Guest

Preview Tool

Cisco Bug: CSCsz25424 - Ingress LC crashes while sending IPv6 host not reachable packets to XBMA

Last Modified

Sep 23, 2017

Products (1)

  • Cisco Carrier Routing System

Known Affected Releases

3.9.0.BASE

Description (partial)

Symptom:

Problem Symptom(s):
=================
The issue can present itself via several potential symptoms.
On any given system, these symptoms may be observed individually, or may be observed in combinations.

	1)	A line card may reset with the following message:

				rx_xbma[77]: %L2-QM_E5-4-INTERRUPT :  INGRESSQ: LQEQ  error : reg value 0x2

	2)	A line card may reset with the following message:

				rx_xbma[77]: %L2-QM_E5-4-INTERRUPT :  INGRESSQ: SRAM error : reg value 0x18 

		Please note that the "reg value" can vary (0x18, 0x11, 0x10, 0xa, 0x2) within this message.

	3)	A SPA may be forcefully reset due to a Heartbeat failure via a sequence similar to the following:

		%L2-SPA-3-PLIM_HEARTBEAT_ERR : SPA-1XCHOC12/DS0: bay 1 heartbeat failed, expected seq# 4563883 received
seq# 4563882, Time since last message 31s
		%L2-SPA-3-PLIM_HEARTBEAT_ERR : SPA-1XCHOC12/DS0: bay 1 heartbeat failed, expected seq# 4563883 received
seq# 4563882, Time since last message 32s
		%L2-SPA-3-PLIM_HEARTBEAT_ERR : SPA-1XCHOC12/DS0: bay 1 heartbeat failed, expected seq# 4563883 received
seq# 4563882, Time since last message 33s
		%L2-SPA-3-PLIM_HEARTBEAT_ERR : SPA-1XCHOC12/DS0: bay 1 heartbeat failed, expected seq# 4563883 received
seq# 4563882, Time since last message 34s
		%L2-SPA-3-PLIM_HEARTBEAT_ERR : SPA-1XCHOC12/DS0: bay 1 heartbeat failed, expected seq# 4563883 received
seq# 4563882, Time since last message 35s
		%L2-SPA-3-PLIM_HEARTBEAT_ERR : SPA-1XCHOC12/DS0: bay 1 heartbeat failed, expected seq# 4563883 received
seq# 4563882, Time since last message 36s
		%L2-SPA-3-PLIM_HEARTBEAT_ERR : SPA-1XCHOC12/DS0: bay 1 heartbeat failed, expected seq# 4563883 received
seq# 4563882, Time since last message 37s
		%L2-SPA-3-PLIM_HEARTBEAT_ERR : SPA-1XCHOC12/DS0: bay 1 heartbeat failed, expected seq# 4563883 received
seq# 4563882, Time since last message 38s
		%L2-SPA-3-PLIM_HEARTBEAT_ERR : SPA-1XCHOC12/DS0: bay 1 heartbeat failed, expected seq# 4563883 received
seq# 4563882, Time since last message 39s
		%L2-SPA-3-PLIM_HEARTBEAT_ERR : SPA-1XCHOC12/DS0: bay 1 heartbeat failed, expected seq# 4563883 received
seq# 4563882, Time since last message 40s
		%L2-SPA-5-PLIM_RELOADING : Reloading SPA-1XCHOC12/DS0 in subslot 1 after heartbeat failure ...


Warning Symptom(s):
=================
In addition to the primary symptoms noted above, there are potential warning indicators that may be
exhibited.
Please note that these warning symptoms do not result in the immediate reset of any line card and/or SPA. 
However, if present, it is an indication that the underlying condition may be occurring, 
and that the system is therefore exposed to any one of the primary symptoms noted above.

	w-i)	A message is received of the following form:

			mq_send failed: Message too long  : pkg/bin/g_spa_: (PID=xxxxx) : 
				

	w-ii)	Counters on a SPA interface discontinue functioning.

		Repeated invocations of "show interface <>"  while traffic is flowing
		will show that the counters have stopped incrementing.


	w-iii)	Heartbeat failure messages appear of short-term duration (0s or 1s)

		%L2-SPA-3-PLIM_HEARTBEAT_ERR : SPA-1XCHOC12/DS0: bay 1 heartbeat failed, expected seq# 3175920 received
seq# 2863311530, Time since last message 1s

		Please note that this warning symptom is very similar to the primary symptom noted above.
		The distinguishing characteristic is the cumulative length of the "Time since last message".

		If the length is of a short duration (0, 1, or 2 seconds), then the system may recover without
reloading the SPA.
		If this is the case, then the message should be viewed as an indication that the issue *may* be occurring,
		and if it repeats in the future, could result in the reset of the affected SPA.

		If the length is of a long duration ( > 30 seconds), then it is an indication that the issue *has*
occurred,
		and that a forceful reset of the SPA is imminent.




Conditions:



The issue is related to Engine-5 line cards only (SIP-401, SIP-501, SIP-601).
The issue is independent of the number and combination of SPAs that are installed on the SIP.
The issue is triggered by the combination and timing of ingress slow-path traffic containing specific
characteristics.
The suspect traffic can originate from any SPA attached to the affected line card.
The issue may impact that same SPA, or any other SPA attached to the line card.
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.