**** BEGIN LOGGING AT Sun Sep 11 02:59:56 2022 Sep 11 11:26:43 Is negative one not allowed in the PWM lib. you wrote, @zmatt? Sep 11 11:27:34 It is not important. I am just trying to learn how to use it on fading and I see that 0.0 to 1.0 is allowed only. Sep 11 11:29:51 I can perform a negative float but the source does not allow me to fade the LED (I think). Sep 11 11:33:12 scratch that...I cannnot perform a negative float in my source. I can put in -1, i.e. for i in range(-1, 0, 1): Sep 11 11:33:37 Anyway, no rush. Sep 11 11:57:52 blinky != fade Aw! Sep 11 12:03:44 negative? what would that even mean? 0.0 = output is continuously low, 1.0 = output is continuously high, 0.5 = output is 50% low, 50% high Sep 11 12:04:48 Oh. Sep 11 12:04:53 I thought I needed negative. Sep 11 12:04:55 Hmm. Sep 11 12:04:57 Let me try. Sep 11 12:07:28 So... https://pastebin.com/q60nvfHk is wrong? Sep 11 12:08:37 I tried w/ 0.5 too in the for loop. Sep 11 12:08:49 It could be this LED. Sep 11 12:10:36 0.5 in the for loop creates an error: TypeError: 'float' object cannot be interpreted as an integer Sep 11 12:15:40 the while-loop and its contents look fine, the for-loops and their contents look nonsense Sep 11 12:15:48 Right. Sep 11 12:15:58 I know. I am still testing to find a way to fade. Sep 11 12:16:12 duty_cycle? Sep 11 12:17:13 I mean...I am reading the comments. Sep 11 12:17:22 I still do not understand completely. Sep 11 12:18:29 So, configure(duty_cycle=40000)? Sep 11 12:18:39 something like https://pastebin.com/64t4epPu ? Sep 11 12:18:46 I will try it. Sep 11 12:19:03 ignore duty_cycle, it's just an internal thing, using value is more user-friendly Sep 11 12:20:49 Okay. It almost worked. See here: TypeError: range expected at least 1 argument, got 0 Sep 11 12:20:52 That was the error. Sep 11 12:21:26 Oops. Let me see if I typed incorrectly. Sep 11 12:21:33 you did :P Sep 11 12:22:21 btw 50 Hz is really low as PWM frequency Sep 11 12:22:27 Yep. Oh. Sep 11 12:22:39 I was about to say that my LED is crud. Sep 11 12:22:53 Let me try a higher value for frequency. Sep 11 12:23:34 Now, I got all sorts of errors. Sep 11 12:24:10 I did notice one mistake: I changed LED to led but missed one (the disable) Sep 11 12:24:47 Oh. Okay. https://pastebin.com/TNiuX32j Sep 11 12:24:52 Let me try that too. Sep 11 12:25:48 ugh, that looks like the pwm output is being annoying about changing the frequency Sep 11 12:25:55 frequency only works... Sep 11 12:25:59 1. If it is under 100. Sep 11 12:26:05 2. above 0. Sep 11 12:26:14 So far at least. Sep 11 12:26:47 The lower the frequency, the faster the LED fades on/off. Sep 11 12:26:53 I seem to recall there's a bug in the driver where once you set the frequency of both outputs of one ehrpwm device (which must always use the same frequency), you can no longer ever change the frequency even if both outputs are disabled Sep 11 12:27:01 frequency has no effect on fade time Sep 11 12:27:10 Hmm. Sep 11 12:27:15 I can see it! Sep 11 12:27:31 Well, it is more of a blink now instead of a fade. Sep 11 12:27:31 using a low frequency just causes visible flicker Sep 11 12:27:36 Right. Sep 11 12:27:46 Dang. Sep 11 12:27:53 So, dang. I guess that is it. Sep 11 12:28:17 I will read about the sysfs pwm interface some more and keep trying. Sep 11 12:28:34 I am taking some diploma class online for C programming now. So, back to the books! Sep 11 12:29:09 Hopefully, they teach me something that will become stuck instead of this fly-by-night tutorials that I keep reading? Sep 11 12:30:16 https://pastebin.com/afEByU58 convenience wrapper for fading the led Sep 11 12:30:26 to fix the issue with changing the frequency, just reboot Sep 11 12:31:19 Okay. So, once I get the frequency down pat, keep the boot. But every time I need to change frequency, I need to reboot? Sep 11 12:32:13 if I remember correctly you can change the frequency fine as long as you only ever use one of the two outputs of a pwm instance, e.g. *either* /dev/bone/pwm/1/a *or* /dev/bone/pwm/1/b but not both Sep 11 12:32:23 this is assuming the issue is what I think it is Sep 11 12:32:45 Oh. Okay. Sep 11 12:33:14 once you use both, you can't change the frequency anymore... at least, that was the bug I seem to remember, I'm too lazy to test it right now Sep 11 12:33:31 No issue. I am only using p9.16 so far. Sep 11 12:33:42 so, /dev/bone/pwm/1/b/. Sep 11 12:35:09 rebooting! Sep 11 12:41:42 NameError: name 'seconds' is not defined Sep 11 12:41:50 I defined it and it just turns on. Sep 11 12:42:34 hmm, am I not allowed to use a previous parameter in a default argument for the next parameter in python? Sep 11 12:42:42 I know javascript is fine with doing that, maybe python isn't