Guest

Preview Tool

Cisco Bug: CSCun19289 - FI mgmt0 dropping traffic coming from blades behind that FI

Last Modified

Nov 27, 2020

Products (1)

  • Cisco Unified Computing System

Known Affected Releases

2.2(1b)B 2.2(1c)C

Description (partial)

Symptom:
The behavior of the mgmt0 interface is changed in 2.2.

The following results are seen using 2.2(1b) and Windows 2012R2, where the vnic was set to prefer a specific Fabric. Fabric Failover enabled.

- A blade connected to Fabric B, sending a ping to mgmt0 of FI-B (same VLAN as blade) -> FAIL
- A blade connected to Fabric B, sending a ping to mgmt0 of FI-A (same VLAN as blade) -> OK
- A blade connected to Fabric A, sending a ping to mgmt0 of FI-B (same VLAN as blade) -> OK
- A blade connected to Fabric A, sending a ping to mgmt0 of FI-A (same VLAN as blade) -> FAIL

This has important negative consequences as depending on which fabric the traffic is sourced, the FI's virtual IP might be unavailable. This is particularly bad when you want to use UCS Director or any other software on the blades that wants to talk to the FI's mangement IP.

This behavior is not seen on <= 2.1.

Seen at a customer deploying UCS Director and reproduced in TAC lab.

Conditions:
First noticed on 2.2(1b) when blade IP is in same VLAN as the mgmt0 interface
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.