If a private interface has faulted or is down, certain commands like "cluster show", "storage disk list" would show an error in their outputs
Article ID: 100043765
Updated On:
Description
Examples
Incorrect output:
access7410> cluster show
Node State CPU(15 min) ens192(15 min) ens193(15 min) ens224(15 min) ens225(15 min)
% rx(MB/s) tx(MB/s) rx(MB/s) tx(MB/s) rx(MB/s) tx(MB/s) rx(MB/s) tx(MB/s)
---- ----- ----------- -------- -------- -------- -------- -------- -------- -------- --------
access7410_02 RUNNING 4.63 0.58 0.15 0.00 0.00 0.00 0.00 0.00 0.00
access7410> storage disk list
Disk access7410_02
========= ==============
emc0_0389 OK
emc0_0390 OK
Correct output:
access7410> cluster show
Node State CPU(15 min) ens192(15 min) ens193(15 min) ens224(15 min) ens225(15 min)
% rx(MB/s) tx(MB/s) rx(MB/s) tx(MB/s) rx(MB/s) tx(MB/s) rx(MB/s) tx(MB/s)
---- ----- ----------- -------- -------- -------- -------- -------- -------- -------- --------
access7410_02 RUNNING 5.14 27.29 0.57 0.00 0.00 0.00 0.00 0.00 0.00
access7410_01 RUNNING 3.34 1.21 0.17 0.00 0.00 0.00 0.00 0.00 0.00
access7410> storage disk list
Disk access7410_01 access7410_02
========= ============== ==============
emc0_0389 OK OK
emc0_0390 OK OK
Cause
There is no failover of private IPs between the private interfaces. So when the first private interface(priveth0 in releases before VA 7.4.1 or interface on which internal IPs are assigned) goes down, it causes the connection to break while executing the commands on remote nodes. Veritas Access uses only the first private interface to communicate within the nodes of the cluster. This is done so as to not disturb the distribution of LLT traffic
Resolution
To avoid this issue, a private bond can be created between private interfaces via the CLISH command "network bond priv-create".
Issue/Introduction
