Diagnosing AWS IoT Communication failure

RE: Diagnosing AWS IoT Communication failure

Emily Grace — Tue, 28 Oct 2025 04:57:43 GMT

Diagnosing AWS IoT communication failures often involves checking device connectivity, message broker configurations, and authentication policies. Network latency and MQTT topic mismatches can also be culprits. In complex deployments like Tower Boxes Melbourne, maintaining clear monitoring dashboards and structured logging ensures quick isolation of issues and smoother IoT performance across all connected systems.

RE: Diagnosing AWS IoT Communication failure

Achim Kraus — Sat, 23 Mar 2024 06:38:28 GMT

> The actual TLS record size is not visible in the log. Is customer able to check and reconfigure the TLS record size server is using.

Searching for that issue other tickets already mentions the solution for this years ago. AFAIK mbedtls supports that extension so it may be more a question to configure the modem build to use it in a matching way.

RE: Diagnosing AWS IoT Communication failure

Hakon — Mon, 26 Feb 2024 10:30:29 GMT

Comment from modem team regarding the modem trace;

Attached picture points to a problem in DNS query but modem log doesn’t reveal any issues in DNS.

However, there seem to be problems in TLS data transmission. In all TLS connections the device (modem) closes TLS connection with error cause “Too long SSL fragment has been received from a network” meaning that server has sent a TLS packet that does not fit to devices TLS buffer. For the TLS buffer sizes there is following limitation mentioned in modem release notes:
Limitations
- TLS/DTLS Secure socket buffer size is 2kB

The actual TLS record size is not visible in the log. Is customer able to check and reconfigure the TLS record size server is using.

RE: Diagnosing AWS IoT Communication failure

Hakon — Fri, 16 Feb 2024 16:14:15 GMT

[quote user="Ivan Popov"]Here, in the archive there are some traces from today, including the required .bin file:[/quote]

Thanks, I will take a look.

[quote user="Ivan Popov"]Do you think it might give me some hints about what's going on if I go and compare both samples?[/quote]

They use entirely different libraries, so I doubt it is very useful.

RE: Diagnosing AWS IoT Communication failure

Ivan Popov — Wed, 14 Feb 2024 13:40:52 GMT

Here, in the archive there are some traces from today, including the required .bin file:

devzone.nordicsemi.com/.../aws_2D00_iot_2D00_traces_2D00_2024_2D00_02.14.zip

Meanwhile, as I was browsing the code samples related to AWS I noticed that there is a new (at least to me) sample related to the subject matter:

The one in the Zephyr samples tree. Do you think it might give me some hints about what's going on if I go and compare both samples?

Regards

RE: Diagnosing AWS IoT Communication failure

Hakon — Fri, 02 Feb 2024 09:56:04 GMT

[quote user="Ivan Popov"]I can add another trace file but from that I'm seeing in WireShark, the information would be the same as in the previously attached trace.[/quote]

Can you also include the raw trace (.bin file)?

[quote user="Ivan Popov"]Also in the Cellular Monitor I get this while executing the modem trace:[/quote]

Yes, RSRP is -81, threshold is around -115 so this should be fine.

RE: Diagnosing AWS IoT Communication failure

Ivan Popov — Wed, 31 Jan 2024 15:28:53 GMT

Hi,

Thanks for the input. Is this output relevant in some way?

Also in the Cellular Monitor I get this while executing the modem trace:

I can add another trace file but from that I'm seeing in WireShark, the information would be the same as in the previously attached trace.

RE: Diagnosing AWS IoT Communication failure

Hakon — Wed, 31 Jan 2024 15:22:02 GMT

Hello,

have you checked the signal quality? You can run the command AT+CESQ? to read various parameters including RSRP and RSRQ.

RE: Diagnosing AWS IoT Communication failure

Ivan Popov — Sun, 28 Jan 2024 16:13:31 GMT

UPDATE

I followed the Cellular Monitor section in the documentation. As an added value I updated both the SIP and SoC firmwares to the latest versions (with no change in the behaviour).

I could test with a set of three applications:

The main application (App A) I'm using - it is the aws_iot sample configured the proper way and with additions only appending sensor readings (ext_sensor.h & ext_sensor.c) to the shadow update - the MQTT (or overall connectivity) functionality was not touched.
Basic MQTT client (App B) - only the aws_iot sample as it is configured with my credentials (tested was previously working - not touched since then).
The full Nordic Asset Tracker v2 (App C) - the sample from the repository only configured with my credentials - the MQTT (or overall connectivity) functionality was not touched.

The Wireshark for App A and App B showed sets of these:

App A

App B

App C

I can see that there is some resetting of the connection on the TCP protocol level but I'm not familiar with this type of logs and traces and I'm not sure how to extract the relations between the messages and the functionality in the code.

Here I provide the relevant trace in two file formats in full

devzone.nordicsemi.com/.../aws_2D00_iot_2D00_mqtt_2D00_connection_2D00_problem.zip

I hope there are enough hints in there so I can get a proper advise and help to figure out the problem. I'll provide any additional information if required.

Almost forgot to mention that on the serial terminal the behviour looks like this:

Regards!

RE: Diagnosing AWS IoT Communication failure

Ivan Popov — Sun, 28 Jan 2024 14:01:35 GMT

Yes, thanks!
I needed to catch up a bit on the subject so I followed this documentation section and this DevAcademy lesson. I have some finding to share and I'll do it in an update post later.

Regards

RE: Diagnosing AWS IoT Communication failure

AlekFlatley — Fri, 26 Jan 2024 17:38:03 GMT

You can take a modem trace and analyze them using the LTE Link Monitor tool available in nRF Connect for Desktop. From there you can narrow down the exact problem with network connectivity and NB-IoT mode. getting over it