Need help

[MarquisMCMXCIV]

I know nothing about programming and ordered all the parts thinking I would figure it out, but it's been a week and I can't find a way to make it work. I followed all the steps from adafruit, transfered the libraries onto the board, copied/pasted the code.py and secrets.py (with my information in it) but I always get error messages. I know it is connected to my wifi, but from there I don't know how to diagnose the problems I am facing.

A desesperate pilot that just wants a cool addition to his living room decor..

Thanks for the help

[resmithn]

What are the error messages you are seeing?

[MarquisMCMXCIV]

Traceback : File "code.py", line 144, in AttributeError: .show(x) removed. Use .root_group = x

And after saving the code.py on the board, it keeps resetting the matrix portal non stop.

[resmithn]

Below are the lines before/after line 144 in code.py. I think "show" has changed in CircuitPython 9x, so question is what version are you using on the M4. Found this on Github which may help, here is the link: https://github.com/adafruit/circuitpython/issues/8499 And the comment within was this:

show is still valid in 8.x and will be the last version it is valid. transitioning to total deprecation of display.show in 9.x in favor of display.root_group do something like this instead

main_group = displayio.Groupdisplay.root_group = main_group

If you are in fact using Circuitpython 9.x then try changing line 144 to: display.root_group = g

lines from code.py below

139 label3_long='' 138 139 # Add the labels to the display 140 g = displayio.Group() 141 g.append(label1) 142 g.append(label2) 143 g.append(label3) 144 matrixportal.display.show(g) 145 146 # Scroll the plane bitmap right to left (same direction as scrolling text) 144 def plane_animation():

On Thu, Sep 19, 2024 at 10:07 AM MarquisMCMXCIV @.***> wrote:

Traceback : File "code.py", line 144, in AttributeError: .show(x) removed. Use .root_group = x

And after saving the code.py on the board, it keeps resetting the matrix portal non stop.

— Reply to this email directly, view it on GitHub https://github.com/smartbutnot/flightportal/issues/26#issuecomment-2361088195, or unsubscribe https://github.com/notifications/unsubscribe-auth/AK4MZVT3R4GUNTCJ2T5KCQ3ZXLLBDAVCNFSM6AAAAABOONKRHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNRRGA4DQMJZGU . You are receiving this because you commented.Message ID: @.***>

[resmithn]

this is corrected line, got concatenated in email main_group = displayio.Group display.root_group = main_group

[resmithn]

Just an additional comment. I am using Circuitpython 8.1.0 . I tried 9.x with this code and received a lot of errors. You may want to consider loading the uf2 file for 8.1.0 and also use the libraries for that version.

here is a link to download from Adafruit: https://circuitpython.org/board/matrixportal_m4/ Scroll down to Previous versions.

[MarquisMCMXCIV]

I was using latest version

Don't know if I did all the steps you said correctly, but if so, those are the messages I get now.

Traceback (most recent call last): File "code.py", line 5, in File "adafruit_matrixportal/matrixportal.py", line 33, in File "adafruit_matrixportal/network.py", line 32, in File "adafruit_portalbase/network.py", line 28, in File "adafruit_io/adafruit_io.py", line 30, in File "adafruit_minimqtt/adafruit_minimqtt.py", line 37, in ImportError: no module named 'adafruit_connection_manager'

[resmithn]

You didn't mention what version of CircuitPython you loaded. I added some additional comments about CP 9.x and this code not working.. I think this project was written for CircuitPython 8.x or maybe even 7.x so I would suggest you load CircuitPython 8.1.0 as I mentioned in the subsequent comments I made. That is what I have loaded on the M4 and it works okay, aside from the other issues that are in the issues thread, i.e. board resets.

On Thu, Sep 19, 2024 at 11:28 AM MarquisMCMXCIV @.***> wrote:

Don't know if I did all the steps you said correctly, but if so, those are the messages I get now.

