NetScaler HA Odd Issue

[Stan Svetec]

Hi,

I have 2 5901's both on the latest 12.x Maintenance Release.  The 5901's are in a HA pair.  I have an FTP server fronted by a VIP that sits on these 5901's.

 

When NS1 is the Active node, connectivity to the VIP (by way of PING) is OK for a period of time, before it no longer responds.

When I force a failover to NS2, connectivity to the VIP is constant and does not fail.

When I fail back over to NS1, connectivity is restored briefly, but soon stops responding.

 

Aside from the NSIP, both NetScalers are configured identically, however, NS1 stops serving the FTP whilst NS2 has no problems.

 

A little more information.

As I type this, NS1 is active and FTP is not responding.  When I disable/enable the VIP on NS1, FTP starts responding again.  For how long?  I'm not sure since I'm not timing it.

 

Thoughts?

[Raman Kaushik]

sounds like an IP conflict.

 

You can check:

>shell

#nsconmsg -K /var/nslog/newnslog -d consmsg | grep -i conflict

 

or

 

>shell

#zgrep -i conflict /var/log/ns.log*

[Stan Svetec]

On 4/29/2019 at 11:51 AM, Raman Kaushik said:

sounds like an IP conflict.

 

You can check:

>shell

#nsconmsg -K /var/nslog/newnslog -d consmsg | grep -i conflict

 

or

 

>shell

#zgrep -i conflict /var/log/ns.log*

It was less a conflict and perhaps more of a just a dodgy IP.  Once I recreated the VIP with a new IP, connectivity was rock solid.