15 — High Availability

flowchart TB
    subgraph Region1["Region: us-east-1"]
        subgraph AZ1["AZ 1"]
            WEB1["Web Server A"]
            DB_PRIMARY["RDS Primary<br/>(Read/Write)"]
        end
        subgraph AZ2["AZ 2"]
            WEB2["Web Server B"]
            DB_STANDBY["RDS Standby<br/>(Failover)"]
        end
        subgraph AZ3["AZ 3"]
            WEB3["Web Server C"]
        end

        ALB["Application Load Balancer"] --> WEB1 & WEB2 & WEB3
        DB_PRIMARY <-->|"Synchronous Replication"| DB_STANDBY
    end

    Users["🌍 Users"] --> Route53["Route53<br/>DNS + Health Checks"]
    Route53 --> ALB

    style Region1 fill:#7c3aed,color:#fff
    style AZ1 fill:#3b82f6,color:#fff
    style AZ2 fill:#059669,color:#fff
    style AZ3 fill:#f59e0b,color:#fff
    style ALB fill:#6366f1,color:#fff

flowchart TB
    subgraph ActivePassive["Active-Passive"]
        AP_LB["Load Balancer"]
        AP_Active["🟢 Active<br/>Handles all traffic"]
        AP_Passive["🔴 Passive (Standby)<br/>Synced, not serving"]
        AP_LB --> AP_Active
        AP_Active -.->|Failover| AP_Passive
    end

    subgraph ActiveActive["Active-Active"]
        AA_LB["Load Balancer"]
        AA_Node1["🟢 Node 1<br/>50% traffic"]
        AA_Node2["🟢 Node 2<br/>50% traffic"]
        AA_LB --> AA_Node1 & AA_Node2
    end

    style ActivePassive fill:#3b82f6,color:#fff
    style ActiveActive fill:#059669,color:#fff
    style AP_LB fill:#7c3aed,color:#fff
    style AA_LB fill:#7c3aed,color:#fff

flowchart TB
    Stateless["Stateless Service<br/>Web / API"] --> SLB["Load Balancer"]
    SLB --> S1["Web Server 1"]
    SLB --> S2["Web Server 2"]
    SLB --> S3["Web Server 3"]
    S1 & S2 & S3 --> SharedDB["Shared DB / Cache"]

    Stateful["Stateful Service<br/>Database"] --> Primary["Primary<br/>Read/Write"]
    Primary -->|Replication| Replica1["Replica 1<br/>Read-only"]
    Primary -->|Replication| Replica2["Replica 2<br/>Read-only"]
    Primary -.->|Auto Failover| Replica1

    style Stateless fill:#3b82f6,color:#fff
    style Stateful fill:#f59e0b,color:#fff
    style SharedDB fill:#059669,color:#fff
    style Primary fill:#ef4444,color:#fff

flowchart TB
    Design["Design for Failure"] --> Eliminate["Eliminate Single Points<br/>of Failure (SPOF)"]
    Design --> Redundancy["Add Redundancy<br/>N+1, N+2, multi-AZ"]
    Design --> Graceful["Graceful Degradation<br/>Degrade features, don't crash"]
    Design --> Isolation["Fault Isolation<br/>Bulkheads, circuit breakers"]

    Eliminate --> Examples1["Multiple servers<br/>Multiple AZs<br/>Multiple regions"]
    Redundancy --> Examples2["Standby DBs<br/>Replica caches<br/>Spare capacity"]
    Graceful --> Examples3["Read-only mode<br/>Serving stale cache<br/>Showing error page vs crashing"]
    Isolation --> Examples4["Service per pod<br/>Bounded queues<br/>Separate thread pools"]

    style Design fill:#7c3aed,color:#fff

flowchart LR
    Incident["💥 Incident<br/>System goes down"] --> RTO_Period["⏱️ RTO Period<br/>Time to restore service"]
    RTO_Period --> Restored["✅ Restored<br/>System back online"]

    LastBackup["💾 Last backup<br/>(RPO point)"] --> DataLoss["❌ Data lost<br/>(between backup and incident)"]
    DataLoss --> Incident

    style Incident fill:#ef4444,color:#fff
    style RTO_Period fill:#f59e0b,color:#fff
    style Restored fill:#059669,color:#fff
    style LastBackup fill:#3b82f6,color:#fff