Traceback (most recent call last): File "code.py", line 5, in File "adafruit_matrixportal/matrixportal.py", line 33, in File "adafruit_matrixportal/network.py", line 32, in File "adafruit_portalbase/network.py", line 28, in File "adafruit_io/adafruit_io.py", line 30, in File "adafruit_minimqtt/adafruit_minimqtt.py", line 37, in ImportError: no module named 'adafruit_connection_manager'

— Reply to this email directly, view it on GitHub https://github.com/smartbutnot/flightportal/issues/26#issuecomment-2361344193, or unsubscribe https://github.com/notifications/unsubscribe-auth/AK4MZVQRD4OQHPYHXTPMJ4TZXLUTFAVCNFSM6AAAAABOONKRHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNRRGM2DIMJZGM . You are receiving this because you commented.Message ID: @.***>

[MarquisMCMXCIV]

Like I said, I was using latest vesrsion which was 9.X and now I tried puting 8.1.0 like suggested, same for the libraries and now I get the messages previously sent. Not even sure if all I did, while "trying" to install 8.1.0 was correctly done as I am a newbie to programming and very limited in my knowledge..

Sorry if what I say doesn't make sense to you

[resmithn]

Appears the "libraries" are out of sync. Did you use this link to get the 8.x library download. https://circuitpython.org/libraries

[resmithn]

This is what I have loaded in the lib folder for 8.1.0

[MarquisMCMXCIV]

yes, got the libraries from the link you sent me and just added the missing libraries to my board.

now that's what I got

Code stopped by auto-reload. Reloading soon. soft reboot

Auto-reload is on. Simply save files over USB to run them or enter REPL to disable. code.py output: Check and reconnect WiFi Connect attempt 1 of 10 Reset ESP... Attempt WiFi connect... Successfully connected.

Does it mean it's working and now I just need to wait for a plane to fly over?? :)

[resmithn]

Looks Good!! Look in the code.py around line 419 for

print("Get flights...")

flight_id=get_flights() w.feed() remove the # from the print statement, just a visual that you got that far. Comment if you need anything further.

[MarquisMCMXCIV]

Ok I'll do that, thanks!

Now I don't know why, but I tried changing the bounds box coordinates to another place closer to the airport to test if it's working, but when doing so, the wifi connection doesn't seem to be working anymore..

I don't see what's the relation between bounds box and wifi connection.

[resmithn]

Any messages displayed? What I do is put a few print statements in the code [some are already there, just commented out] to see what is happening. You are right, bounds box shouldn't affect wifi.

[MarquisMCMXCIV]

Managed to figure out what was the problem with the bounds box.. by removing the # now it constantly says get flights... which is nice, but it still doesn't show any flight information on the display when a flight is overflying my house. I did enlarge the scanning zone to increase the chances of getting something,

[resmithn]

This is the bound box tool I use. https://norbertrenner.de/osm/bbox.html Use mouse wheel to resize the map to your location, then click "draw" upper left. Bottom right shows the latitude/longitude, add to secrets file paying attention to the exact order of the coordinates.

And make sure to include the "-" sign for longitude, if present. My secrets file looks like this: 'bounds_box' : '27.5892,27.1892,-82.8524,-82.3524' This is the order the coordinates should be in the secrets.py file: top latitude, bottom latitude, left longitude, right longitude

In the code.py file around statement line 395, uncomment the print for "flight found". This will let you know that a flight was found. Also around line 307 should be "print("Flight is called "+flight_number)" which will let you know a flight was found. If you get the flight number displayed it should scroll the flight info. Please let me know if you need further help.

[MarquisMCMXCIV]

Sorry for the delay, just came back from a flight and followed your intructions.

IT WORKS! thank you very much for your help, really apreciated 😁

[resmithn]

Glad to hear it!

On Sat, Sep 21, 2024 at 9:12 AM MarquisMCMXCIV @.***> wrote:

Sorry for the delay, just came back from a flight and followed your intructions.

