cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
41989
Views
86
Helpful
20
Replies

3650 PoE issue - Tstart error detected

andrewswanson
Level 7
Level 7

Hello

I am seeing an issue with PoE on a switch stack (WS-C3650-48FD-S running 03.06.05E).

Issue affects both non Cisco and Cisco devices. Not all switches in the stack are affected by this - each switch has a single PWR-C2-1025WAC PSU and the power consumption is low for the affected switch (652.6 W remaining with 122.4 W used)

The affected port(s) has the following status:

Gi7/0/24 auto faulty 0.0 n/a n/a 30.0

Cabling has been checked and issue can happen with long/short patch cables (as with bug CSCsw18530 https://bst.cloudapps.cisco.com/bugsearch/bug/CSCsw18530/?referring_site=bugquickviewclick)

debug output when issue occurs:

%ILPOWER-7-DETECT: Interface Gi7/0/24: Power Device detected: Cisco PD
%ILPOWER-5-IEEE_DISCONNECT: Interface Gi7/0/24: PD removed
%ILPOWER-3-CONTROLLER_PORT_ERR: Controller port error, Interface Gi7/0/24: Power Controller reports power Tstart error detected

A restart will "solve" the issue but what can be done to avoid this in future. Disabling and re-enabling inline power and the interface doesn't help either.

Thanks
Andy

20 Replies 20

Hi,

 

I have the same issue with a 3650-48PD (3.6.6E)...

anyone has find a solution ?

 

thank you

Sam

 

Hi All,

I was getting same issue on port Fa0/23, I change the port to other free switch port and issue was resolved.

andrewswanson
Level 7
Level 7

Upgraded the WS-C3650-48PD stacks to 03.07.05E and I'm still seeing this issue. The faulty status is appearing on interfaces where devices aren't PoE capable. I can now clear this faulty status with a "shutdown/no shutdown" which is a bit better than reloading the switch.

Cheers
Andy

I've got this atm with a 3650, I was at 03.07.05E, Moved to 16.9.3 and still experiencing the issue.

resets or reloads will not resolve this. 

 

 

Same error report on C3560-24PS  for me, only restarting doesn't resolve the issue. Every other top port is affected F0/1, F0/3 etc.

 

*Mar 1 00:04:01.700: %ILPOWER-7-DETECT: Interface Fa0/3: Power Device detected: IEEE PD
*Mar 1 00:04:01.776: %ILPOWER-3-CONTROLLER_PORT_ERR: Controller port error, Interface Fa0/3: Power Controller reports power Tstart error detected

 

I have loaded the latest recommended IOS on and it has not resolved the issue.12.2(55)SE12 

POST says inline power tests passed.

This switch is being scrapped. 

RachelGomez161999
Spotlight
Spotlight

Steps to fix this error-

A determination of how much power a PD really draws in the field is not very easy. When the power controller on the switch detects there is more power being drawn on a port, it shuts the port down and notifies Cisco IOS® of the fact that the PD has exceeded the maximum allocated power. In Cisco IOS you can see the currently drawn power usage per port with the show power inline <interface> detail command.

3850_4#sh power inline Te 3/0/44 detail
Interface: Te3/0/44
Inline Power Mode: auto
Operational status: on
Device Detected: yes
Device Type: Ieee PD
IEEE Class: 3
Discovery mechanism used/configured: Ieee and Cisco
Police: off
Power Allocated
Admin Value: 60.0
Power drawn from the source: 15.0
Power available to the device: 15.0
Actual consumption
Measured at the port: 6.1
Maximum Power drawn by the device since powered on: 6.2
Absent Counter: 0
Over Current Counter: 0
Short Current Counter: 0
Invalid Signature Counter: 0
Power Denied Counter: 0
Power Negotiation Used: IEEE 802.3at LLDP
LLDP Power Negotiation --Sent to PD-- --Rcvd from PD--
Power Type: Type 2 PSE Type 1 PD
Power Source: Primary PSE
Power Priority: low high
Requested Power(W): 12.7 12.7
Allocated Power(W): 12.7 12.7
Four-Pair PoE Supported: Yes
Spare Pair Power Enabled: No
Four-Pair PD Architecture: Shared
The measured value shown at the port in this output is measured by the controller. This information is gathered every few seconds and gives some indication about the drawn power. The value shown with Maximum Power drawn appears useful to troubleshoot Imax errors, but unfortunately that is just a historical display of what the Maximum power drawn by the device has been. If an Imax error occurs, the Power drawn at that time is not reported back to Cisco IOS and will not be displayed there.

