Loading...
Loading...
ASR9xx router may unexpectedly reload after an MPLS pseudowire is brought up. Messages similar to the following may be observed in the crashinfo file: *Jan 10 23:57:35.891: %LDP-5-NBRCHG: LDP Neighbor x.x.x.x:0 (1) is UP *Jan 10 23:57:38.162: %LDP-5-NBRCHG: LDP Neighbor x.x.x.x:0 (2) is UP *Jan 10 23:57:39.321: %LDP-5-NBRCHG: LDP Neighbor x.x.x.x:0 (3) is UP *Jan 10 23:57:39.590: %XCONNECT-5-PW_STATUS: MPLS peer x.x.x.x vcid 30, VC state UP *Jan 10 23:57:40.507: %XCONNECT-5-PW_STATUS: MPLS peer x.x.x.x vcid 30, VC state DOWN, PW Err *Jan 10 23:57:45.787: %XCONNECT-5-PW_STATUS: MPLS peer x.x.x.x vcid 30, VC state DOWN *Jan 10 23:57:45.787: %XCONNECT-5-PW_STATUS: MPLS peer x.x.x.x vcid 30, VC state UP Exception to Fastpath Thread: Frame pointer 0x3010E510, PC = 0x0
ASR9xx router (e.g. ASR902, ASR903, ASR907, ASR920) in use. MPLS pseudowires are configured. There is a possibility for the crash to occur if the system processes a punted packet before initialization of the pseudowire has fully completed while it is transitioning to an "up" state.
No workaround.
This crash is caused by a timing issue. During initialization of the pseudowire, an interface descriptor block (IDB) in memory must be populated with various information about the pseudowire. There is a possibility that a punted packet could come to IOS for processing before the IDB is fully populated, resulting in an invalid memory access and crash.
Click on a version to see all relevant bugs
Cisco Integration
Learn more about where this data comes from
BugZero Plan
Streamline upgrades with automated vendor bug scrubs
BugZero Prevent
Wish you caught this bug sooner? Get proactive today.