IT WORKS! thank you very much for your help, really apreciated 😁

— Reply to this email directly, view it on GitHub https://github.com/smartbutnot/flightportal/issues/26#issuecomment-2365181253, or unsubscribe https://github.com/notifications/unsubscribe-auth/AK4MZVRH3PDLVRWQTDMOOLLZXVWCDAVCNFSM6AAAAABOONKRHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNRVGE4DCMRVGM . You are receiving this because you commented.Message ID: @.***>

[resmithn]

Thought I would share this revised code.py file with you. I changed the label positions [centered] and some of the colors. Also added a banner when a flight is found [before the plane animation] and another banner when no flight is found. You can change the "flight found" banner for your location or whatever you like. I just think this looks better than the original layout. Comment if you have any questions and the attached file should run on your setup. code.zip

[MarquisMCMXCIV]

Thanks, I'll try it!

Also if that can be helpful, I THINK when it started working and displaying flights over, the board was not resetting like you guys are talking in the other issue. I don't know if it could have something to do with the internal memory of the board being saturated or something like that.. hope that helps.

[resmithn]

Yes, very helpful. Thanks. There is a line in the while loop [towards the bottom of the code] that will show memory use. You can remove the "#" to have it show up in the serial monitor. Usually the code will stop if the board is out of memory. I was just running the native version and the board was still disconnecting, I haven't a clue why this all of sudden started happening.

On Sat, Sep 21, 2024 at 12:58 PM MarquisMCMXCIV @.***> wrote:

Thanks, I'll try it!

Also if that can be helpful, I THINK when it started working and displaying flights over, the board was not resetting like you guys are talking in the other issue. I don't know if it could have something to do with the internal memory of the board being saturated or something like that.. hope that helps.

— Reply to this email directly, view it on GitHub https://github.com/smartbutnot/flightportal/issues/26#issuecomment-2365252843, or unsubscribe https://github.com/notifications/unsubscribe-auth/AK4MZVQI2ECKNFRBHWUZR3DZXWQTFAVCNFSM6AAAAABOONKRHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNRVGI2TEOBUGM . You are receiving this because you commented.Message ID: @.***>

[MarquisMCMXCIV]

I am still running the original code.. it says memory free 38864.

[resmithn]

Just checked the code.py modified version that I sent you: memory free 30080.

On Sat, Sep 21, 2024 at 1:34 PM MarquisMCMXCIV @.***> wrote:

I am still running the original code.. it says memory free 38864.

— Reply to this email directly, view it on GitHub https://github.com/smartbutnot/flightportal/issues/26#issuecomment-2365263424, or unsubscribe https://github.com/notifications/unsubscribe-auth/AK4MZVSOMGYYMAQCE7WH2B3ZXWU23AVCNFSM6AAAAABOONKRHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNRVGI3DGNBSGQ . You are receiving this because you commented.Message ID: @.***>

[MarquisMCMXCIV]

When it restarts there is a little python flashing in the top left corner. Is there a way to avoid that? Cause apart from that I can't tell it is restarting, it still works fine but this is kind of annoying.

[resmithn]

Unfortunately no, just the nature of the circuitpython

On Sat, Sep 21, 2024 at 2:23 PM MarquisMCMXCIV @.***> wrote:

When it restarts there is a little python flashing in the top left corner. Is there a way to avoid that? Cause apart from that I can't tell it is restarting, it still works fine but this is kind of annoying.

— Reply to this email directly, view it on GitHub https://github.com/smartbutnot/flightportal/issues/26#issuecomment-2365277733, or unsubscribe https://github.com/notifications/unsubscribe-auth/AK4MZVR6GBQYCJIO2NJ6PODZXW2RXAVCNFSM6AAAAABOONKRHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNRVGI3TONZTGM . You are receiving this because you commented.Message ID: @.***>

[resmithn]

Just an FYI. I had an ESP32-S3 processor [ 8 MB flash, 2 MB of SRAM] board and was playing around with a Word Clock [64x64]. Then yesterday a new issue appeared on this thread, someone trying to run the flight code on an S3, bought by mistake. The S3 handles wifi and requests a little different than the M4. I took the Word Clock code particulars for the S3 and plugged into the M4 Flight code and loaded to the S3. It worked and the surprise is: No disconnects!! So the issue that all are having with the M4 disconnects appears to be something in the code the M4 doesn't like.