Sep 11 12:42:59 Good question! Sep 11 12:43:06 Let me look that up. Sep 11 12:43:37 https://pastebin.com/AEt0Ssdt fix Sep 11 12:47:35 Yeppers. That did it. Sep 11 12:47:38 Fading! Sep 11 12:47:41 and I guess it may be nicer to just always use the current led value as from-value instead of having to specify it manually: https://pastebin.com/RKEHTcm0 Sep 11 12:48:42 Okay. Sep 11 12:48:47 I will test it. Sep 11 12:50:41 That works too. My led is a bust. Sep 11 12:50:47 It is so dim. Sep 11 12:54:13 Hey @zmatt: When messing w/ CAN1 on the BBB, would I need to have it plugged into another piece of hardware to read specifics on the routes of the CAN interface? Does that make sense? Sep 11 12:54:28 Or... Sep 11 12:55:01 Can I just plug in the tranceiver to the BBB on CAN1 and get readings or does it need to be plugged into the OBD-II port on a vehicle? Sep 11 12:56:04 I tried to set up wireshark so I can visually see what CAN1 is doing on the BBB w/ my tranceiver. Sep 11 12:56:15 Wireshark does not run. Too much consumption probably. Sep 11 20:56:42 i am working with pulse audio and want to use a microsd as a directory. Sep 11 20:56:58 anyone familiar with the situation? Sep 11 20:57:13 ?? what do these two things have to do with each other? Sep 11 20:57:42 (also, pulseaudio is something you generally only use on desktop systems) Sep 11 20:58:16 I have debian installed on the beaglebone and can't find the microsd directory path Sep 11 20:59:45 there isn't one unless you do one of 1. mount it manually 2. configure an fstab entry to mount it automatically, or 3. install some service that mounts removable media automatically Sep 11 21:00:21 can you guide me on completing this process? Sep 11 21:01:01 I saw a tutorial online on buidling a multi room audio system with Google Home and MPD using the beaglebone black Sep 11 21:02:02 I did every step but got stuck trying to change the directory to my SD card so that Pulse Audio could pull music from the SD card instead of the default directory Sep 11 21:02:10 ah Sep 11 21:02:54 you mean where mpd can pull music from Sep 11 21:03:01 yes sir Sep 11 21:03:18 create an empty directory where you want the mountpoint for the sd card to be, e.g. /mnt/sd Sep 11 21:03:56 I don't know how Sep 11 21:04:07 sudo mkdir -p /mnt/sd Sep 11 21:05:03 i did it and then it asked for the password Sep 11 21:05:23 it's asking for the password of the account you're logged in as Sep 11 21:05:43 i entered it and its back to the debian@beaglebone Sep 11 21:06:05 now add the following line to the /etc/fstab config file: Sep 11 21:06:52 /dev/mmcblk0p1 /mnt/sd vfat uid=mpd,fmask=0133,dmask=0022,nofail 0 0 Sep 11 21:07:33 you can edit the config file e.g. with: sudo pico /etc/fstab Sep 11 21:08:07 i will type it now Sep 11 21:08:51 there is two lines already Sep 11 21:08:55 once you've added that line, it should mount automatically at boot. you can also mount it immediately (without having to reboot) using: sudo mount /mnt/sd Sep 11 21:08:58 should i just type under them Sep 11 21:08:59 yes, don't touch any existing lines Sep 11 21:10:53 i am done typing it Sep 11 21:10:59 exit and save? Sep 11 21:11:13 note that this will give mpd write-access to the card... I don't know if it has any use for that (e.g. for organizing music), but if it doesn't or you don't want it to have write-access to the card, use "uid=debian" instead of "uid=mpd" in that line Sep 11 21:12:02 (the uid option in this line is used to specify which user on the system will "own" the contents of the sd card) Sep 11 21:12:18 after that yes save and exit Sep 11 21:12:49 and then, like I said, sudo mount /mnt/sd Sep 11 21:13:15 I have the link for the tutorial Sep 11 21:13:45 I found it 10 minutes ago, I know how to use google ;) Sep 11 21:13:56 apologies Sep 11 21:14:15 I didn't intend to depreciate your knowledge Sep 11 21:14:43 so I will type sudo mount /mnt/sd Sep 11 21:15:15 yeah, once you've executed that, the contents of your sd card should be accessible at the path /mnt/sd Sep 11 21:16:12 it says wrong fs type, bad option, bad superblock on /dev/mmcblk0p1, missing codepage or helper program, or other error Sep 11 21:16:25 that suggests the sd card doesn't have a FAT filesystem on it Sep 11 21:16:38 format to fat? Sep 11 21:16:42 do you have any idea how the card has been formatted? Sep 11 21:16:55 fat32 i believe Sep 11 21:17:04 well it doesn't seem like it Sep 11 21:17:12 try: sudo file -s /dev/mmcblk0p1 Sep 11 21:17:19 i used a windows laptop Sep 11 21:17:29 i will format to fat32 now Sep 11 21:17:59 nevermind it was ExFAT Sep 11 21:18:46 Should i use a mac to format it perhaps? Sep 11 21:19:02 I'd suggest using fat32 Sep 11 21:19:15 you can use the