AP 9120 temperature

[Moudar]

Hi

Many of 9120 APs shows very high temperature. Even if schools are closed this week, the APs still generate logs about very high temperature. Here is an example.

ov  1 19:26:44 root: Thermal-Control-Unit - Sensor : Local , Current-status : good, Current-Temperature: 60 C, Highest-recorded : 63 C
Nov  1 19:26:44 root: Thermal-Control-Unit - Sensor : Remote, Current-status : good, Current-Temperature: 46 C, Highest-recorded : 49 C
Nov  1 19:56:45 root: Thermal-Control-Unit - Sensor : Local , Current-status : good, Current-Temperature: 60 C, Highest-recorded : 63 C
Nov  1 19:56:45 root: Thermal-Control-Unit - Sensor : Remote, Current-status : good, Current-Temperature: 46 C, Highest-recorded : 49 C

Nov  2 00:26:52 root: Thermal-Control-Unit - Sensor : Local , Current-status : good, Current-Temperature: 60 C, Highest-recorded : 63 C
Nov  2 00:26:52 root: Thermal-Control-Unit - Sensor : Remote, Current-status : good, Current-Temperature: 46 C, Highest-recorded : 49 C
Nov  2 00:56:52 root: Thermal-Control-Unit - Sensor : Local , Current-status : good, Current-Temperature: 60 C, Highest-recorded : 63 C
Nov  2 00:56:52 root: Thermal-Control-Unit - Sensor : Remote, Current-status : good, Current-Temperature: 46 C, Highest-recorded : 49 C

 What is the difference between local and remote. These APs are in schools where the room temperature is not exeeding 22 Celsius degrees.

This is a different AP:

Nov  2 04:28:34 root: Thermal-Control-Unit - Sensor : Remote, Current-status : good, Current-Temperature: 47 C, Highest-recorded : 51 C
Nov  2 04:58:34 root: Thermal-Control-Unit - Sensor : Local , Current-status : good, Current-Temperature: 60 C, Highest-recorded : 64 C
Nov  2 04:58:34 root: Thermal-Control-Unit - Sensor : Remote, Current-status : good, Current-Temperature: 47 C, Highest-recorded : 51 C
Nov  2 05:28:35 root: Thermal-Control-Unit - Sensor : Local , Current-status : good, Current-Temperature: 59 C, Highest-recorded : 64 C

According to Cisco:

Cisco Catalyst 9120AXI

●  Nonoperating (storage) temperature: -22° to 158°F (-30° to 70°C)
●  Nonoperating (storage) altitude test: 25˚C, 15,000 ft.
●  Operating temperature: 32° to 122°F (0° to 50°C)
●  Operating humidity: 10% to 90% (noncondensing)
●  Operating altitude test: 40˚C, 9843 ft.
Note: When the ambient operating temperature exceeds 40°C, the access point will shift from 4x4 to 2x2 on both the 2.4-GHz and 5-GHz radios, uplink Ethernet will downgrade to 1 Gigabit Ethernet; however, the USB interface will remain enabled

Cisco Catalyst 9120AXE and 9120AXP

●  Nonoperating (storage) temperature: -22° to 158°F (-30° to 70°C)
●  Nonoperating (storage) altitude test: 25˚C, 15,000 ft.
●  Operating temperature: -4° to 122°F (-20° to 50°C)
●  Operating humidity: 10% to 90% (noncondensing)
●  Operating altitude test: 40˚C, 9843 ft.

What is the effect of changing from 4x4 to 2x2 on clients. Many IPad clients have complained about bad connection specially in classrooms with 9120 AP!

Here is another AP log:

 2 07:25:15 root: Thermal-Control-Unit - Sensor : Local , Current-status : good, Current-Temperature: 61 C, Highest-recorded : 66 C
Nov  2 07:25:15 root: Thermal-Control-Unit - Sensor : Remote, Current-status : good, Current-Temperature: 47 C, Highest-recorded : 52 C

[Rich R]

The datasheet specs are for the environment the AP is installed in, not the AP internal temperatures.

The internal temperatures depend on where the sensor is measuring - CPU is usually the hottest.
You can tell from those logs "Current-status : good" that the temp is within normal operating thresholds for that component so nothing to worry about.

For a better understanding of Spatial Streams take a look at:
https://www.netally.com/network-performance/spatial-streams/
https://www.mbreviews.com/what-are-spatial-streams/
https://www.digitalairwireless.com/articles/blog/wi-fi-spatial-streaming-explained
You can find plenty of other blogs, videos etc with more info.
Short answer 2x2 = lower speed and capacity.

> "Many IPad clients have complained about bad connection"
Can you define bad connection?
This could mean slow, high packet loss, frequent disconnection etc etc etc.  You need to be really specific about the problem.
Again plenty of docs to refer to but at a very high level make sure software is up to date (refer to TAC recommended link below) and make sure client OS and drivers are up to date.
Check your WLC config with the Config Analyzer (link below).
Make sure clients are using 5GHz bands not 2.4GHz.
If you're trying to use 80MHz or 160MHz channel width that's generally pointless and more likely to cause problems - reduce to 40MHz channel width.
Make sure you have enough AP capacity for the number of clients and the bandwidth they're using.
Optimise the data rates to improve airtime utilisation.
https://www.cisco.com/c/en/us/products/collateral/wireless/catalyst-9800-series-wireless-controllers/guide-c07-743627.html#Lowdatarates
Also refer to:
https://www.cisco.com/c/en/us/td/docs/solutions/CVD/Campus/cisco-campus-lan-wlan-design-guide.html
https://www.cisco.com/c/dam/en_us/solutions/industries/docs/education/cisco_wlan_design_guide.pdf

[Moudar]

What makes me think about these logs first is that they are coming much often, so I need to exclude that these logs mean a problem.

If the logs about temperature are excluded, what about this last one which also comes often:

Nov  4 12:56:47 kernel: [*11/04/2023 12:56:47.2736] chatter: simple_action packet too short (expected 0xec00, payload 28)
Nov  4 13:08:42 kernel: [*11/04/2023 13:08:42.2008] chatter: simple_action packet too short (expected 0xd300, payload 18)
Nov  4 15:32:16 kernel: [*11/04/2023 15:32:16.5033] chatter: simple_action packet too short (expected 0x3200, payload 18)
Nov  4 20:40:36 kernel: [*11/04/2023 20:40:36.9308] chatter: simple_action packet too short (expected 0x2700, payload 18)
Nov  4 20:48:32 kernel: [*11/04/2023 20:48:32.8159] chatter: simple_action packet too short (expected 0x2000, payload 34)
Nov  4 20:48:38 kernel: [*11/04/2023 20:48:38.8579] chatter: simple_action packet too short (expected 0x5700, payload 34)

[Rich R]

Might be significant, might not be.  If you start looking at debug logs you'll see all sorts of messages that only have meaning to the software developers.  If you want those investigated then open a TAC case but they'll probably say to ignore those.

Did you make sure your software is up to date?
Currently AireOS should be 8.10.190.0 and current IOS-XE (9800) should be 17.9.4 + SMUs + APSPs but always refer to TAC recommended for latest info.