[MarquisMCMXCIV]

I was at my parents house yesterday when I was setting it up first. I had the reset issue during the day, but then when I got home in the evening, I entered my place's wifi information and found out that compared to earlier that day the board was struggling to connect but when it finally did and after finding flights, it was not resetting anymore. Absolutely no idea why...

[resmithn]

I have run out of things to try. I have print statements in the code and best I can tell when it calls "get flights", that is where the disconnect happens. Course you can no longer see the serial monitor to see what happened. The try/except block in get flights looks okay, but yet to see a message from there. So I will just leave it as is. Thanks for your input.

On Sun, Sep 22, 2024 at 4:20 PM MarquisMCMXCIV @.***> wrote:

I was at my parents house yesterday when I was setting it up first. I had the reset issue during the day, but then when I got home in the evening, I entered my place's wifi information and found out that compared to earlier that day the board was struggling to connect but when it finally did and after finding flights, I found out that it was not resetting anymore. Absolutely no idea why...

— Reply to this email directly, view it on GitHub https://github.com/smartbutnot/flightportal/issues/26#issuecomment-2366949167, or unsubscribe https://github.com/notifications/unsubscribe-auth/AK4MZVS7DOJNKDWFUJRFPHTZX4Q7TAVCNFSM6AAAAABOONKRHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNRWHE2DSMJWG4 . You are receiving this because you commented.Message ID: @.***>

[MarquisMCMXCIV]

Sorry to bother you again..

Quick update on what I noticed. I cleared the board entirely and reinstalled everything back on it and put it back to the test.

Yesterday it has been running flawlessly without any resets after flights were detected, and this morning when I woke up it has started to reset again.. so my understanding is that it works fine at the beginning when the board is "empty", but then at some point after some time and after multiple flights are detected, it starts to reset everytime a flight is scanned.

Hope that helps.

[resmithn]

No worries, thanks for the update! Yes, have experienced the same thing. Other thing I've noticed is as long as there are flights found one right after the other and no json error, board doesn't reset. But I still can't isolate the disconnect in the code on the M4. I have that modified version running on an S3, no errors and no disconnects, none. Been trying to help S3 code? in the issues list [StarTrailMeteor https://github.com/StarTrailMeteor], interesting fwiw.

On Wed, Sep 25, 2024 at 11:15 AM MarquisMCMXCIV @.***> wrote:

Sorry to bother you again..

Quick update on what I noticed. I cleared the board entirely and reinstalled everything back on it and put it back to the test.

Yesterday it has been running flawlessly without any resets after flights were detected, and this morning when I woke up it has started to reset again.. so my understanding is that it works fine at the beginning when the board is "empty", but then at some point after some time and after multiple flights are detected, it starts to reset everytime a flight is scanned.

Hope that helps.

— Reply to this email directly, view it on GitHub https://github.com/smartbutnot/flightportal/issues/26#issuecomment-2374385246, or unsubscribe https://github.com/notifications/unsubscribe-auth/AK4MZVRE2HGWXHJTTG2S6NTZYLHQBAVCNFSM6AAAAABOONKRHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNZUGM4DKMRUGY . You are receiving this because you commented.Message ID: @.***>

[MarquisMCMXCIV]

Tried that code but wifi connection for the s3 code doesn't seem to be working on M4. Now the other problem I am facing (I feel I am taking a lot of your time lately haha) with the original code, I don't know if it's because I live in an apartment and that there is many wifis around, but compared to when I first did set it up at my parents house, after the resets, it was instantaneously reconnecting to wifi and scanning for next flights which was making the reset almost unnoticeable and I could've live with it. But now since I've set it at my place, everytime a reset is happening, it's not reconnecting and I have to force resets until it does using the reset button on the board..

Any idea??