beaglebone to format is Sep 11 21:19:18 I only had two options on windows Sep 11 21:19:28 NTFS or ExFat Sep 11 21:19:34 I will try to format via ma Sep 11 21:19:38 Mac Sep 11 21:20:30 the way to format it on the beaglebone would be: sudo mkfs.vfat -n MUSIC /dev/mmcblk0p1 Sep 11 21:20:45 ("MUSIC" here is the name given to the disk) Sep 11 21:21:08 formatting using a mac should work to Sep 11 21:21:11 probably Sep 11 21:21:43 it says I can format MS-DOS(FAT) Sep 11 21:21:50 yep that's fine Sep 11 21:23:04 i have reset and mounted Sep 11 21:23:08 btw, note that once the card is mounted you should preferably not just remove it without unmount it first using "sudo umount /mnt/sd" (note that's "umount", not "unmount") Sep 11 21:23:52 I will keep that in mind and save the line Sep 11 21:24:30 anyway, you can now just configure /mnt/sd as music directory in mpd's config Sep 11 21:24:41 as explained in the tutorial Sep 11 21:24:59 Do i put the music straight into the drive or should i put it in a folder? Sep 11 21:25:13 how you organize your music on the sd card is up to you Sep 11 21:26:53 if you configured uid=debian in the fstab line (which in retrospect I should probably have used instead of uid=mpd) then you can also use any SFTP program to transfer files via the network to the sd card while it's mounted on the beaglebone Sep 11 21:28:03 I have the beagleboard connected via ethernet and USB but i tried using the ip address given on the site and it didn't work Sep 11 21:29:02 192.168.7.2 should work if connected via usb to windows or linux, 192.168.6.2 should work if connected via usb to mac or linux Sep 11 21:29:45 when connected via ethernet it'll get an ip address from the router on your network. depending on the router, the beaglebone may be findable using its hostname "beaglebone" Sep 11 21:30:33 using the hostname "beaglebone.local" should work on any mac or linux system regardless of how the beaglebone is connected (usb or ethernet) Sep 11 21:30:52 i will try now Sep 11 21:31:02 it might also work on windows 10+, not 100% sure about that though (I don't use windows) Sep 11 21:31:55 I have windows 11 Sep 11 21:36:23 can i save this chat? Sep 11 21:36:34 https://libera.irclog.whitequark.org/beagle/2022-09-11 Sep 11 21:38:38 https://libera.irclog.whitequark.org/beagle/2022-09-11#1662926202-1662928315; more specifically Sep 11 21:40:58 I unmounted the SD but when I put it back in, I had to mount it through a command. Sep 11 21:41:05 yes Sep 11 21:41:23 There is no automatic way to do this? Sep 11 21:41:31 auto mount? Sep 11 21:42:35 not a super easy way. there exist services that do that for you like udisks but I don't really have experience with them Sep 11 21:43:23 no worries Sep 11 21:43:37 I will just remember to unmount and mount Sep 11 21:43:40 it will however automatically (try to) mount it at startup Sep 11 21:45:17 another option would be to enable file sharing on your music directory on windows and access it directly from mpd, but that would require your windows pc to be available to be able to play music. I'm guessing the intention here is to have a self-contained solution Sep 11 21:46:34 correct Sep 11 22:16:01 Hello again Sep 11 22:16:48 I just launched MPD interface and on the bottom is says ncmpcpp: Timeout while connecting Sep 11 22:17:09 also no active mpd connection Sep 11 22:17:12 I don Sep 11 22:17:24 i don't know what has to be done Sep 11 22:21:16 are you running ncmpcpp on the beaglebone or on another computer? in the latter case you need to provide it with the hostname or ip address of the beaglebone using the --host option Sep 11 22:22:36 beaglebone Sep 11 22:23:26 I am getting the media from the SD card so do i needed to provide it with ip address Sep 11 22:23:35 also how to we happen to do this? Sep 11 22:23:57 then presumably mpd isn't actually running, check with systemctl status mpd Sep 11 22:24:34 ?? Sep 11 22:24:41 I don't know what you meant by that last question Sep 11 22:24:47 i have tried yet Sep 11 22:25:06 I will check the status Sep 11 22:25:26 I was just saying that if you run an mpd client (such as ncmpcpp) on a different computer to control mpd on your beaglebone, you'll need to pass the hostname/ip of the beaglebone to the mpd client Sep 11 22:25:46 if you run an mpd client on the beaglebone itself then that's not needed, it'll default to connecting to mpd on the same device Sep 11 22:26:18 active says failed Sep 11 22:26:31 Failed to start Music Daemon Sep 11 22:26:50 also code=exited, status =1/Failure Sep 11 22:27:16 yeah so something is wrong, e.g. perhaps due to a configuration mistake... you'll need to check the logs to see what's wrong. I can't help you in detail right now, I kinda