cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Showing results for 
Search instead for 
Did you mean: 

Community Tip - Want the oppurtunity to discuss enhancements to PTC products? Join a working group! X

Setting Up VMware Load Balancer with a ThingWorx High Availability Deployment

nhunter
7-Bedrock

Setting Up VMware Load Balancer with a ThingWorx High Availability Deployment

Hi All,

 

First post on the forum.

 

I was wondering if anyone could provide guidance on setting up Thingworx HA with a VMware Load balancer, similar to the Azure guide I found in the IoT Tips section.

 

I've used the following guide to date, which references a HAProxy config.

Load Balancers for ThingWorx High Availability (ptc.com)

 

What i'm finding is that I get SSO timeouts when navigating to the URL but only when the two Thingworx nodes are online during my testing.

 

When reviewing the connected users I see multiple sessions for my name, each time I refresh the SSO error I get a new session.

 

I understand through talking to PTC Tech support that this is a Load Balancer Persistence (Sticky Sessions) issue.

 

Tech support call raised with PTC and VMware to resolve, but if anyone has seen this before i'd be grateful to hear how you resolved it.

 

nhunter_0-1649941487794.png

Thingworx HA Setup Version 9.3

Ping Federate (9.3.3) CAS / SP = ADFS

Thingworx Foundation x2

Connection Servers x2

Integration Runtime x2

Zookeeper x3

Ignite x2

 

Thanks in advance.

Neil

ACCEPTED SOLUTION

Accepted Solutions

Hi Dmitry,

 

Sure, I hope the below helps others in time with the same problem.

 

The team found that if we changed the algorithm used for the CXServer and Platform on the Load balancer pools to IP-HASH the problem was resolved.

 

Regards

 

Neil

View solution in original post

2 REPLIES 2

Hi Neil @nhunter!

 

Did you manage to sort this out?

Not that I have any clues / ideas to suggest, just genuinely interested in the topic.

Hi Dmitry,

 

Sure, I hope the below helps others in time with the same problem.

 

The team found that if we changed the algorithm used for the CXServer and Platform on the Load balancer pools to IP-HASH the problem was resolved.

 

Regards

 

Neil

Announcements


Top Tags