[resmithn]

Not a problem. Are you seeing any messages from the wifi function, number of retries, failed to connect... It will keep trying to connect up to ten times. How close is the M4 to your router. You can check your wifi signal with options in the following link. How to Check/View WiFi Signal Strength in Windows 11? - Winsides.com https://winsides.com/how-to-check-view-wifi-signal-strength-in-windows-11/ But I doubt the signal strength is an issue.

I have the S3 code running and as said before no issues, no connect issues, it just runs and runs without disconnects. This is the connect code for the original M4, you should see messages in the serial monitor from the print statements.

def checkConnection(): print("Check and reconnect WiFi") attempts=10 attempt=1 while (not esp.status == adafruit_esp32spi.WL_CONNECTED) and attempt<attempts: print("Connect attempt "+str(attempt)+" of "+str(attempts)) print("Reset ESP...") w.feed() wifi.reset() print("Attempt WiFi connect...") w.feed() try: wifi.connect() except OSError as e:

print(e.class.name+"--------------------------------------") print(e) attempt+=1 if esp.status == adafruit_esp32spi.WL_CONNECTED: print("Successfully connected.") else: print("Failed to connect.")

On Wed, Sep 25, 2024 at 1:53 PM MarquisMCMXCIV @.***> wrote:

Tried that code but wifi connection for the s3 code doesn't seem to be working on M4. Now the other problem I am facing (I feel I am taking a lot of your time lately haha) with the original code, I don't know if it's because I live in an apartment and that there is many wifis around, but compared to when I first did set it up at my parents house, after the resets, it was instantaneously reconnecting to wifi and scanning for next flights which was making the reset almost unnoticeable and I could've live with it. But now since I've set it at my place, everytime a reset is happening, it's not reconnecting and I have to force resets until it does using the reset button on the board..

Any idea??

— Reply to this email directly, view it on GitHub https://github.com/smartbutnot/flightportal/issues/26#issuecomment-2374779750, or unsubscribe https://github.com/notifications/unsubscribe-auth/AK4MZVU56O4HZ27AZPSRZGTZYLZ75AVCNFSM6AAAAABOONKRHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNZUG43TSNZVGA . You are receiving this because you commented.Message ID: @.***>

[MarquisMCMXCIV]

I am not home now but basically, just to make sure I understand, your telling me to take that "connect" part and to replace the one in the your S3 code and that it is supposed to be running perfectly?

For my wifi connection issue, looks like it's trying one time only and then the board is crashing and resetting everytime but without any successful connection.

Again thanks for your help!

[resmithn]

No that code is the M4 code, you said you were running the original code. Was just pointing out the messages you should see on connect. Don't change anything.

On Wed, Sep 25, 2024 at 2:50 PM MarquisMCMXCIV @.***> wrote:

I am not home now but basically, just to make sure I understand, your telling me to take that "connect" part and to replace the one in the your S3 code and that it is supposed to be running perfectly?

For my wifi connection issue, looks like it's trying one time only and then the board is crashing and resetting everytime but without any successful connection.

Again thanks for your help!

— Reply to this email directly, view it on GitHub https://github.com/smartbutnot/flightportal/issues/26#issuecomment-2374903501, or unsubscribe https://github.com/notifications/unsubscribe-auth/AK4MZVWQZOBUIBEF7IDZOLDZYMAVXAVCNFSM6AAAAABOONKRHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNZUHEYDGNJQGE . You are receiving this because you commented.Message ID: @.***>

[MarquisMCMXCIV]

I don't know if that can be of any help for you, but I asked chat gpt and that's what it came out with:

It sounds like the board might be running out of memory or experiencing a watchdog timeout, especially since the script is handling large amounts of data from the FlightRadar24 API. Let's go through a few potential causes and fixes:

1. Memory Management

Microcontrollers like the MatrixPortal M4 have limited memory, and you're working with relatively large JSON files. Even though you are managing memory with a fixed json_bytes array, there could still be memory fragmentation or exhaustion over time, especially when multiple objects are dynamically allocated and freed (such as JSON parsing, networking objects, and display labels).

