Guest

Preview Tool

Cisco Bug: CSCun38334 - network flap lead to cache no capacity

Last Modified

Jun 21, 2020

Products (2)

  • Cisco Videoscape Distribution Suite for Television
  • Cisco TV Streamer Application

Known Affected Releases

3.0(3) 3.0(3)ES1

Description (partial)

Symptom:
The issue observed here was after few days the CGW's on the customer start reporting 503 error on its Http layer which resulted in streams getting dropped off

Conditions:
the issue was with the 3rd Party Streamers like Enrich/Seachange/MX which start requesting the C2 Transfer after the TransferId Timeout which is off 500ms and CGW's responded with a TCP GONE. And within a short span of seconds the streamers from the same source ip and port combination start requesting another Transfer Request which resulted in CGW to not respond for those transfers and unable to clear the bandwidth held for the second transfer.
cserver TCP stack assumes the source ip&port combination can be reused with a 10seconds delay(with a tunable tcp_msl_seconds) which reset the TCP session and clears the flags. As the second request was came in ~4seconds which resulted in bandwidth over utilization
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)
  • Status
  • Severity
  • Known Fixed Releases
  • Related Community Discussions
  • Number of Related Support Cases
Bug information is viewable for customers and partners who have a service contract. Registered users can view up to 200 bugs per month without a service contract.