Add recommended sysctl and etcd defaults to the Talos machine configuration.

Proposed defaults

machine:
  sysctls:
    # TCP orphan handling
    net.ipv4.tcp_orphan_retries: "3"
    net.ipv4.tcp_fin_timeout: "30"

    # Network backlog
    net.core.netdev_max_backlog: "5000"
    net.core.netdev_budget: "600"
    net.core.netdev_budget_usecs: "8000"

    # TCP keepalive (early detection of dead connections)
    net.ipv4.tcp_keepalive_time: "600"
    net.ipv4.tcp_keepalive_intvl: "10"
    net.ipv4.tcp_keepalive_probes: "6"

  etcd:
    extraArgs:
      quota-backend-bytes: "8589934592"
      max-request-bytes: "10485760"

Why

• tcp_orphan_retries=3: Talos default is 0, meaning a single lost ACK causes orphaned TCP connections to accumulate indefinitely. With DRBD creating many peer connections, this leads to port exhaustion.
• tcp_fin_timeout=30: Reduces TIME_WAIT duration for faster port recycling.
• netdev_max_backlog/budget: Prevents packet loss under high network load from DRBD replication traffic.
• tcp_keepalive: Enables faster detection of dead DRBD peer connections (10min idle + 60s probe window instead of default 2h+75s).
• etcd quota-backend-bytes=8Gi: Prevents etcd running out of space with large LINSTOR CRD datasets.
• etcd max-request-bytes=10Mi: Allows larger CRD objects to be stored.