Cisco Bug: CSCvt31934 - AA VNF ownership transfer can happen independent of local/geo switchover
May 31, 2020
- Cisco Elastic Services Controller
Known Affected Releases
Symptom: ESC's VNF load sharing (ownership) manager will transfer all VNFs managed by a cluster node to a new node on detection of a critical service fault. Once VNF ownership begins the process completes until all VNFs are transferred. Under high load conditions it is possible for a spurious node fault to occur and clear quickly without causing a switchover. Since ownership transfer can not be terminated once started, a condition can latch where both the initial and new cluster elements are actively managing a VNF. This dual ownership syndrome can exhibit the following behaviours on new NB management requests against the VNF: 1> A new request can result in FAILED_TEMP (ETSI) as ownership transfer is in process (recoverable). 2> A new request is processed by both ESC nodes causing the operation to fail and loss of downstream management ability. Conditions: ESC is under high load or is resource constrained causing a spurious service health check failure.
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