VMware Cloud Community
82Titleist
Contributor
Contributor
Jump to solution

VDS active/active uplink behavior with failback option "no"

All,

I am encountering the situation that is specified in this blog: http://tinyurl.com/hvheosx

I have tested the configuration with active/active uplinks (2) and the failback option to "no", but it always results in failing back to the original uplink.  I am able to configure the uplink policy to active/passive, and the configuration works as specified, but the configuration seems less ideal as new connections wouldn't be balanced across the uplinks.  Has there been an update to the VDS from 5.0 that rectifies this behavior? 

Thanks!

Tags (2)
Reply
0 Kudos
1 Solution

Accepted Solutions
npadmani
Virtuoso
Virtuoso
Jump to solution

Nic teaming policy Failback set to Yes or No would only give you effect when you set NIC Teaming Active and Standby.

From the Failback drop-down menu, select whether a physical adapter is returned to active status after recovering from a failure.

If failback is set to Yes, the default selection, the adapter is returned to active duty immediately upon recovery, displacing the standby adapter that took over its slot, if any.

If failback is set to No for a standard port, a failed adapter is left inactive after recovery until another currently active adapter fails and must be replaced.

Ref: vSphere 6.0 Documentation Center

if you are using Active-Active then it's Load Distribution methods which you are using like 'Originating Virtual port ID', 'Source Mac Hash', etc.. would take care of distributing load among available active adapters. It has nothing to do with what you have setup in Failback part of this policy.

So, even the version 6 of VDS is also having same behaviour, no changes in this part.

Narendra Padmani VCIX6-DCV | VCIX7-CMA | VCI | TOGAF 9 Certified

View solution in original post

Reply
0 Kudos
2 Replies
npadmani
Virtuoso
Virtuoso
Jump to solution

Nic teaming policy Failback set to Yes or No would only give you effect when you set NIC Teaming Active and Standby.

From the Failback drop-down menu, select whether a physical adapter is returned to active status after recovering from a failure.

If failback is set to Yes, the default selection, the adapter is returned to active duty immediately upon recovery, displacing the standby adapter that took over its slot, if any.

If failback is set to No for a standard port, a failed adapter is left inactive after recovery until another currently active adapter fails and must be replaced.

Ref: vSphere 6.0 Documentation Center

if you are using Active-Active then it's Load Distribution methods which you are using like 'Originating Virtual port ID', 'Source Mac Hash', etc.. would take care of distributing load among available active adapters. It has nothing to do with what you have setup in Failback part of this policy.

So, even the version 6 of VDS is also having same behaviour, no changes in this part.

Narendra Padmani VCIX6-DCV | VCIX7-CMA | VCI | TOGAF 9 Certified
Reply
0 Kudos
82Titleist
Contributor
Contributor
Jump to solution

I see your point, and that is exactly what I was looking to validate from what I am seeing during tests.


After some thought, I am pretty confident that the reason it is "failing back" is because of the load balancing policy is being re-applied.  If I have route by originating port ID (or any other policy), the virtual switch configuration is going to want to pin to the previous uplink it was using.  I think it is more of a misinterpretation of the "failback" option on my part.

Reply
0 Kudos