- 3,714,823 Users
- 2,242,634 Discussions
- 7,845,082 Comments
Forum Stats
Discussions
Categories
- Industry Applications
- 3.2K Intelligent Advisor
- Insurance
- 1K On-Premises Infrastructure
- 364 Analytics Software
- 32 Application Development Software
- 1.7K Cloud Platform
- 700.5K Database Software
- 17.4K Enterprise Manager
- 7 Hardware
- 167 Infrastructure Software
- 91 Integration
- 52 Security Software
Best practise to configure network redundancy for LDOM and Zones

Hi,
I would like to ask for best practice idea to configure network redundancy using EMOC.
I have CDOM with 8 NIC. (1 x MGMT, 1 x BCK, 2 x APP, 2 x DB, 2 x WEB). No IPMP/AGGR created at this level.
In EMOC, i create network domain for each vlan and attach the network to my multiple LDOM. For example, 1 LDOM with 1 x MGMT (with IP), 2 x APP (Interface not plumb)
Inside this LDOM, i will have 10 zones that will use the APP segment.
I would like my zone to have full network redundancy for the segment with 2xNIC, what is the best way and how to do it.
Thank you.
Regards,
BH
Answers
-
Hello BH,
You should be able to create an I/O service domain depending on the T series HW where the service I/O has assigned one of the network interface sets and the primary has the other. Then you create a VSW for the primary and the secondary for those networks with the appropriate VLAN information. Then for the ldom guest your assigning one NIC for the same network from the primary and secondary domains. This should allow network connectivity if primary or secondary instances has troubles. This works as well with HBA's if your using fiber channel for Storage.
The OVM for Sparc documentation talks about these I/O, root and SRIOV types of domains and the Ops Center documentation covers configuring via Ops Center.
The other concept unique to Ops Center is the server pool where multiple cdom hosts all configured the same regarding the Storage Libraries and networks associated to them are made into a Server Pool a logical construct in Ops Center. If any single cdom host becomes unavailable the OC environment can recognize and try and auto-migrate a guest from the original host to an alternate cdom host if enough cpu and memory resources are available.
Main Ops Center 12.4.0 (current release) documentation URL
https://docs.oracle.com/cd/ops-center-12.4/
Enterprise Manager Ops Center Virtualization Reference
5 Oracle VM Server for SPARC
6 Server Pools
I hope this helps
-
That's a really good answer.