need my hands available to eat my pizza ;) Sep 11 22:27:33 no worries! Sep 11 22:27:56 Would you like me to chat in ten minutes? Sep 11 22:28:28 or at what time can I reach out again? Sep 11 22:28:45 if you don't mind as I don't want to cause any inconvience Sep 11 22:28:47 I suggest just sticking around, maybe I'll have time later, maybe someone else has time Sep 11 22:28:55 ok Sep 11 22:30:08 if your beaglebone has internet access via ethernet, try: journalctl -u mpd | pastebinit to create a url to share your mpd's logs Sep 11 22:35:01 got it! Sep 11 22:35:40 https://paste.debian.net/1253543/ Sep 11 22:36:11 well, it's reporting a syntax error in your /etc/mpd.conf config file Sep 11 22:37:42 what would you recommend? Sep 11 22:37:59 i would recommend fixing the mistake in the config file :D Sep 11 22:38:14 it looks like you probably forgot a " Sep 11 22:38:46 i don't know how as i have no knowledge about this Sep 11 22:39:46 open up the config file in a text editor like you've done before, check the line mentioned in the error (line 324) Sep 11 22:40:23 it's undoubtedly a line you've edited previously Sep 11 22:41:13 is it the mpd config file? Sep 11 22:41:15 if you can't figure out what's wrong with the config file, share it using: pastebinit /etc/mpd.conf Sep 11 22:41:42 the error is literally telling you which file: "Error in /etc/mpd.conf line 324" Sep 11 22:42:16 apologies as i didnt know Sep 11 22:42:45 lack of knowledge is fine, lack of reading what's right in front of you isn't :P Sep 11 22:43:08 understood sir Sep 11 22:44:18 how do we skip to line 324 Sep 11 22:45:18 using pico? no idea, I don't really have experience using pico, I use a different text editor Sep 11 22:47:07 according to wikipedia, adding +324 as option to pico (e.g. "sudo pico +324 /etc/mpd.conf") should make it start on that line Sep 11 22:50:56 got it Sep 11 22:51:41 i don't see anything wrong Sep 11 22:51:44 looks fine Sep 11 22:52:41 i have the link for the file Sep 11 22:53:18 https://paste.debian.net/1253546/ Sep 11 22:54:34 I do see that line 332 is missing an ending ", and all three auto_output blocks are missing a closing } Sep 11 22:54:42 *audio_output blocks Sep 11 22:55:50 yeah the error is slightly misleading, because it hasn't seen a } yet it thinks on line 324 it's still inside the audio_output block you started at line 319 Sep 11 22:56:43 and because it's still inside an audio_output block, it's expecting a line of the form option "value" which line 324 isn't Sep 11 22:58:50 can i share the revised file? Sep 11 23:00:04 you don't need to ask that, just give the link :P Sep 11 23:00:33 https://paste.debian.net/1253547/ Sep 11 23:00:49 yep that looks better Sep 11 23:01:06 i will try running it again Sep 11 23:03:14 still the same error Sep 11 23:03:22 i will get the error logs once more Sep 11 23:06:47 line 239 this time Sep 11 23:07:29 https://paste.debian.net/1253540/ Sep 11 23:07:40 oh yeah, you uncommented line 239 and 247 but there's no (uncommented) settings in there Sep 11 23:08:40 should i just close the quotes on the lines? Sep 11 23:09:38 remember that lines that start with a # are just comments, they're ignored, so this is what your config file actually looks like to mpd: https://pastebin.com/raw/eYCAfQbD Sep 11 23:09:55 that empty audio_output { } block you've got is not valid Sep 11 23:10:24 so just comment out those two lines (line 239 and 247) Sep 11 23:11:07 i will attempt to try again Sep 11 23:12:32 it worked but i may have forgotten to mount the sd Sep 11 23:12:35 let me check Sep 11 23:14:53 it works but now it isn't importing any music from the sd card Sep 11 23:14:54 note that ncmpcpp probably by default opens in the (initially empty) queue window, the file browser is window 3 (accessed by pressing 3) .. at least it's like that for ncmpc so I assume also for ncmpcpp Sep 11 23:15:07 yes Sep 11 23:15:39 it says empty Sep 11 23:15:56 although i have one song on the sd card Sep 11 23:16:14 what does ls /mnt/sd show? Sep 11 23:17:06 how do i leave the ncmpcpp menu to check Sep 11 23:17:12 q Sep 11 23:18:04 i restarted the board Sep 11 23:18:14 * set_ watches to learn more! Sep 11 23:19:23 no such file or directory Sep 11 23:19:35 uhh what Sep 11 23:19:56 i am confused since we did everything initially Sep 11 23:20:06 i also checked if it was mounted Sep 11 23:20:43 sure you didn't just make a typo? Sep 11 23:20:49 soap operas in mcu instances! Sep 11 23:21:18 set_, can you wait with your usual word salad contributions till some other time? Sep 11 23:21:27 Sure. Yes sir. Sep 11 23:21:33 its ls/mnt/sd Sep 11 23:21:49 Guest48: ls /mnt/sd Sep 11 23:22:06 L or i Sep 11 23:22:15 lowercase L Sep 11 23:22:26 and the space between