Isilon OneFS: Int-b/Failover interface is not enabled on the cluster

In the OneFS WebUI, on clicking a node ID, we may notice a failover interface lit amber:

Running ifconfig on the node may show only one internal interface [int - a] having the flag ISIINTERNAL. Also you notice that if int-b/failover is not enabled, the interface does not have an IP address.

ix0: flags=1008843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST,ISIINTERNAL> metric 0 mtu 9000
options=e407bb<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU,VLAN_HWCSUM,TSO4,TSO6,LRO,VLAN_HWTSO,RXCSUM_IPV6,TXCSUM_IPV6>
        ether f8:f2:1e:3e:fb:ec
        inet 192.X.X.X netmask 0xffffff00 broadcast 192.168.1.255 zone 1     ==================> IP address for int-a
        inet6 fe80::faf2:1eff:fe3e:fbec%ix0 prefixlen 64 scopeid 0x1 zone 1
        inet6 fdfe:9042:c53d:0:faf2:1eff:fe3e:fbec prefixlen 64 zone 1
        nd6 options=21<PERFORMNUD,AUTO_LINKLOCAL>
        media: Ethernet autoselect (10Gbase-Twinax <full-duplex,rxpause>)
        status: active

ix1: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
options=e407bb<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU,VLAN_HWCSUM,TSO4,TSO6,LRO,VLAN_HWTSO,RXCSUM_IPV6,TXCSUM_IPV6>
        ether f8:f2:1e:3e:fb:ed
        inet6 fe80::faf2:1eff:fe3e:fbed%ix1 prefixlen 64 scopeid 0x2 zone 1    ==================> No IP seen for Int-b
        inet6 fda7:e6ee:2e09:0:faf2:1eff:fe3e:fbed prefixlen 64 zone 1
        nd6 options=21<PERFORMNUD,AUTO_LINKLOCAL>
        media: Ethernet autoselect (10Gbase-Twinax <full-duplex,rxpause>)
        status: active

You can confirm this by checking the /etc/ifs/array.xml file on the node:

<?xml version="1.0" encoding="utf-8"?>
<isilon version="0x8010000">
  <name>clustername</name>
  <joinmode>manual</joinmode>
  <interfaces>
    <interface name="int-a" flags="enabled_ok cluster_traffic" index="0">
      <netmask>255.X.X.X</netmask>
      <ipranges>
        <iprange>
          <low>192.X.X.X</low>
          <high>192.X.X.X</high>
        </iprange>
      </ipranges>
      <mtu>9000</mtu>
    </interface>
    <interface name="int-b" flags="disabled" index="1"> =============================> int-b disabled
      <netmask>255.X.X.X</netmask>
      <ipranges>
        <iprange>
          <low>192.X.X.X</low>
          <high>192.X.X.X</high>
        </iprange>
      </ipranges>
      <mtu>9000</mtu>
    </interface>
    <interface name="lpbk" flags="disabled failover" index="0"> ======================> Failover disabled
      <netmask>255.X.X.X</netmask>
      <ipranges/>
      <mtu>9000</mtu>
    </interface>
  </interfaces>

As shown in the snippet above, clicking the failover interface notifies that int-b/failover is not enable in the cluster, and no IP ranges are specified for failover:

In order to rectify this, kindly specify the failover range [if int-b range is not specified, kindly populate this range as well]. Once the range is provided, enabled the state of the interface as shown below:

Once the state of the interface is enabled, click submit.

NOTE: When you submit the change, you are prompted for a simultaneous reboot of the entire cluster in order to commit the change. Kindly perform these steps in a maintenance window to avoid disruption to end users.

Once all the nodes are back up after the reboot, the int-b and failover interfaces should be enabled.

Step-by-step procedure for the above:

* OneFS 8.x:

In the WebUI:

Step 1: Cluster Management > Internal Network > int-b/Failover
Step 2: If the failover interface is lit amber, verify int-b and the failover IP range is populated. If not, specify the int-b and failover IP ranges.
Step 3: Enable the state of the interface if it is disabled.
Step 4: Submit the changes made.