Guest

Preview Tool

Cisco Bug: CSCup38079 - Nexus6000:ECC error chk for BdStatsTable cause Hardware Failure on ports

Last Modified

Nov 06, 2020

Products (2)

  • CiscoPro Workgroup EtherSwitch Software
  • CiscoPro Workgroup EtherSwitch Software

Known Affected Releases

6.0(2)N2(3) 6.0(2)N2(4)

Description (partial)

Symptom:
In Nexus 6000 switches running 6.0(2)N2(3) or higher, certain ports belonging to a ASIC will be marked as hardware failure. Ports in either group of 3(if interfaces are in 40G mode) or in 12(if interfaces are in 10g) mode will be brought down. A message such as following is seen

2014 May 20 06:52:04 6004-A %$ VDC-1 %$ %NOHMS-2-NOHMS_DIAG_ERROR: Module 3: Runtime diag detected major event: Forwarding ASIC failure: Ethernet3/1 Ethernet3/2 Ethernet3/3 

<snip>
Eth2/12/4     1       eth  access down    SFP not inserted            10G(D) --
Eth3/1        --      eth  routed down    Hardware failure            40G(D) -- <<<---
Eth3/2        --      eth  routed down    Hardware failure            40G(D) -- <<<---
Eth3/3        1       eth  access down    Hardware failure            40G(D) -- <<<---
Eth3/4        --      eth  routed up      none                        40G(D) --
<snip>

Conditions:
Seen in normal operation. If you are running into this bug, the ASIC will be brought down due to incorrect ECC check handling. This can be verified with following command

6004-A# show hardware internal bigsur event-history errors
<snip>
19) Event:E_DEBUG, length:35, at 532298 usecs after Tue May 20 06:52:04 2014
    [100] EDMA - DMA fialed for asic: 0



20) Event:E_DEBUG, length:93, at 480405 usecs after Tue May 20 06:52:04 2014
    [100] bigsur_handle_ecc_intr(): slot 2 asic 0 set to faulty, 2bit ECC detected int_src_id 956



21) Event:E_DEBUG, length:95, at 480199 usecs after Tue May 20 06:52:04 2014
    [100] ECC Failure: slot 2 asic 0 [src=956, addr = 0x01000800, flags = 1][single_bit? NO][loc=0]



22) Event:E_DEBUG, length:96, at 884740 usecs after Tue May 20 06:51:48 2014
    [100] ECC Failure: slot 2 asic 0 [src=956, addr = 0x01000820, flags = 7][single_bit? YES][loc=0]
<snip>

Related Community Discussions

&lt;保留&gt; Nexus 6000 シリーズ、特定のASICにてパケットロスが発生する事象
2017年x月x日(初版) TAC SR Collection 主な問題 Nexus 6000 シリーズにおいて、特定のASICに所属するポートでパケットロスが発生する場合があります。 原因 本問題は CSCuj27098 にて報告されており、特定ASIC上で Parity error (single bit error) が検知された場合にError修正が行われるべきですが、これが正しく動作しません。結果、該当のASIC上で処理されるパケットが破棄される場合があります。 ※ASIC に所属するポートは以下のコマンドで確認できます。 N6000# show hardware internal bigsur all-ports Bigsur Port Info: Port     |asic|inst|inst| name     |idx |slot|asic|eport|logi|flag|adm|opr|if_index|diag|ucVer ---------+----+----+----+-----+----+----+---+---+--------+----+----- ...
Latest activity: Jan 16, 2017
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.