Skip to Main Content

Infrastructure Software

Announcement

For appeals, questions and feedback about Oracle Forums, please email oracle-forums-moderators_us@oracle.com. Technical questions should be asked in the appropriate category. Thank you!

Interested in getting your voice heard by members of the Developer Marketing team at Oracle? Check out this post for AppDev or this post for AI focus group information.

Sun cluster 3.3 in guest LDOM - disable IPMP for public

871253Mar 12 2013 — edited Mar 14 2013
Hi,
i have setup HA-NFS with Sun cluster 3.3 in two guest LDOMs on two separate T4-2. The guest LDOMs get one VNET for interconnect and one VNET for public; the vswitch for public is 2x10g port aggregated on the control domain, so there is no need to have IPMP for public interface in the ldom. But sun cluster automatically does IPMP for public even with one vnet as failover however is it coming up as failed.

I have the exact same setup for another cluster in LDOMs and public interface IPMP with one vnet is working fine. Can't seem to get this one working. any tips?

NODE 1:

vnet2: flags=19000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4,NOFAILOVER,FAILED> mtu 1500 index 3
inet 192.168.130.141 netmask ffffff80 broadcast 192.168.130.255
groupname sc_ipmp0
ether 0:14:4f:f8:b9:47
vnet2:1: flags=1011040843<UP,BROADCAST,RUNNING,MULTICAST,DEPRECATED,IPv4,FAILED,FIXEDMTU> mtu 1500 index 3
inet 192.168.130.150 netmask ffffff80 broadcast 192.168.130.255


cat /etc/hostname.vnet2
nfsp0gux001 group sc_ipmp0 -failover


clrs status -v

=== Cluster Resources ===

Resource Name Node Name State Status Message
------------- --------- ----- --------------
nfsp-rs nfsp0gux002 Offline Offline
nfsp0gux001 Online Online - Service is online.

nfsp-hastp-rs nfsp0gux002 Offline Offline
nfsp0gux001 Online Online

nfsp-lh-rs nfsp0gux002 Offline Offline
nfsp0gux001 Online Degraded - IPMP Failure.

Node 2:

vnet2: flags=19000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4,*NOFAILOVER,FAILED*> mtu 1500 index 3
inet 192.168.130.142 netmask ffffff80 broadcast 192.168.130.255
groupname sc_ipmp0
ether 0:14:4f:f8:9f:25

Comments

Alex-D

Thank you, great post!

Regards

StephenCross-Oracle

The "Build the OpenStack provider" step is not necessary.  There are pre-built release binaries for the openstack provider which Terraform will automatically fetch when you run terraform init if the terraform configuration contains any `openstack` resources:

$ terraform init  Initializing provider plugins... - Checking for available provider plugins on https://releases.hashicorp.com... - Downloading plugin for provider "openstack" (1.3.0)...
1102108

I have a question about this post. Do you know why the repository I can use for the internal version of Oracle Linux (OBI)? I am asking because I recently created a cluster with Kubernetes and Oracle Cloud Infrastructure using that image version but I did not find Terraform in the mandatory repositories. If I use ol7_developer will update my kubectl to 1.10 causing several issues

Greetings,
Gabe

Avi Miller-Oracle

I have a question about this post. Do you know why the repository I can use for the internal version of Oracle Linux (OBI)? I am asking because I recently created a cluster with Kubernetes and Oracle Cloud Infrastructure using that image version but I did not find Terraform in the mandatory repositories. If I use ol7_developer will update my kubectl to 1.10 causing several issues

Greetings,
Gabe

We only provide Terraform in the ol7_developer channel. If you do not want to upgrade kubectl you will need to exclude it in your Yum configuration. You can review the Oracle Linux 7 Administrator's Guide on instructions on how to configure yum to include or exclude certain packages per repo.

1 - 4
Locked Post
New comments cannot be posted to this locked post.

Post Details

Locked on Apr 11 2013
Added on Mar 12 2013
4 comments
2,707 views