'OK' is tangling in the Receive Buffer (IDFGH-10582)

[diplfranzhoepfinger]

Answers checklist.

• I have read the documentation for esp-protocols components and the issue is not addressed there.
• I have updated my esp-protocols branch (master or release) to the latest version and checked that the issue is present there.
• I have searched the issue tracker for a similar issue and not found a similar issue.

General issue report

Code: https://github.com/diplfranzhoepfinger/esp-protocols-tree-master-components-esp_modem-examples-simple_cmux_client/tree/feature/AT_CPSI

Log: https://gist.github.com/diplfranzhoepfinger/dde959578cb954f23908c07fb10b463a

we observe the following:

the Command "AT+CPSI?" is sent out.
the Response is coming inside the 500ms Timeout.
but the OK is only received after we send out another "AT" with dce->sync();

this happens in CMUX Mode.

is the receive Buffer not flushed while waiting for the OK ? Or what can be the Reason ?

[diplfranzhoepfinger]

Log: https://gist.github.com/diplfranzhoepfinger/e7ba744e4f6ccc54e1119f9439e0ec86

Code: https://github.com/diplfranzhoepfinger/esp-protocols-tree-master-components-esp_modem-examples-simple_cmux_client/tree/feature/AT_CPSI2

we observe that this is NOT happening in Command Mode. 

only in CMUX Mode.

[diplfranzhoepfinger]

Now test to do this multiple times. 

Code: https://github.com/diplfranzhoepfinger/esp-protocols-tree-master-components-esp_modem-examples-simple_cmux_client/tree/feature/AT_CPSI3

Log: https://gist.github.com/diplfranzhoepfinger/9cef72ff68f18592195d9048ac48fad0

[diplfranzhoepfinger]

@david-cermak any news ? 

SIMCOM side or Espressif side ?

[david-cermak]

Those are probably leftovers from the previous command. Does it cause any trouble?

We can flush the buffers before sending a new command.

[diplfranzhoepfinger]

Those are probably leftovers from the previous command. Does it cause any trouble?

yes, as we get "Inquiring UE system information ERROR or TIMEOUT" 

because the "OK" is not coming. 

We can flush the buffers before sending a new command.

that would not solve anything. as the Parser needs the OK to parse right, also it is Part of the Command.

[david-cermak]

yes, as we get "Inquiring UE system information ERROR or TIMEOUT"

it was probably timeout, so the OK was actually coming, but didn't make it before timeout. And the OK response was buffered and appeared in the next command processing.

Clearing buffers before commands is probably the right thing to do, but wouldn't help much (just gives ~4 bytes to buffer size available)

[diplfranzhoepfinger]

yes, as we get "Inquiring UE system information ERROR or TIMEOUT"

it was probably timeout, so the OK was actually coming, but didn't make it before timeout. And the OK response was buffered and appeared in the next command processing.

Thw Question is:

in Command Mode we never see this Problem. 

in CMUX Mode we see this Problem even if we make a really long timeout to this Command, like 15s. 

and the OK comes once we issue a "AT" to the Modem. 

so we are unsure if this is a SIMCOM Modem or a esp-modem Problem.

[david-cermak]

in CMUX Mode we see this Problem even if we make a really long timeout to this Command, like 15s.

Could you please connect another UART-USB converter to modem Tx line and check if the OK comes before or after those 15s?
It might be theoretically also possible that the modem flow control is preventing the SIMCOM device to continue? (ESP32 is not ready to receive? in that case #312 should fix it as well)

[diplfranzhoepfinger]

ah, i will test immediately with #312 so we know.

[diplfranzhoepfinger]

Test here:
https://github.com/diplfranzhoepfinger/esp-protocols-tree-master-components-esp_modem-examples-simple_cmux_client/tree/feature/AT_CPSI4

Again no IP:
https://gist.github.com/diplfranzhoepfinger/a49c4325bcb75f270b93b49a96dfce81

Test again: working:
https://gist.github.com/diplfranzhoepfinger/7656fcdd4f1ac351ca72b3d092762a59

(it looks like not working, but this is because the Parser is not fully written)

the OK is no more tangling. That is good. Write the Parser is just work.

[david-cermak]

If I understand correctly the change in #312 fixed this issue, correct?

If not please reopen

[diplfranzhoepfinger]

i will test just now. Am Mo., 9. Okt. 2023 um 18:04 Uhr schrieb david-cermak < ***@***.***>:

…

If I understand correctly the change in #312 <#312> fixed this issue, correct? If not please reopen — Reply to this email directly, view it on GitHub <#323 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AJAHMXLRQIG7BC3A4CCHWLDX6QOAFAVCNFSM6AAAAAA2AN5WMCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONJTGI4DIMJYGI> . You are receiving this because you authored the thread.Message ID: ***@***.***>

-- Diplom-Informatiker Franz Höpfinger e.K. Reibersdorf 16a 84419 Schwindegg