This article provides an example of a Shelf Manager log entry, and explains how to find out more information through basic Shelf Manager commands. It assumes a working knowledge of the 5000 chassis Shelf Manager functionality and software.
Each FRU or Field Replaceable Unit within a 5000 chassis has in-built sensors designed to inform the Shelf manager of the health of the device. The Shelf Manager uses this information to monitor and take action, for example, generating a trap or log entry.
Solution
This example uses a FortiGate-5001A blade. Each Fortinet blade has a number of sensors which capture the temperature, the voltage, and the hot-swap status. These sensors report to the Shelf Manager situations where the threshold values are exceeded so that appropriate action can be taken.
Each sensor has 3 thresholds :
- Upper Non-Critical Threshold
- Upper Critical Threshold
- Upper Non-Recoverable Threshold
Each time the sensor exceeds a threshold a log entry will be created.
If the Upper Critical Threshold is exceeded, a critical alarm will be generated in the log and the Critical Alarm light illuminated on the Shelf Alarm Panel.
To view the log and issue the example commands you must be in the Command Line Interpreter, this is achieved by logging into the Shelf Manager and typing CLIA.
# clia
Pigeon Point Shelf Manager Command Line Interpreter CLI> |
An example Log Entry is outlined below – this was obtained by viewing the log using the [clia sel] command. .
0x0018: Event: at Apr 27 17:45:19 2010; from:(0x98,0,0); sensor:(0x01,14); event:0x1(asserted): "Upper Non-Critical", 0x07 0xFF 0xFF
To decode the events in the log you can use the verbose command [clia sel –v] which provides additional information for each log entry.
0x0018: Event: at: Apr 27 17:45:19 2010; from IPM
Controller: 0x98, LUN: 0, Channel: 0
"Temperature" (0x01) sensor # 14
"Threshold" (0x01) event Asserted
"Upper Non-Critical Going High"
This log entry states that on [Fru = 98] the sensor [14] has reported exceeding the ‘’Upper Non-Critical’’ temperature threshold.
To further decode the event in the log, below are some example commands
1. Confirm which FRU is generating the alarm by verifying the FRU information.
CLI> fru 98
| 98: FRU # 0 Entity: (0xa0, 0x60) Hot Swap State: M4 (Active), Previous: M3 (Activation In Process), Last State Change Cause: Normal State Change (0x0) Device ID String: "FG5001A" |
This highlights that the alarm has come from a temperature sensor on a FG5001A blade.
2. Check the Sensor Threshold values
CLI>threshold 98 14
| 98:
LUN: 0, Sensor # 14 ("DS75 Temp") Type: Threshold (0x01), "Temperature" (0x01) Upper Non-Critical Threshold, Raw Data: 0x32 Processed data: 50.000000 degrees C Upper Critical Threshold, Raw Data: 0x41 Processed data: 65.000000 degrees C Upper Non-Recoverable Threshold, Raw Data: 0x50 Processed data: 80.000000 degrees C |
From this information we can determine that the Upper Non-Critical threshold has been exceeded, the threshold value is 50 degrees C.
3. Now we check the current value of the Sensor
CLI>sensordata 98 14
| 98:
LUN: 0, Sensor # 14 ("DS75 Temp") Type: Threshold (0x01), "Temperature" (0x01) Belongs to entity (0xc1, 0x61): FRU # 1 (AMC # 1) Status: 0xc0 All event messages enabled from this sensor Sensor scanning enabled Initial update completed Raw data: 47 (0x2f) Processed data: 47.000000 degrees C Status: 0x00 |
This indicates at the time the sensor was verified it reported 47 degrees which is close to the threshold value.
Conclusion & Recommended Actions
To summarize the log entry indicates that a sensor on a Fortinet blade has reported to the Shelf Manager that it has exceeded a temperature threshold but not to a critical level. It should be noted that the Shelf Manager constantly monitors temperature sensor information from all installed devices and will automatically raise and low the fan tray speeds to allow for dynamic cooling. It may therefore be that no action is required in this situation.
Depending on the frequency, one follow-up action would be to verify the fan trays to see if they are all working correctly and producing optimal cooling.
