Article ID: 13353
Products
SecureStack C2, firmware 3.00.50 through 5.01.03.0007
SecureStack B2, firmware 2.00.16 through 4.01.03.0007
Symptoms
The switch reboots.
The current.log (
5487) contains a "NIM: Timeout event(3)" item, which will be followed by a "Start of Code" item, then with later firmware will be followed by a "Last switch reset caused by power cycle" item.
Here are separate examples taken from different C2 units... <57> MAY 31 17:06:41 STK1 BOOT[238395136]: nim_events.c(184) 177 %%
NIM: Timeout event(3), intIfNum(1) remainingMask = 0x00000000 0x00040000 0x00000000
<57> MAY 31 17:08:58 STK1 BOOT[268434944]: bootos.c(424) 178 %%
Start of Code - Build:03.00.52 Date:Tue May 10 18:02:44 2005
<160> FEB 07 15:24:46 0.0.0.0-1 NIM[237596088]: nim_events.c(180) 23279 %%
NIM: Timeout event(3), intIfNum(8) remainingMask = 0x00000000 0x00040000 0x00000000
<57> FEB 07 15:26:54 STK1 BOOT[268434944]: bootos.c(491) 61 %%
Start of Code - Build:03.01.91 Date:Thu Nov 17 16:54:09 2005
<110> FEB 07 15:27:57 STK1 BOOT[223494312]: edb_bxs_api.c(305) 62 %%
Last switch reset caused by power cycle
<160> AUG 15 13:04:50 10.10.0.102-1 NIM[232463848]: nim_events.c(187) 145 %%
NIM: Timeout event(3), intIfNum(201) remainingMask = 0x00000800 0x00000000 0x00000000
<57> AUG 15 13:54:59 STK1 BOOT[268434944]: bootos.c(504) 90 %%
Start of Code - Build:03.03.33 Date:Tue Apr 4 11:03:56 2006
<110> AUG 15 13:55:20 STK1 BOOT[217972848]: edb_bxs.c(670) 91 %%
Last switch reset caused by power cycle
<160> JUN 28 15:56:12 10.17.4.2-1 NIM[219512128]: nim_events.c(187) 68 %%
NIM: Timeout event(3), intIfNum(24) remainingMask = 0x00000000 0x00000000 0x00040000 0x00000400
<57> JUN 28 15:56:09 2007 STK1 BOOT[268430848]: bootos.c(652) 59 %%
Start of Code - Build:05.00.69 Date:Thu Jun 14 06:06:03 2007
<110> JUN 28 15:57:11 2007 STK1 BOOT[191646280]: edb_bxs_api.c(353) 60 %%
Last switch reset caused by power cycle
<160>Feb 11 17:24:08 10.10.0.102-1 NIM[200312408]: nim_events.c(187) 1365 %%
NIM: Timeout event(3), intIfNum(109) remainingMask = 0x00000000 0x02000000 0x00044000 0x00000000
<57> FEB 11 17:24:10 2008 STK1 BOOT[268430848]: bootos.c(689) 107 %%
Start of Code - Build:05.01.01.0047 Date:Mon Jan 28 16:16:04 2008
<110> FEB 11 17:24:32 2008 STK1 BOOT[177738192]: edb_bxs.c(761) 108 %%
Last switch reset caused by power cycle
Here are separate examples taken from different B2 units... <160> MAR 07 20:00:49 0.0.0.0-1 NIM[238614256]: nim_events.c(187) 37 %%
NIM: Timeout event(3), intIfNum(5) remainingMask = 0x00000000 0x00040000 0x00000000
<57> MAR 07 20:02:57 STK1 BOOT[268434944]: bootos.c(506) 38 %%
Start of Code - Build:02.00.20 Date:Fri Jan 13 10:09:01 2006
<110> MAR 07 20:03:59 STK1 BOOT[224186160]: edb_bxs_api.c(305) 39 %%
Last switch reset caused by power cycle
<160> SEP 13 20:17:15 10.10.100.161-1 NIM[223607472]: nim_events.c(187) 63 %%
NIM: Timeout event(3), intIfNum(24) remainingMask = 0x00000000 0x00000000 0x00040000 0x00000000
<57> SEP 13 20:17:06 2010 STK1 BOOT[268430848]: bootos.c(640) 25 %%
Start of Code - Build:04.00.26 Date:Wed Jan 17 18:27:37 2007
<110> SEP 13 20:18:08 2010 STK1 BOOT[195755392]: edb_bxs_api.c(353) 26 %%
Last switch reset caused by power cycle
<160> MAR 04 12:44:26 10.12.229.243-1 NIM[208914392]: nim_events.c(187) 61 %%
NIM: Timeout event(3), intIfNum(16) remainingMask = 0x00000000 0x02000000 0x00040000 0x00000400
<57> MAR 04 12:42:05 2008 STK1 BOOT[268430848]: bootos.c(680) 59 %%
Start of Code - Build:04.01.01.0040 Date:Tue Dec 11 15:45:01 2007
<110> MAR 04 12:43:08 2008 STK1 BOOT[189289584]: edb_bxs_api.c(353) 60 %%
Last switch reset caused by power cycle
Cause
This is a port Link Down event which is due to some internal application(s) (i.e. 802.1x/DOT1X, IGMP Snooping/IGMP_SNOOPING, LAG/DOT3AD, LLDP/LLDP, MAC Authentication/MACA, MAC Locking/PORT_MACLOCK, MSTP/DOT1S, OSPF/OSPF, SFLOW/SFLOW, VRRP/VRRP, etc; as coded within the remainingMask values) not responding to port (see
12864 for intIfNum translation) communication.
There are several timeout event versions, but the "event(3)" variety is immediately followed by a system reset ("Start of Code"). Upon subsequent bootup, diagnostics misleadingly indicate that the reset occurred due to a power cycle (
5766).
Solution
For the C2, upgrade to firmware 5.01.04.0001 or higher.
For the B2, upgrade to firmware 4.01.04.0001 or higher.
Release notes state, in the 'Firmware Changes and Enhancements' section:
code:Corrected an erroneous interface message timeout reset (NIM timeout event) caused during management changes of complex interface configurations.
The NIM timeout interval has been increased from 600 to 1200 seconds,
and any messages that may remain will be in an enhanced form.
Some Spanning Tree examples...
code:NIM: Timeout event(L7_PORT_ACTIVE) on unit(1) slot(0) port(17) (intIfNum(17)) for components(DOT1S)
means that unit 1's port 17 failed to go ACTIVE in the DOT1S (MSTP) component.
code:NIM: Timeout event(FORWARDING) on unit(2) slot(0) port(40)(intIfNum(92)) for components(DOT1S)
means that unit 2's port 40 failed to go FORWARDING in the DOT1S (MSTP) component.
code:NIM: Timeout event(NOT_FORWARDING) on unit(1) slot(0) port(48)(intIfNum(48)) for components(DOT1S)
means that unit 1's port 48 failed to go NOT_FORWARDING in the DOT1S (MSTP) component.
code:NIM: Timeout event(UP) on unit(3) slot(0) port(21)(intIfNum(125)) for components(DOT1S)
means that unit 3's port 21 failed to go UP in the DOT1S (MSTP) component.
code:NIM: Timeout event(DOWN) on unit(3) slot(0) port(23)(intIfNum(127)) for components(DOT1S)
means that unit 3's port 23 failed to go DOWN in the DOT1S (MSTP) component.
If this is accompanied by more concrete behavioral symptoms, please
contact the GTAC for assistance.