This counter counts how many lock requests could not be satisfied right away and had to queue and wait for the resource to become available. This is one of the clearest direct indicators of lock contention severity, since it counts actual delays experienced by requesting processes. For example, if this counter rises sharply during a specific batch window, the admin can correlate that time period with whichever job is holding contested locks the longest.