ls and /mnt/sd is not optional :P Sep 11 23:22:45 its showing my file now Sep 11 23:23:04 let try to run ncmpcpp Sep 11 23:23:51 still says empty Sep 11 23:24:02 should the audio file be a certain format? Sep 11 23:24:26 I mean, it should be a format mpd supports, but I'm pretty sure it supports quite a lot Sep 11 23:24:36 i used mp3 Sep 11 23:24:42 that's definitely fine Sep 11 23:24:56 what seems to be the issue then? Sep 11 23:25:09 the config file looked correct Sep 11 23:25:14 the name was right Sep 11 23:25:21 yeah, I have no idea Sep 11 23:25:42 but do we have to add ls to the path Sep 11 23:25:49 maybe that is a solution? Sep 11 23:25:52 ??? Sep 11 23:26:18 the path was only written as /mnt/sd Sep 11 23:26:29 ls is a command to list files Sep 11 23:27:13 try telling mpd to rescan the music directory using: mpc update --wait Sep 11 23:27:32 I would assume it does that automatically at startup but maybe I'm wrong about that Sep 11 23:27:36 that would be in the central console Sep 11 23:27:41 in the shell yeah Sep 11 23:27:57 there's also a way to trigger that from within ncmpcpp but I don't know the key for that Sep 11 23:30:50 it says command not found Sep 11 23:31:30 install with sudo apt-get install mpc Sep 11 23:31:52 (mpc is a very simple mpd client useful for sending individual commands to mpd) Sep 11 23:32:09 installing now Sep 11 23:32:13 its done Sep 11 23:34:25 tried ncmpcpp and it still shows 0 Sep 11 23:34:39 in its file browser tab? Sep 11 23:34:54 yes Sep 11 23:35:31 then not sure what's going on, sorry... I'm also kinda out of support-giving-energy Sep 11 23:35:58 i did a restart as it solves some common issues Sep 11 23:36:43 Have no fret, I am here for you! Sep 11 23:37:01 set_: please try to not confuse people more than they already are :P Sep 11 23:37:01 Let me go and look up what exactly is going on! Sep 11 23:37:06 Fine. Sep 11 23:37:20 like, I know you mean well, but... Sep 11 23:37:23 @zmatt: If you are too tired, it is going to be an all nighter! Sep 11 23:37:51 I know, I know. Sep 11 23:38:13 I mean well but I revert and octupus stuff. Sep 11 23:38:25 zmatt I apologize for the inconvenience I have caused Sep 11 23:39:12 it is to the highest regard I octopus salad the equation. Sep 11 23:39:13 I am a noob at this stuff and you definitely helped close the gap in my understanding Sep 11 23:39:33 Hey guys, having an issue with a GPS unit with an FTDI driver... used to work fine before and provided a /dev/ttyUSB0 serial port but no longer does on the new units... old driver was "Prolific USB-to-Serial Comm Port" on Win and new one is listed as "Prolific PL2303GC USB Serial COM Port"... just leaving it here in case someone has any insight and Sep 11 23:39:34 ran into the same issue. Thanks! Sep 11 23:39:35 Okay. Fine, I give up. BBL! Sep 11 23:49:40 am I allowed to try help yet? Sep 11 23:51:00 ARM may have a driver...I will research it and see if it is something that comes w/ a driver or not on Win 7/8/10/11? Sep 11 23:51:45 I have looked around for a driver and have just been going around in circles Sep 11 23:52:06 The specific device is BU-353S4 Sep 11 23:54:51 Oh. I found there is a prolificusa site. I am sure they have drivers but I think, from what I read, the older driver may work w/ newer OSs. Sep 11 23:55:05 Let me dig around some more. Sep 11 23:55:55 thanks! As soon as my BB is up, I will copy past the dmesg --follow output from one vs the other Sep 11 23:56:17 Yeppers. prolificusa.com has drivers but further down the page. Not the first one! Sep 11 23:56:35 BB-Flash: Use pastebin.com instead. Sep 11 23:56:46 Do not paste the entire dmesg in this chat, please. Sep 11 23:57:02 ha ha, got it... no I won't I will just grab the end of it :-D Sep 11 23:57:34 Use pastebin.com. Sep 11 23:57:57 Please. Some people get tired of me chatting nonsense. I can only do so much. Sep 11 23:58:15 Speaking of...where did everyone go? Sep 11 23:58:45 This forum is usually this quiet...periodically there is some chatter but otherwise quiet Sep 11 23:58:48 [ 182.320279] usb 1-1: Product: USB-Serial Controller D Sep 11 23:58:49 [ 182.320286] usb 1-1: Manufacturer: Prolific Technology Inc. Sep 11 23:58:49 [ 182.406577] usbcore: registered new interface driver usbserial_generic Sep 11 23:58:50 [ 182.406667] usbserial: USB Serial support registered for generic Sep 11 23:58:50 [ 182.458412] usbcore: registered new interface driver pl2303 Sep 11 23:58:51 [ 182.458534] usbserial: USB Serial support registered for pl2303 Sep 11 23:58:51 [ 182.458846] pl2303 1-1:1.0: pl2303 converter detected Sep 11 23:58:52 [ 182.541310] usb 1-1: pl2303 converter now attached to ttyUSB0 Sep 11 23:59:26 pl2303. Okay. What is the issue? Sep 