A virtual waiting room is a web page in which buyers wait in an online queue, when far more people want tickets than a ticketing system can serve at one time. It is also called an online queue or a virtual queue. It lets buyers into the ticket shop a few thousand at a time, in order. While they wait, it also checks for bots, programs that try to buy tickets in bulk so that touts can resell them at a higher price.
Why is it needed? A ticketing system can serve only a certain number of buyers at once. Each buyer loads seat maps, holds seats and pays, and each of these steps uses the system's computers and its database. When tickets for a famous concert go on sale at noon, millions of people may arrive in the same minute. This is a problem of high concurrency. Without a waiting room, the whole site would slow down or crash, and nobody would get a ticket.
The waiting room is a simple page that is cheap to serve. It is usually served from a content delivery network (CDN), whose servers in many cities can show the same page to millions of people. Each arriving buyer gets a place in the queue. When that buyer's turn comes, the waiting room gives the buyer's browser a signed pass, a small piece of data that the ticket shop can check but that nobody can forge. The pass lets the buyer into the shop for a limited time.
Imagine a concert by a famous international band at a stadium in Mumbai, with 50,000 tickets. Sales open at 12 noon, and 12 lakh people are on the waiting page by then. Everyone who arrived before noon is given a random place in the queue, so that arriving at 11:30 gives no advantage over arriving at 11:59. The waiting room lets in 5,000 buyers every minute. Each buyer gets a pass that is valid for 10 minutes, and may buy at most 4 tickets. Once inside, buyers choose seats, which are then held as described in seat selection and reservation. By about 12:45, all 50,000 tickets are sold. The waiting room then tells everyone still in the queue that the concert is sold out.
Meanwhile, bot management software checks each buyer in the queue. It looks for signs of a program rather than a person, such as hundreds of requests from the same group of computers, or a browser that behaves too fast to be human. Suspected bots are removed from the queue, or asked to solve a challenge. Limits of 4 tickets for each account and each phone number make bulk buying harder too.
The online ticketing platform decides how fast to let people in, depending on how many buyers its systems can serve. Some platforms are also exploring dynamic pricing for such sales, which is a separate question from the queue itself.