Solution:

• Manually Freeing Memory: You can attempt to free memory after handling each flight or after every loop iteration by using the gc.collect() function. You are already calling it, but try adding it in more places where memory usage spikes, such as after JSON parsing or network calls.

# Call after handling each flight's data
gc.collect()

• Reduce JSON Data Size: Ensure that you're fetching only the necessary parts of the JSON. Currently, you're cutting off the "trail" portion, but further optimization might be possible by reducing fields in the request headers or response.

2. Watchdog Timeout

The watchdog is set to a 16-second timeout (w.timeout=16). If the JSON fetching, parsing, or scrolling is taking longer than expected, it could trigger the watchdog and reset the board.

Solution:

• Feed the Watchdog More Frequently: Ensure you’re feeding the watchdog frequently enough, especially during tasks that take longer, such as scrolling text or fetching data.

w.feed()

• Increase Watchdog Timeout: If your tasks take longer than 16 seconds, consider increasing the watchdog timeout to give more headroom.

w.timeout = 30  # Increase timeout to 30 seconds

3. Network Instability or Large API Responses

The crash could also be caused by intermittent network issues or large API responses. The code tries to handle these scenarios with retries and error handling, but there might be cases where it still causes the board to reset.

Solution:

• Retry Network Requests: Instead of crashing, ensure you handle network errors gracefully and attempt retries.

try:
    response = requests.get(url=FLIGHT_SEARCH_URL, headers=rheaders).json()
except (RuntimeError, OSError, HttpError, requests.OutOfRetries) as e:
    print("Network issue:", e)
    checkConnection()
    return False

• Smaller API Queries: Try to limit the number of flights or the size of the response by setting appropriate parameters in the API call.

# Try setting a lower limit, if applicable
FLIGHT_SEARCH_TAIL = "&limit=1"  # Already set, but double-check API docs for further optimization

4. Reduce Display Complexity

The plane animation and scrolling text might consume CPU cycles and memory.

Solution:

• Simplify Animations: If you notice frequent crashes, try commenting out the plane_animation() function or reduce the frame rate of the animation:

# Reduce plane speed to make the animation more lightweight
PLANE_SPEED = 0.1  # Increase this value to slow down the animation and reduce CPU load

• Reduce Text Scrolling Speed: Similarly, slow down the text scrolling or skip it altogether if the text fits within the screen without scrolling.

TEXT_SPEED = 0.1  # Increase this to slow down scrolling, reducing CPU load

5. Optimize Display Updates

Continuously updating the display with new objects and text can cause memory issues, especially on a limited system.

Solution:

• Minimize Display Object Changes: Try to reuse display objects like labels and avoid continuously creating or updating them. Instead of re-creating text labels, change only their .text property, which you already seem to be doing. Also, ensure you are not adding/removing objects from the displayio.Group frequently.

6. Watch the Logs and Test Incrementally

Use the logs printed during the crash to determine where the board is resetting. Gradually comment out parts of the code and reintroduce them one at a time to see what specific section might be causing the reset.

---

By optimizing memory usage, adjusting the watchdog timeout, and simplifying the display and network tasks, you can likely resolve the crashing issue. Would you like me to go into more detail on how to log and debug the crash further?

[resmithn]

Thanks, most of those are in the code" w.feed, gc_collect. AI is pretty smart! I think we should take this conversion off Github. I have some alternatives to the M4 you may be interested in, or maybe not. Email me if you are interested, an alternative is running code.py [modified] on Raspberry Pi4 or Windows with native Python [3.11], not MicroPython. my email: @.*** Thanks for your replies.

