CQ instance down still request going to the respective web server in clustered environment | Community
Skip to main content
Level 4
October 16, 2015
Solved

CQ instance down still request going to the respective web server in clustered environment

  • October 16, 2015
  • 6 replies
  • 1795 views

Hi,

I have a clustered CQ environment. There are 2 CQ servers in my cluster hosted on different web servers.

I have shutdown one of the CQ instance but web server is still up and running which is hosting that CQ instance.

When I hit the page, it gives me an error - Bad gateway, that means though CQ instance is down request is still going to web server.

Shouldn't it be that if one CQ instance is down, request should go to another web server with second cluster node?

Regards,

Shallu Rohilla

This post is no longer active and is closed to new replies. Need help? Start a new post to ask your question.
Best answer by joerghoh

Then configure your loadbalancer not to check the healthstate of the webserver/dispatcher, but to check the health of your CQ instance sitting behind the webserver/dispatcher. The probably most easiest way is to append a "?" at the end of the healthcheck URL.

6 replies

October 16, 2015

Hi Shallu,

Have you Added both of your CQ instances in the "/renders" section of your Dispacther Configuration?. This could be due to Web server not being able to Connect to your Running CQ instance. Dispatcher Logs could also help you to fix it.

Sham_HC
Level 10
October 16, 2015

In front of web servers you should have load balancer to check the healthiness & redirect to available one.  Or at dispatcher you need to configure cross reference. 

Level 4
October 16, 2015

Hi,

Both the web servers have different Apache configurations and have renders pointing to their individual servers for clustering.

Regards,

Shallu Rohilla

Level 6
October 16, 2015

So Dispatcher1 is connected to Publish1 and Dispatcher2 is connected to Publish2 with no cross reference. A pure Left and Right setup.

And then you have a load balancer in front of the Dispatchers and turn of the Publish1 but Dispatcher 1 is still serving content. Did I get you right?

/Ove

Level 4
October 16, 2015

Thats correct.

joerghoh
Adobe Employee
joerghohAdobe EmployeeAccepted solution
Adobe Employee
October 16, 2015

Then configure your loadbalancer not to check the healthstate of the webserver/dispatcher, but to check the health of your CQ instance sitting behind the webserver/dispatcher. The probably most easiest way is to append a "?" at the end of the healthcheck URL.