery (page 45), it doesn't support scheduling automatic firmware updates (page 70), or support scheduling automatic backup server profiles (Page 78).
Hope this helps.
DELL-Pupul M
2 Intern
•
1K Posts
0
September 25th, 2018 23:00
Hi,
Could you please specify the update method you are using? Is it FTP or a local repository?
bernilrz
1 Rookie
•
7 Posts
0
September 28th, 2018 01:00
The OME is using "Latest component firmware versions on Dell.com" as Catalog Source.
The iDRAC is using whatever OME is telling it to use. On IPv4 it is NFS (Port 2049/tcp), so I'd assume for IPv6 it will be the same.
bernilrz
1 Rookie
•
7 Posts
0
September 28th, 2018 10:00
Note that when you decide to have OME configure SNMP traps on discovery it also adds the literal _IPv4_ address as trap destination on discovered IPv6-only devices. I think this is likely the same root cause, if OME would send the command to mount the firmware store using a literal IPv4 address it would also fail.