Guest

Preview Tool

Cisco Bug: CSCur86643 - Seeing Amba MBE error during MFG SNT test

Last Modified

Jan 29, 2017

Products (3)

  • Cisco Network Convergence System 6000 Series Routers
  • Cisco NCS 6008 - 8-Slot Chassis
  • Cisco IOS XR Software

Known Affected Releases

5.0.1.LC

Description (partial)

Symptom:
DDR - MBE errors seen on Line Cards (reported on fia).  
When this error count exceeds the threshold, ASIC goes through PON reset and if the errors continue to occur, after 5 PON resets of the ASIC, board gets reloaded.  

Error messages on console:
----------------------------------------
LC/0/6/CPU0:Oct  9 17:24:30.306 : fia_driver[244]: %PLATFORM-CIH-3-ASIC_ERROR_SPECIAL_HANDLE_THRESH : fia[1]: A mbe error has occurred causing  packet drop transient. CMIC.CMIC_CMC0_IRQ_STAT3.MMU.Interrupt_Register.DramOppCrcErrInt  Threshold has been exceeded ^M^M
LC/0/7/CPU0:Oct  9 17:24:30.306 : fia_driver[255]: %PLATFORM-CIH-3-ASIC_ERROR_SPECIAL_HANDLE_THRESH : fia[3]: A mbe error has occurred causing  packet drop transient. CMIC.CMIC_CMC0_IRQ_STAT3.MMU.Interrupt_Register.DramOppCrcErrInt  Threshold has been exceeded ^M^M
LC/0/7/CPU0:Oct  9 17:24:30.442 : npu_driver[176]: %PLATFORM-NPU-3-SW_ERROR : Declared fault to FM, Interlaken DOWN for NPU Instance 3. Bringing interfaces down ^M^M
LC/0/6/CPU0:Oct  9 17:24:30.443 : npu_driver[189]: %PLATFORM-NPU-3-SW_ERROR : Declared fault to FM, Interlaken DOWN for NPU Instance 1. Bringing interfaces down ^M^M
LC/0/7/CPU0:Oct  9 17:24:30.443 : npu_driver[176]: %L2-PLIM_ETHER-2-RX_LF : Interface HundredGigE0_7_CPU0_7_0, Detected Local Fault ^M^M
LC/0/6/CPU0:Oct  9 17:24:30.443 : npu_driver[189]: %L2-PLIM_ETHER-2-RX_LF : Interface HundredGigE0_6_CPU0_2_0, Detected Local Fault ^M^M
LC/0/6/CPU0:Oct  9 17:24:30.443 : npu_driver[189]: %L2-PLIM_ETHER-2-RX_LF : Interface HundredGigE0_6_CPU0_3_0, Detected Local Fault ^M^M
LC/0/7/CPU0:Oct  9 17:24:30.443 : npu_driver[176]: %L2-PLIM_ETHER-2-RX_LF : Interface HundredGigE0_7_CPU0_6_0, Detected Local Fault ^M^M
LC/0/7/CPU0:Oct  9 17:24:30.444 : ifmgr[288]: %PKT_INFRA-LINK-3-UPDOWN : Interface HundredGigE0/7/0/7, changed state to Down ^M^M
LC/0/7/CPU0:Oct  9 17:24:30.444 : ifmgr[288]: %PKT_INFRA-LINK-3-UPDOWN : Interface HundredGigE0/7/0/6, changed state to Down ^M^M
LC/0/6/CPU0:Oct  9 17:24:30.444 : ifmgr[185]: %PKT_INFRA-LINK-3-UPDOWN : Interface HundredGigE0/6/0/2, changed state to Down ^M^M
LC/0/6/CPU0:Oct  9 17:24:30.445 : ifmgr[185]: %PKT_INFRA-LINK-3-UPDOWN : Interface HundredGigE0/6/0/3, changed state to Down ^M^M
LC/0/6/CPU0:Oct  9 17:24:31.516 : fia_driver[244]: %PLATFORM-CIH-2-ASIC_ERROR_PON_RESET : fia[1]: A mbe error has occurred causing  packet drop transient. CMIC.CMIC_CMC0_IRQ_STAT4.IPT.Interrupt_Register.CrcErrPkt  Threshold has been exceeded ^M^M
LC/0/7/CPU0:Oct  9 17:24:31.521 : fia_driver[255]: %PLATFORM-CIH-2-ASIC_ERROR_PON_RESET : fia[3]: A mbe error has occurred causing  packet drop transient. CMIC.CMIC_CMC0_IRQ_STAT4.IPT.Interrupt_Register.CrcErrPkt  Threshold has been exceeded ^M^M
LC/0/6/CPU0:Oct  9 17:25:49.495 : ifmgr[185]: %PKT_INFRA-LINK-3-UPDOWN : Interface HundredGigE0/6/0/2, changed state to Up ^M^M
LC/0/6/CPU0:Oct  9 17:25:49.497 : ifmgr[185]: %PKT_INFRA-LINK-3-UPDOWN : Interface HundredGigE0/6/0/3, changed state to Up ^M^M
LC/0/7/CPU0:Oct  9 17:25:50.070 : ifmgr[288]: %PKT_INFRA-LINK-3-UPDOWN : Interface HundredGigE0/7/0/6, changed state to Up ^M^M
LC/0/6/CPU0:Oct  9 17:25:50.302 : fia_driver[244]: %PLATFORM-CIH-3-ASIC_ERROR_SPECIAL_HANDLE_THRESH : fia[1]: A mbe error has occurred causing  packet drop transient. CMIC.CMIC_CMC0_IRQ_STAT3.MMU.Interrupt_Register.DramOppCrcErrInt  Threshold has been exceeded ^M^M
LC/0/6/CPU0:Oct  9 17:25:50.439 : npu_driver[189]: %PLATFORM-NPU-3-SW_ERROR : Declared fault to FM, Interlaken DOWN for NPU Instance 1. Bringing interfaces down ^M^M
LC/0/6/CPU0:Oct  9 17:25:50.440 : npu_driver[189]: %L2-PLIM_ETHER-2-RX_LF : Interface HundredGigE0_6_CPU0_2_0, Detected Local Fault ^M^M
LC/0/6/CPU0:Oct  9 17:25:50.440 : npu_driver[189]: %L2-PLIM_ETHER-2-RX_LF : Interface HundredGigE0_6_CPU0_3_0, Detected Local Fault ^M^M
LC/0/6/CPU0:Oct  9 17:25:50.440 : ifmgr[185]: %PKT_INFRA-LINK-3-UPDOWN : Interface HundredGigE0/6/0/2, changed state to Down ^M^M

Conditions:
When traffic is sent across Amba, sometimes we may see MBE errors on the Ingress Amba (causing traffic drop).
This is due to a flaw in the DDR tuning code which results in marginal/non-optimal tuning values causing the failures on DDR. This impacts the traffic flowing through the Amba that is reporting these errors.
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.