Load balancing web service by rejecting connections

A non-transitory computer readable storage medium storing instruction, the instruction when executed by a processor cause the processor to:

• receive, at a primary port of a first web server of a pool of web servers, a request from a client via a load balancer;
• responsive to a first request queue of the first web server being full, send, by the first web server, a rejection notification to the load balancer to allow the load balancer to quickly resend the rejected request to another web server in the pool of web servers;
• receive the request at the secondary port of a web server of a pool of web servers, the request sent to the secondary port of the web server by a load balancer in response to determining that the request has been rejected more than a threshold number of times from various web servers in the pool of web servers;
• and
• responsive to receiving the request at the secondary port and the request queue being full:
• place the request in a secondary queue of the first web server, and
• perform at least one of:
• (i) processing the secondary queue ahead of the first request queue, (ii) placing the secondary queue ahead of the first request queue, and (iii) responsive to a space opening in the first request queue, place the request from the secondary queue in the first request queue.

A system comprising:

• a processor;
• and
• a non-transitory compute readable storage medium, the instructions when executed by the processor, cause the processor to:
• receive, at a primary port of a first web server of a pool of web servers, a request from a client via a load balancer;
• responsive to a first request queue of the first web server being full, send, by the first web server, a rejection notification to the load balancer to allow the load balancer to quickly resend the rejected request to another web server in the pool of web servers;
• receive the request at the secondary port of a web server of a pool of web servers, the request sent to the secondary port of the web server by a load balancer in response to determining that the request has been rejected more than a threshold number of times from various web servers in the pool of web servers;
• and
• responsive to receiving the request at the secondary port and the request queue being full:
• place the request in a secondary queue of the first web server, and
• perform at least one of:
• (i) processing the secondary queue ahead of the first request queue, (ii) placing the secondary queue ahead of the first request queue, and (iii) responsive to a space opening in the request queue, place the request from the secondary queue in the first request queue.