11 23:59:43 ls -l /dev/ttyUSB0 Sep 11 23:59:58  314.827567] usb 1-1: new full-speed USB device number 4 using musb-hdrc Sep 11 23:59:58 [ 314.976747] usb 1-1: New USB device found, idVendor=067b, idProduct=23a3, bcdDevice= 1.05 Sep 11 23:59:59 [ 314.976766] usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3 Sep 11 23:59:59 [ 314.976775] usb 1-1: Product: USB-Serial Controller Sep 12 00:00:00 [ 314.976782] usb 1-1: Manufacturer: Prolific Technology Inc. Sep 12 00:00:00 [ 314.976789] usb 1-1: SerialNumber: DZBVb106315 Sep 12 00:00:20 Hmm. Sep 12 00:00:28 so the first is the one which works fine... the one with a newer driver does not give a ttyUSB0 Sep 12 00:00:36 Right. Sep 12 00:00:43 The newer driver does not work. Use an older driver. Sep 12 00:00:50 At least, that is the consensus. Sep 12 00:01:25 It's usually plug and play and I have never messed with it Sep 12 00:01:45 If you got search pl2303 Windows 11, there are a bunch of ideas on ms's site. Sep 12 00:02:23 They all lead to the driver at prolificusa.com but the prolificusa people say, and MS's site says, use the driver further down the page. Sep 12 00:02:54 I think it may be the last entry on the page. Sep 12 00:03:11 it actually is a MS thing. Sep 12 00:04:34 the prolificusa site says that it is WHQA or whatever certified. I guess it is okay? Sep 12 00:04:37 Is that relevant for BB too? I was hunting around for an older PL2303 Linux driver to see if I can override what's getting loaded automatically when it's plugged in Sep 12 00:05:13 I think not. It is the driver and how Windows is seeing it, using it, and communicating to it. Sep 12 00:05:23 Now. I could be wrong. I only know so much. Let me check. Sep 12 00:06:38 To my knowledge pl2303 comes with the BB debian OS and that's how I have always gotten a ttyUSB0 when the unit is plugged in. With these newer units, I did not, so I plugged into my Windows PC and saw that there were two different drivers being loaded - pl2303 v/s pl2303gc Sep 12 00:06:42 usb to rs232? Sep 12 00:06:53 Yep that's correct Sep 12 00:07:01 Okay. Sep 12 00:07:21 The FTDI chip in the device handles that automatically Sep 12 00:08:16 So, there is a problem w/ the newer driver says the source of the product. Now, if Linux help, good. I am sure there are drivers for linux but... Sep 12 00:08:54 What's the issue you are facing? Maybe I can offer some help in some fashion... I am no kernel expert and pretty much work on out of the box configuration but have deployed BBs in hundreds of live production environments Sep 12 00:09:11 Nice! Sep 12 00:09:30 No issue here except for GenTooMan having me drop his name around here... Sep 12 00:10:28 Are you using Linux or this bridge to open a terminal? Sep 12 00:10:54 Sorry. Let me better ask that question... Sep 12 00:11:19 ha ha, I was wondering what you meant:-)  I am chatting from my Windows PC, but yeah connected via BB to the USB GPS unit Sep 12 00:11:28 Are you using a Linux OS development env. or are you just trying to terminal into the BBB via the bridge? Sep 12 00:11:39 yeah I am doing a putty into the BB Sep 12 00:11:48 From Win 11? Sep 12 00:11:52 Win 10 Sep 12 00:11:57 Okay. Sep 12 00:12:24 So, I think there is a driver but I do not think the BBB needs this driver to communicate just yet. I am about small steps. Sep 12 00:13:01 yeah BB usually never needed a driver and directly usually gave me ttyUSB0 Sep 12 00:14:15 oh... /dev/ttyUSB0 is for usb. Yes. Or it used to be...have you tried /dev/ttyACM0 or /dev/ttyGCM0 <<< guessing on this last one (sort of). Sep 12 00:14:30 Let me plug in. Sep 12 00:14:49 oh? Sep 12 00:14:56 You do not find /dev/ttyUSB0? Sep 12 00:15:27 Yeah I didn't want to confuse why I described what's going on in Device Manager but there seems to be a correlation... one device when plugged in shows up as pl2303, the other shows as pl2303gc Sep 12 00:15:49 the one with pl2303 when plugged in into BB provides the ttyUSB0, the pl2303gc does not Sep 12 00:15:54 One of those are the updated (supposedly nonworking driver). Sep 12 00:16:03 yep Sep 12 00:16:21 but yeah no way to connect to it, at least I have not figured it as of now Sep 12 00:17:14 Hmm. Okay. Sep 12 00:17:19 Try the header on it. Sep 12 00:17:58 I think it is tty0. Sep 12 00:18:33 Here they are back to back Sep 12 00:18:35  1397.299563] usb 1-1: new full-speed USB device number 5 using musb-hdrc Sep 12 00:18:36 [ 1397.448448] usb 1-1: New USB device found, idVendor=067b, idProduct=2303, bcdDevice= 4.00 Sep 12 00:18:36 [ 1397.448467] usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=0 Sep 12 00:18:37 [ 1397.448476] usb 1-1: Product: USB-Serial Controller D Sep 12 00:18:37 [ 1397.448483] usb 1-1: Manufacturer: