PowerScale: Isilon: Management interface would be unreachable after OneFS upgrade.

No IP in the Management network pool would be responsive to the client. The access to the WebUI/ssh by Management IP would fail. 

Cluster-1# isi network pools list
ID                           SC Zone                       IP Ranges                   Allocation Method
---------------------------------------------------------------------------------------------------------
groupnet0.mgmt.mgmt          cluster_mgmt_xyz.com    10.11.X.X-10.11.X.X static

For Example, Here the management interface is bge0, which is active but though unresponsive post OneFS upgrade. 

Cluster-1# isi_for_array -s ifconfig bge0
Cluster-1: bge0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
Cluster-1:      ether 00:60:16:9e:X:X
Cluster-1:      inet 10.11.X.X netmask 0xffffff00 broadcast 10.11.X.X zone 1
Cluster-1:      inet 10.11.X.X netmask 0xffffff00 broadcast 10.11.X.X zone 0
Cluster-1:      media: Ethernet autoselect (1000baseT <full-duplex>)
Cluster-1:      status: active
Cluster-2: bge0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500
Cluster-2:      ether 00:60:16:9e:X:X
Cluster-2:      inet 10.11.X.X netmask 0xffffff00 broadcast 10.11.254.255 zone 1
Cluster-2:      media: Ethernet autoselect (1000baseT <full-duplex>)
Cluster-2:      status: active

Ping, from client system would result in "Request timed out" with 100% packet loss for any Management network Pool IP. 

C:\Users\admin>ping 10.11.x.x

Pinging 10.11.x.x with 32 bytes of data:
Request timed out.
Request timed out.
Request timed out.
Request timed out.

Ping statistics for 10.11.x.x:
    Packets: Sent = 4, Received = 0, Lost = 4 (100% loss).

 

1. Log in to an ssh session to a node in the cluster
2. Manually switch the Management interface bge0 with the help of below commands: 

Cluster-1#ifconfig bge0 down >>Taking the interface down
Cluster-1#ifconfig bge0 up  >>Taking the interface up. 
 
To switch the management interface for another node: 
Cluster-1# isi_for_array -n2 ifconfig bge0 down
Cluster-1# isi_for_array -n2 ifconfig bge0 up

3. Once you switch the management interface in question, the client system would be able to ping the management interfaces successfully. 

C:\Users\adminby>ping 10.11.X.X

Pinging 10.11.X.X with 32 bytes of data:
Reply from 10.11.X.X: bytes=32 time<1ms TTL=128
Reply from 10.11.X.X: bytes=32 time<1ms TTL=128
Reply from 10.11.X.X: bytes=32 time<1ms TTL=128
Reply from 10.11.X.X: bytes=32 time<1ms TTL=128

Ping statistics for 10.11.X.X:
    Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
Approximate round-trip times in milli-seconds:
    Minimum = 0ms, Maximum = 0ms, Average = 0ms

As per engineering, customers should avoid using bge0. 
Please see the below article as well: 

{Isilon} Gen 6: The bge0 (MGMT) interface should be used for SSH only. | Dell US