As can be seen in the example, the value allocated to the port is 15W. This is the cutoff value that gets programmed onto the interface. Prior to Cisco bug ID CSCuy7423, the Icutoff value is programmed regularly on a port. Every time a CDP packet is received the value will be reprogrammed. After Cisco bug ID CSCuy74231 (fixed in Cisco IOS-XE 3.6.5E and 3.7.5 or later) this programming has been optimized. This reduces the possibility of a "failure" in reprogramming the Icutoff value which leads to an Imax error.

Programming of Icutoff value can be shown via two commands. Either via the trace where the log can be gathered historically or a debug can be enabled to log a debug message when it occurs. The commands to get this are:

show mgmt-infra trace message platform-mgr-poe <switch x>
debug platform poe
The show trace command can only be executed if the active switch in the stack is PoE capable. Otherwise, this command is needed in order to first connect to the PoE member switch in the stack to execute it:

session switch <x>
*May 20 00:34:04.445:CDP-PA: Packet received from AP2 on interface TenGigabitEthernet3/0/44
**Entry found in cache**
*May 20 00:34:04.445: %IOSXE-7-PLATFORM: MEMBER: 3 process platform_mgr: PoE Info: Dequeued POE SPI msg ver 1 if_id 73003723793629284
num_ports 1 req_id 650 msg_type 20
*May 20 00:34:04.452: %IOSXE-7-PLATFORM: MEMBER: 3 process platform_mgr: PoE Info: E_ILP_SET_CUTOFF if_id 73003723793629284
*May 20 00:34:04.452: %IOSXE-7-PLATFORM: MEMBER: 3 process platform_mgr: PoE Info:port 44 icutoff power 15000
*May 20 00:34:04.452: %IOSXE-7-PLATFORM: MEMBER: 3 process platform_mgr: PoE Info: re_poe_set_icutoff_current port 44 power 15000
*May 20 00:34:04.452: %IOSXE-7-PLATFORM: MEMBER: 3 process platform_mgr: PoE Info: scale factor 22 for power 15000
*May 20 00:34:04.452: %IOSXE-7-PLATFORM: MEMBER: 3 process platform_mgr: PoE Info: POE_SET_CUTOFF_CURRENT_SCALE_FACTOR sent
for port 44 (e:11)
As mentioned earlier, it is a complex process to diagnose Imax errors. There is not much information logged at the time an Imax error occurs. The controller shuts the port down and the PD would have typically lost all logs in regards to what it was doing at the time it drew more power than allocated. Measurement of the drawn power by a port in the field is not easy, but with static allocated power a determination could be made. By statically allocating more power than would be requested dynamically, it is possible to determine how much more power the PD would draw that would trigger the Icutoff threshold to be exceeded. A static maximum power consumption can be configured on a switch port with the command power inline static max <value>.

3850_4#sh run int te 3/0/44
interface TenGigabitEthernet3/0/44
power inline static max 20000
end

3850_4#sh power inline te 3/0/44 detail
Interface: Te3/0/44
Inline Power Mode: static
Operational status: on
Device Detected: yes
Device Type: Ieee PD
IEEE Class: 3
Discovery mechanism used/configured: Ieee and Cisco
Police: off
Power Allocated Admin Value: 20.0
Power drawn from the source: 20.0
Power available to the device: 20.0

 

This may help you,

Rachel Gomez

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community:

Review Cisco Networking products for a $25 gift card