Description

RPD core dump due to very high OSPF SFP runs (more than 350K) in a large scale of network in a rare case, it is a day-1 issue in all platform. RDP will restart automatically after crash. 

Symptoms

Large number of SPF runs might trigger this issue.

 

 

 

 

{master}

 

c798977@rdnvej160pee0344> show ospf overview

 

Instance: master

 

 Router ID: 144.133.144.135

 

 Route table index: 0

 

 LSA refresh time: 50 minutes

 

 Traffic engineering

 

 SPRING: Enabled

 

  SRGB Config Range :

 

   SRGB Start-Label : 600000, SRGB Index-Range : 65535

 

  SRGB Block Allocation: Success

 

   SRGB Start Index : 600000, SRGB Size : 65535, Label-Range: [ 600000, 665534 ]

 

  Node Segments: Enabled

 

  Ipv4 Index : 24701

 

 Post Convergence Backup: Enabled

 

  Max labels: 5, Max spf: 100, Max Ecmp Backup: 8

 

 DoNotAge uncapable

 

  Indication LSAs received: 1

 

 Area: 0.0.0.0

 

  Stub type: Not Stub

 

  Authentication Type: None

 

  Area border routers: 40, AS boundary routers: 85

 

  Neighbors

 

   Up (in full state): 2

 

  DoNotAge uncapable

 

   Indication LSAs received: 1

 

 Topology: default (ID 0)

 

  Prefix export count: 0

 

  Full SPF runs: 334760                  <<<<<<<<<<<<<

 

  SPF delay: 0.200000 sec, SPF holddown: 5 sec, SPF rapid runs: 3

 

  Backup SPF: Not Needed

Solution

The issue was resolved in PR/1739123, Juniper software engineer team identified the root cause and fix the programming reference count issue.

Modification History

1.0 Draft version

Related Information

PR/1739123