Prolific Technology Inc. Sep 12 00:18:38 [ 1397.449871] pl2303 1-1:1.0: pl2303 converter detected Sep 12 00:18:38 [ 1397.457165] usb 1-1: pl2303 converter now attached to ttyUSB0 Sep 12 00:18:39 [ 1404.829389] usb 1-1: USB disconnect, device number 5 Sep 12 00:18:39 [ 1404.839840] pl2303 ttyUSB0: pl2303 converter now disconnected from ttyUSB0 Sep 12 00:18:40 [ 1404.840057] pl2303 1-1:1.0: device disconnected Sep 12 00:18:40 [ 1412.943552] usb 1-1: new full-speed USB device number 6 using musb-hdrc Sep 12 00:18:41 [ 1413.092502] usb 1-1: New USB device found, idVendor=067b, idProduct=23a3, bcdDevice= 1.05 Sep 12 00:18:41 [ 1413.092521] usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3 Sep 12 00:18:42 [ 1413.092529] usb 1-1: Product: USB-Serial Controller Sep 12 00:18:42 [ 1413.092536] usb 1-1: Manufacturer: Prolific Technology Inc. Sep 12 00:18:43 [ 1413.092543] usb 1-1: SerialNumber: DZBVb106315 Sep 12 00:18:46 oh you think it's tty0? Sep 12 00:18:54 No. Sep 12 00:19:13 I think tty0 is the 'FTDI' header on the BBB. Sep 12 00:19:50 I think it gets routed to S0. Sep 12 00:19:53 UART0. Sep 12 00:20:14 so, /dev/ttyS0 should be UART0, right? Sep 12 00:21:16 I have tried ttyS0 Sep 12 00:21:42 So, if you use your USB to FTDI on the header for /dev/ttyS0 and use config-pin p9.whatever uart && config-pin p9.whatever uart, then you can ultimately bash in. Sep 12 00:22:06 That is one idea w/out using the older driver... Sep 12 00:22:21 I mean...USB == PWR, GND, UART! Sep 12 00:22:53 Oh. Sep 12 00:23:09 It is like their driver is connecting and then disconnecting from another driver. Sep 12 00:24:02 ok, I deploy in production environments and there is no scope to change stuff around unfortunately so if it does not work with the plug in USB, I will have to look for some other option and I am looking for it... just that the lead times on different components are months now Sep 12 00:24:30 OTG. Okay, I am getting somewhere. Sep 12 00:24:49 Right. Sep 12 00:25:08 So it looks like PL2303GC is connecting in some other way to BB that it's not leading to registering on ttyUSB0 Sep 12 00:25:25 Yeppers. Sep 12 00:25:40 So, the musb gadget thing has changed a bit. Sep 12 00:26:14 I cannot recall what happened but the whole identity of WiFi and Ethernet has changed, too. Sep 12 00:26:47 So, instead of using connman, people have mitigated to the networkd system for connectivity. Sep 12 00:26:56 That is one aspect. Sep 12 00:27:12 Yeah I am getting somewhere... https://prolificusa.com/product/pl2303gc-usb-full-uart-bridge-controller-gpio/ has a Linux driver that I am downloading Sep 12 00:27:26 Another is usb-gadgets. Sep 12 00:27:42 There was an update somewhere along the lines. Check the forums. That may work. Sep 12 00:28:01 Look for usb-gadgets or something like that... Sep 12 00:28:28 Be careful. Sep 12 00:28:39 I think the driver may ruin your system. Sep 12 00:28:50 So, back up before going forth. Sep 12 00:29:58 No reason to lose valuable data! Sep 12 00:30:06 ok thanks! Looks like there are some linux drivers I can build... I have done this before but not with this device Sep 12 00:30:37 Oh. Sep 12 00:30:41 make or cmake? Sep 12 00:30:44 make Sep 12 00:30:56 This should be good! Sep 12 00:31:17 So, they have a download/install and then a command for make? Sep 12 00:31:28 Check the file, Makefile. Sep 12 00:31:33 Before running make. Sep 12 00:32:10 Linux Kernel 5.5 and above already includes built-in drivers for PL2303G Chip. Sep 12 00:32:11 If you's Linux kernel is v2.6.15 ~v5.4, Sep 12 00:32:11 Please update the PL2303G Linux driver in the following way. Sep 12 00:32:12 1. terminal // open terminal AP. S1.png Sep 12 00:32:12 2. uname -r // found out the nearest Linux kernel version first , S2.png Sep 12 00:32:13 3. make all // make new driver, if you have meet error message during make kernel driver, please send email to us. , S3.png Sep 12 00:32:13 4. sudo cp pl2303.ko /lib/modules/$(uname -r)/kernel/drivers/usb/serial Sep 12 00:32:14     // copy new driver to kernel. S4.png Sep 12 00:32:14 5. sudo gedit /etc/modules // edit modules , S5.png Sep 12 00:32:15 6. pl2303 // add pl2303, save, close modules , S6.png Sep 12 00:32:15 7. reboot // reboot OS , S7.png Sep 12 00:32:16 8. plug in new cable, and then enjoy! Sep 12 00:32:19 sorry to paste walls of text Sep 12 00:32:31 use pastebin.com Sep 12 00:33:10 I am not sure if that will help, number 4. Sep 12 00:33:34 Do you have a large ARM system? Sep 12 00:34:11 So, they do not have a real make file. Sep 12 00:34:19 They have a LKM. Sep 12 00:34:20 Sorry I don't understand what you mean by if I have a large ARM system, the BeagleBone is the only Linux device I use Sep 12 00:34:27 oh. Sep 12 