On Wed, Sep 25, 2024 at 3:48 PM MarquisMCMXCIV @.***> wrote:

I don't know if that can be of any help for you, but I asked chat gpt and that's what it came out with:

It sounds like the board might be running out of memory or experiencing a watchdog timeout, especially since the script is handling large amounts of data from the FlightRadar24 API. Let's go through a few potential causes and fixes:

1. Memory Management

Microcontrollers like the MatrixPortal M4 have limited memory, and you're working with relatively large JSON files. Even though you are managing memory with a fixed json_bytes array, there could still be memory fragmentation or exhaustion over time, especially when multiple objects are dynamically allocated and freed (such as JSON parsing, networking objects, and display labels). Solution:

• Manually Freeing Memory: You can attempt to free memory after handling each flight or after every loop iteration by using the gc.collect() function. You are already calling it, but try adding it in more places where memory usage spikes, such as after JSON parsing or network calls.

Call after handling each flight's datagc.collect()

• Reduce JSON Data Size: Ensure that you're fetching only the necessary parts of the JSON. Currently, you're cutting off the "trail" portion, but further optimization might be possible by reducing fields in the request headers or response.

2. Watchdog Timeout

The watchdog is set to a 16-second timeout (w.timeout=16). If the JSON fetching, parsing, or scrolling is taking longer than expected, it could trigger the watchdog and reset the board. Solution:

• Feed the Watchdog More Frequently: Ensure you’re feeding the watchdog frequently enough, especially during tasks that take longer, such as scrolling text or fetching data.

w.feed()

• Increase Watchdog Timeout: If your tasks take longer than 16 seconds, consider increasing the watchdog timeout to give more headroom.

w.timeout = 30 # Increase timeout to 30 seconds

3. Network Instability or Large API Responses

The crash could also be caused by intermittent network issues or large API responses. The code tries to handle these scenarios with retries and error handling, but there might be cases where it still causes the board to reset. Solution:

• Retry Network Requests: Instead of crashing, ensure you handle network errors gracefully and attempt retries.

try: response = requests.get(url=FLIGHT_SEARCH_URL, headers=rheaders).json()except (RuntimeError, OSError, HttpError, requests.OutOfRetries) as e: print("Network issue:", e) checkConnection() return False

• Smaller API Queries: Try to limit the number of flights or the size of the response by setting appropriate parameters in the API call.

Try setting a lower limit, if applicableFLIGHT_SEARCH_TAIL = "&limit=1" # Already set, but double-check API docs for further optimization

4. Reduce Display Complexity

The plane animation and scrolling text might consume CPU cycles and memory. Solution:

• Simplify Animations: If you notice frequent crashes, try commenting out the plane_animation() function or reduce the frame rate of the animation:

Reduce plane speed to make the animation more lightweightPLANE_SPEED = 0.1 # Increase this value to slow down the animation and reduce CPU load

• Reduce Text Scrolling Speed: Similarly, slow down the text scrolling or skip it altogether if the text fits within the screen without scrolling.

TEXT_SPEED = 0.1 # Increase this to slow down scrolling, reducing CPU load

5. Optimize Display Updates

Continuously updating the display with new objects and text can cause memory issues, especially on a limited system. Solution:

• Minimize Display Object Changes: Try to reuse display objects like labels and avoid continuously creating or updating them. Instead of re-creating text labels, change only their .text property, which you already seem to be doing. Also, ensure you are not adding/removing objects from the displayio.Group frequently.

6. Watch the Logs and Test Incrementally

Use the logs printed during the crash to determine where the board is resetting. Gradually comment out parts of the code and reintroduce them one at a time to see what specific section might be causing the reset.

By optimizing memory usage, adjusting the watchdog timeout, and simplifying the display and network tasks, you can likely resolve the crashing issue. Would you like me to go into more detail on how to log and debug the crash further?

— Reply to this email directly, view it on GitHub https://github.com/smartbutnot/flightportal/issues/26#issuecomment-2375115953, or unsubscribe https://github.com/notifications/unsubscribe-auth/AK4MZVVF4JFDPBOOUMR5NFLZYMHPXAVCNFSM6AAAAABOONKRHOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGNZVGEYTKOJVGM . You are receiving this because you commented.Message ID: @.***>

[MarquisMCMXCIV]

Your email got automomaticay hidden.. mine is jordanmarquis04 at gmail.com

[tschuBBi]

Both of you helped me to realise this project for my wife. THANK YOU. =)