esp32 switch_example issues

[rgb-ryan]

Hello. Happy day

The esp32 device is located in the st-device-sdk-c-ref/apps/esp32/switch_example path
Testing switch_example code.

There were 2 problems during the test.
I would like to ask for your help with the reason and how to solve it.

1. The first issue

2. Test Environment

• phone : samsung galaxy s23
• android ver : 14
• smartthings ver : 1.8.17.22
• app automation conditions

[app image]

2. Issue

1. case 1

• Auto iteration condition 7 hours or so in the app "Routine run but some actions don't run" message and wait for the app in a stopped state.

[app image]

[esp log]

2. case 2

• "mqtt idn't get PINGRESP" ------> app standby after "MQTT connect success" during 21 hours of automation condition repetition. (No message in app)

• "mqtt idn't get PINGRESP" ----> MQTT connection is lost and reconnection is repeated multiple times.
  "MQTT connect success" message and app standby.

[esp log]

[esp log]

[rgb-ryan]

Please save the log file image and look at it.

[junyoun-kim]

Hello @rgb-ryan Sorry for late noticing. Is there still issue ocurring?

[junyoun-kim]

No response from reporter

[rgb-ryan]

@junyoun-kim
Thank you for your response.
The issue is still ongoing. Can you please open an issue and help me?

[junyoun-kim]

Hello @rgb-ryan , it looks like there was network disconnection in esp32. Network disconnection can happen depends on network environment or other factors. So it can't be always guaranteed all command(turn off/trun on) operates successfully.

[rgb-ryan]

@junyoun-kim
hello.

Below is a list of common errors that occur during operation. Are they all due to instability of the wifi network?

After the error message, manually turning it on/off in the 'smartthing app' works fine, but I understand that it is because the network connection is stable when manually operated.

Let's assume that the network is frequently disconnected.

The following problem does not occur in Samsung Electronics home appliances that are currently connected to the same wifi network.
Is it common to supplement the sequence with the procedure below? If there is a better way, please help.

1. Detect network instability or disconnection.
2. Reconnect the network.
3. Retry the failed operation.

------------- most frequent error list ----------------

E (100613346) [IoT]: _iot_net_tls_read(600) > mbedtls_ssl_read = -0x0000

E (100613346) [IoT]: _iot_net_tls_read(600) > mbedtls_ssl_read = -0x004C

E (109648276) [IoT]: _iot_net_tls_read(600) > mbedtls_ssl_read = -0x7880

E (114469726) [IoT]: _iot_net_tls_read(600) > mbedtls_ssl_read = -0x7880

W (117058126) [IoT]: _iot_mqtt_check_alive(877) > mqtt didn't get PINGRESP

E (123974306) [IoT]: _iot_net_tls_read(600) > mbedtls_ssl_read = -0x7880

E (125370226) [IoT]: _iot_net_tls_read(600) > mbedtls_ssl_read = -0x0000

E (125370226) [IoT]: _iot_net_tls_read(600) > mbedtls_ssl_read = -0x004C
W (535496) [IoT]: _iot_mqtt_process_received_publish(457) > duplicated PUB packet 26015

E (77158806) [IoT]: _iot_net_tls_connect(488) > mbedtls_ssl_handshake = -0x6800

E (77158806) [IoT]: _iot_mqtt_connect_net(798) > MQTT net connection failed

E (77158806) [IoT]: _iot_es_mqtt_connect(938) > _iot_es_mqtt_connect error(-1)

E (77158816) [IoT]: iot_es_connect(1102) > failed to connect

E (77158826) [IoT]: _do_iot_main_command(722) > failed to iot_es_connect for communication try count 1 next after 2829 ms

E (77183286) [IoT]: _iot_net_tls_read(600) > mbedtls_ssl_read = -0x7880

I (77183286) [IoT]: _iot_net_show_status(89) > [1073427092] network socket status: sockfd 1726124974 readable 0 writable 54 sock_err 0 errno 0

E (77183296) [IoT]: _iot_es_mqtt_connect(938) > _iot_es_mqtt_connect error(-9)

E (77183306) [IoT]: iot_es_connect(1102) > failed to connect

E (77183316) [IoT]: _do_iot_main_command(722) > failed to iot_es_connect for communication try count 2 next after 4451 ms

E (77203146) [IoT]: _iot_net_tls_connect(488) > mbedtls_ssl_handshake = -0x7280

E (77203146) [IoT]: _iot_mqtt_connect_net(798) > MQTT net connection failed

E (77203146) [IoT]: _iot_es_mqtt_connect(938) > _iot_es_mqtt_connect error(-1)

E (77203156) [IoT]: iot_es_connect(1102) > failed to connect

E (77203166) [IoT]: _do_iot_main_command(722) > failed to iot_es_connect for communication try count 3 next after 8899 ms

[junyoun-kim]

Hello @rgb-ryan , the device SDK try to reconnect when it detects disconnection like PING fail etc. So in most cases, it will reconnect successfully within 1 ~ 2 mins.

[rgb-ryan]

hello @junyoun-kim ,

I understand that stdk automatically reconnects.

The problem I need to solve now is that if the device encounters a network problem while in auto-routine, it stays in standby after reconnecting (auto-routine stops).

How can I solve this issue?

[junyoun-kim]

@rgb-ryan , On your auto-routine example, it looks like routine condition trigger is trun-off. And due to previous routine fail(fail to trun on and trun off after 1 min), it didn't trigger routine condition next time. Auto routine service don't retry failed operation.

[rgb-ryan]

@junyoun-kim ,

I understand that stdk does not retry if the automatic routine fails due to network problems.

I think I need to add code to save the failed routine using stdk and retry after reconnection. Can you help me?

When the automatic reconnection is done, it seems like the state is the same as calling the connection_start(); function.

I need to develop a function that can recover the failed routine after network reconnection due to network connection failure or failed command.

I need help using the stdk library to implement the above function.

[junyoun-kim]

@rgb-ryan There is some misunderstanding. SmartThings App don't retry failed operation not stdk. There is no way from stdk to know if there is failed auto routine operation.