Cisco Bug: CSCtb07649 - bgp nsr restarts in forced cleanup case when counter not reset to '0'
Jan 15, 2017
- Cisco 12000 Series Routers
Known Affected Releases
Symptoms: BGP may restart NSR while doing scoped-sync for a few neighbor (< 75) Condition: Recently we have added a safety mechanism whereby we do not try to do scoped sync on more than 75 neighbors at a time. This was necessiated by the fact that TCP was sending OPER_DOWN for all the neighbors in a session-set if the accounting timed out resulting in BGP trying to do scoped-sync for 1000s of neighbors for which it is not designed. However, the counter we used to check how many nbrs have been added to the scoped-sync ilist was getting reset only when the scoped sync timer fired. In case we do restart NSR when we exceeded the limit, the counter was not getting reset. As a result, if we had received OPER_DOWN for fewer neighbors next time, we were exceeding the limit and starting NSR again.
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