Failure points analysis in the system

The diagram presents the main elements of the infrastructure: API gateway (load balancer, authentication, TLS, request limiting), TinyUrl service with POST and GET methods, cache layer (LRU), distributed SQL shards with replicas, external short link generator implementing real-lock, distributed journal/queue, and an analytical subsystem.

It is necessary to identify critical points of failure:

• Determine which components are single points of failure (SPOF).
• Suggest ways to improve fault tolerance: replication, automatic restart, retry strategies, rollback mechanisms.
• Describe system behavior when a generator, a shard or its replica, cache layer, queue, or analytical module service fails.
• Divide components into stateful and stateless.
• Formulate a list of monitoring tools and alert systems necessary for failure control.