IXSystems X10 HA: Nodes CARP states do not agree

TrueNAS General
1

Good Day All,

I have done some research on this forum and a few other places but I do not see anything on this message: Nodes CARP states do not agree. Can anyone tell me what the problem is and how to resolve it? I believe it is preventing the HA function of the X10 from being enabled. Thank you.

Kindly,

Soul

2

Can you describe the networking configs on both nodes, and whether they’re running Core or Scale? Which OS version?

3

Good Day,

Version: TrueNAS-13.0-U4 and the configuration is static ips. We are running the enterprise addition.

4

This forum is for self-support for CE, not the Enterprise Edition. You should open a support ticket for your issue.

CARP is essentially VRRP. Both nodes need to share the same virtual IP, and all the IPs (both physical & virtual) have to be on the same VLAN.

5

I understand this is for core support, but we lost our maintenance agreement, and we were told by paid support to come here. Do you know how to configure this device using the command line? Is there documentation somewhere that explains how to do that?

6

You might have to wait for someone from TrueNAS Enterprise to post. Adding a link to your other thread so others can try to follow recent changes

7

Failover isn’t supported here, unfortunately. All we can do is point you to the official documentation:

As for CLI, that’s going to use midclt commands to invoke the websocket API, but the ones for failover aren’t documented well. Like @SmallBarky said, you’re going to have to wait for someone from TrueNAS Enterprise to chime in.

8

Reading the other thread, the likely reason is your virtual IPs don’t match the new IP subnet your X10s now inhabit, or there’s a DNS mismatch between the shared virtual name and the shared virtual IP.

9

Thanks Samuel_Tai and SmallBarky. I will look at this documentation and do as you say and await the enterprise guys. The information you shared is a great start!

10

Node CARP states don’t agree would indicate that the two controllers are not able to talk to eachother via the multicast group in the subnet they are both in. In general, this can be a VLAN mismatch, misconfigured interface or even just an unplugged cable.