Some Observations on Network Failures NANOG 15

1
Some Observations on Network FailuresNANOG 15

• Craig Labovitz
• ltlabovit_at_merit.edugt

2
Observations

• Goal Model Internet topological changes
• Lots of strange BGP routing
• Strange BGP routing went away
• What causes remaining BGP topological and policy
  changes?
• Not just count flaps, but study how routing
  tables changes over extended periods
• Not end-to-end

3
Internet Failures Analysis

• Look at default-free BGP announcements from
  multiple large providers
• Long lived (60 of 9 months)
• Consider stable if covered by less specifics
• 15 minute filter window
• Mean-time failure, repair and availability
• Case study regional network

4
What We Did

• Lots of probe machines
• Mae-East, Mae-West, Paix, PacBell, AADs
• A default-free collector at UM
• Routeviews Multi-hop EBGP 6 providers US, Canada,
  Europe and Japan (300,000 routes)
• Case study of regional backbone (OSPF, IBGP/BGP)
• 42 gigabytes and four years of logged routing
  packets

5
RouteTracker

• Peer with ISP routers
• Log all routing packets to disk
• Maintain statistics

6
MTBF
7
Route Fail-Over
8
MTTR
9
Availability
10
Default-Free Route Availability
11
(No Transcript)
12
Backbone MTR
13
Network Failures
Michnet Backbone Failures 11/97 - 11/98
14
Observations

• Internet significantly less availability than
  PSTN (99.99 )
• Low mean time to change

15
Next Steps

• Host other routeviews machines?
• Merit has several FreeBSD desktop boxes
• Looking for peers
• ipma-support_at_merit.edu

16
(No Transcript)