Cisco Bug: CSCua98465 - [BFD]:8k BFDv6 ses not come up for 20 min, when changing TTL threshold.
May 14, 2018
- Cisco ASR 9000 Series Aggregation Services Routers
Known Affected Releases
Symptom: On a box that has 8K BGP ipv6 sessions and these sessions have BFD multihop configured, updates to the protocol such as BFD multihop TTL may take long amount of time to take effect in the hardware. Conditions: The total amount of TCAM space required to program these 8K BGP IPv6 entries in LPTS is 16K. Now, LPTS implements an upper ceiling of 8000 for IPv4 entries and 1400 for IPv6 entries. As a result the remaining 10600 entries will remain in the LPTS retry queue and wont' be programmed in TCAM. These entries are retried constantly increasing the churn in the LPTS and TCAM , thus delaying the processing of any updates exponentially. With the fix the LPTS IFIB tcam entries can grow dynamically to the maximum tcam entries that can be programmed in the specific linecard hardware. The current hardware limits are: TYPHOON SE card: -96K 160-bit entries for IPv4 ( This is shared across all PRM applications such as IFIB, QOS, ACL etc) -16K 640-bit entries for IPv6 ( ditto) TYPHOON TR card: -24K 160-bit entries ( This is shared across all PRM applications such as IFIB, QOS, ACL etc) -4K 640-bit entries ( ditto ) So, with a Typhoon SE card we can have all 16K entries programmed and the problem doesn't occur. However, with a TR card the limit is still 4k and the problem will be seen.
Bug details contain sensitive information and therefore require a Cisco.com account to be viewed.
Bug Details Include
- Full Description (including symptoms, conditions and workarounds)
- Known Fixed Releases
- Related Community Discussions
- Number of Related Support Cases