Blog

Notes from building Osprey against a lab of 192 real routers running three different IGPs. Screenshots are taken live. Where a measurement comes from something the lab cannot hold, the post says where it came from instead.

Series 1: Trust

  1. Three IGPs, One Map

    What 192 routers, three carriers and three routing protocols look like on a single pane of glass
    • osprey
    • ospf
    • is-is
    • eigrp
    • topology
    • multi-tenant

  2. Zero Footprint

    How Osprey sees a network without becoming part of it
    • osprey
    • gre
    • bmp
    • snmp
    • ospf
    • is-is
    • passive-discovery

  3. Hop-by-Hop Truth

    Your path tool draws a corridor. Routers forward hop by hop. Here is the 4,032-pair measurement of the gap.
    • osprey
    • ospf
    • rfc2328
    • rfc3509
    • spf
    • bgp
    • path-analysis

  4. What Osprey Refuses to Guess

    The most important thing I built is a set of refusals
    • osprey
    • design
    • is-is
    • eigrp
    • bgp
    • honesty
    • observability

Series 2: Engineering the Truth

  1. The Same Router, Three Different Truths

    One physical box, several routing identities, and why a node-and-edge graph cannot hold them
    • osprey
    • ospf
    • ospfv3
    • data-model
    • abr
    • asbr
    • topology

  2. When the Network Disagrees With Itself

    Six sources of truth, one answer, and what to do when they contradict each other
    • osprey
    • evidence
    • provenance
    • lldp
    • bmp
    • snmp
    • bgp

  3. The Day My Topology Lied to Me

    An engineering postmortem on my own product, and the lab that caught it
    • osprey
    • postmortem
    • ospf
    • rfc2328
    • rfc3509
    • validation

  4. A Network Engineer's Unit Tests

    How do you test a topology engine when the network itself is the only oracle?
    • osprey
    • testing
    • fuzzing
    • validation
    • vendor-interop

Series 3: From Seeing to Reasoning

  1. What Happens If I Break This?

    A model you trust is only worth having if you can ask it questions it has not been asked before
    • osprey
    • simulation
    • what-if
    • spf
    • ospf
    • eigrp
    • resilience

  2. The Blast Radius

    Every link and every router tested at once, and why graph theory makes it take 13 milliseconds
    • osprey
    • resilience
    • spof
    • tarjan
    • graph-theory
    • dependency
    • bgp

  3. Estimating What You Cannot Measure

    A traffic matrix inferred from link counters, the flow collector I decided not to build, and the honest limits of my own lab
    • osprey
    • traffic
    • tomogravity
    • snmp
    • netflow
    • estimation
    • honesty

  4. The Network Has a Memory

    Reconstructing what the network believed at 14:02, and the difference between what changed and what you learned
    • osprey
    • time-travel
    • snapshots
    • bitemporal
    • bgp
    • forensics

Series 4: The View from Here

  1. The Table Is Not a Tree

    I wrote down three rules for drawing the AS graph before I measured one. A real full table killed all three.
    • osprey
    • bgp
    • as-path
    • visualisation
    • dfz
    • measurement
    • dn42

  2. The Biggest AS Doesn't Exist

    Two ways to rank a BGP table, both reasonable, sharing zero members in the top twelve
    • osprey
    • bgp
    • ranking
    • deaggregation
    • dfz
    • measurement

  3. Your Backbone Is Not the Internet's

    Five networks reach 94% of the origins in my table. Move one AS_PATH hop and it is five different networks.
    • osprey
    • bgp
    • vantage
    • transit
    • dfz
    • measurement

  4. What a Full Feed Actually Costs

    I measured a full table into the production schema instead of estimating it, and the measurement killed one of my own defaults
    • osprey
    • bgp
    • bmp
    • sizing
    • storage
    • churn
    • measurement

Series 5: A Guest in the Protocol

  1. The Router That Must Never Carry a Packet

    Three link-state protocols, three different ways to join a network and be useless to it on purpose
    • osprey
    • ospf
    • ospfv3
    • is-is
    • rfc5340
    • safety
    • passive-discovery

  2. The LSA You Cannot Take Back

    Joining a routing protocol is a handshake. Leaving one ends with a single message you may never get to send.
    • osprey
    • ospf
    • is-is
    • rfc2328
    • rfc5304
    • purge
    • safety

  3. The Protocol With No Database

    Three different kinds of nothing: no database to join, a database nobody implements, and a table with the answers but not the inputs
    • osprey
    • eigrp
    • is-is
    • snmp
    • mib
    • dual
    • refusals

  4. Push, Don't Poll

    The one source that comes to you, and the two things that change about what you are allowed to believe
    • osprey
    • bmp
    • bgp
    • rfc7854
    • rfc8671
    • freshness
    • refusals