Allow long-polling by making the NLB timeout configurable, not hard-coded to 60 seconds.
Thank you for the feedback.
I’ve moved this to the ‘Cloud Services’ forum, since it applies to the network-level load balancer used with Cloud Services.
Please note that Traffic Manager provides a different kind of load-balancing…it’s a DNS-based system used to distributed traffic between different cloud services, typically deployed to different data centers.
Vasu Srinivasan commented
This must be configurable. A very serious limitation, it caused us several days of anxiety trying to figure out the problem.
this is a major issue that needs to be solved.
It has to be configurable!
This is a must have feature for longrunning jobs!
Paul Collins commented
Recently migrated to Azure and this issue is causing serious issues for our clients.
It has to be configurable
This is must!
This is must!
I need fix
John Falcon commented
The same problem. Weird, strange and unpleasant restriction. Hope that will be fixed soon.
Ivan Sokalskyi commented
In most of the cases it's not a problem, but sometimes it's just impossible. Let's say you can just send one HTTP POST request from some 3-rd party platform (commands are written on some script language) to your service for some data.
If you have control over your client, you can configure it to send TCP Keep-Alive's, which will keep the connection open indefinetely.
To send a TCP Keep-Alive every 50 seconds from C#:
webRequest.ServicePoint.SetTcpKeepAlive(true, 50 * 1000, 1000);