00:34:28 Okay. Sep 12 00:34:32 4.19.94-ti-r42 is my kernel Sep 12 00:34:36 Aw. Sep 12 00:34:38 Okay. Sep 12 00:35:11 I think you need a larger ARM system to build your LKM and then port that to the BBB. Sep 12 00:35:27 So I assume I have to use 4.19_ok Sep 12 00:35:35 Right. Sep 12 00:36:08 Don't think I need an ARM system, I have another WiFi adapter for which I build the driver on the BB itself, it takes some time, like 35 min but it does work Sep 12 00:36:08 Well, at least the same C/C++ and gcc versions, presumably. Sep 12 00:36:19 Wozzers. Sep 12 00:36:25 I guess I am wrong. Sep 12 00:36:58 That is a LKM, right. Loadable Kernel Module. Sep 12 00:37:20 I am just going to use psftp to move the folder over and try to follow the instrucitons Sep 12 00:37:44 Okay. Good luck...I would not do that move. Sep 12 00:38:28 If my device gets bricked, not a big deal, I have about 50 more to destroy :-D Sep 12 00:38:43 You can always reflash it. Sep 12 00:39:07 Why are you using kernel 4.19.x now? Sep 12 00:39:26 Did you build it or was it part of the BBB.io peoples' builds? Sep 12 00:43:48 Long story, I usually use BeagleBone Black Wireless, I am forced to use BeagleBone black regular with the Ethernet port because BBBW is not available for like months Sep 12 00:45:13 I am using 10.3 flasher image which by default installs the 4.19 kernel Sep 12 00:45:45 Right. Sep 12 00:46:40 They have updated images. Who knows? This is why I brought up usb-gadgets or whatever it is called. It may just work. Sep 12 00:46:55 I am not sure, for sure. Sep 12 00:47:57 I am sorry. I thought you had an enormous ARM machine to help w/ building and LKM, rebuilding the kernel, and so on. Sep 12 00:48:14 It is hard w/out those machines. Sep 12 00:48:27 TI and bbb.io has them. Sep 12 00:48:36 I think the x15 is one or let me check. Sep 12 00:50:57 ok, just built the driver (or whatever that's called) and about to test! Sep 12 00:51:31 Nice! Sep 12 00:52:39 Yeppers. The x15 has some capabilities for producing but not like the TI boards but this chat does not accept TI built boards (I think). Sep 12 00:52:41 * GenTooMan accidentally drops set_'s name and can't find it now. Sep 12 00:52:49 Here I is! Sep 12 00:52:55 am. Here I am! Sep 12 00:53:22 * set_ wonders where the slaping fish tails ever went? Sep 12 00:53:49 hopefully the tails are still on the poor fish... Sep 12 00:54:15 I know. We got local yocals here fishing and then stomping on the fish instead of eating them. LA? Sep 12 00:54:52 It made national news on my Google AI bout of a news feed! Sep 12 00:56:10 Oh and when I say accept, "accept" means does not support. (I think). Sep 12 00:58:30 did not work :-( Sep 12 00:58:37 Uh. Sep 12 00:58:45 I know...LKM. Rebuild the kernel. Sep 12 00:59:00 just blame set_ works for me... blaming set that is. Sep 12 00:59:08 I think. Let me go and research this a bit more. Sep 12 00:59:22 * set_ blames himself all the time! Sep 12 00:59:30 Wait. I am set_! Sep 12 00:59:48 Is there a way to upgrade the Linux kernel on the BB to 5.5 and above? Sep 12 00:59:56 It will not go past page 24! Sep 12 01:00:05 Yes. Sep 12 01:00:12 Once I do that, I am sure a lot of other stuff will break :-D Sep 12 01:00:13 Reflash to Bullseye! Sep 12 01:00:18 Right! Sep 12 01:00:26 let me check where is bullseye Sep 12 01:00:37 forum.beagleboard.org Sep 12 01:00:53 Do a search for Bullseye. Sep 12 01:01:02 But...I am telling you. Seriously. Sep 12 01:01:14 I am assuming this is it: Buster IoT (without graphical desktop) for BeagleBone on-board eMMC flashing via microSD card Sep 12 01:01:15 AM3358 Debian 10.3 2020-04-06 4GB eMMC IoT Flasher Sep 12 01:01:24 no. Sep 12 01:01:31 forum.beagleboard.org Sep 12 01:01:45 You can use that one if necessary but it is past its prime. Sep 12 01:02:32 Ok... https://beagleboard.org/latest-images Sep 12 01:02:45 Right. Past...Let me go and check. Sep 12 01:02:57 Third from the top I guess... I am relegated to using the onboard 4 GB eMMC Sep 12 01:03:23 Oh. Try forum.beagleboard.org and then search for Buster or Bullseye! Sep 12 01:04:19 I mess w/ bbb.io images only. Shy over here! Sep 12 01:04:34 They have all the best tools for their boards. Sep 12 01:05:30 If I was a real player, I would get an ARM server (BIG BEEF). But, I am too weak minded. Sep 12 01:05:45 Weekly regreting throwing it out! Sep 12 01:06:00 It was in my hands. Now, I am nothing again! Sep 12 01:06:07 Looking for a Bullseye image that will fit in 4 GB Sep 12 01:06:29 console Sep 12 01:06:49 Unfortunately I don't do much hobby work... our production products are usually in the $3-5MM range so it's serious stuff and it has to work Sep 12 01:07:42 Oh. Sep 12 01:07:51 is mm millimeter? Sep 12 01:07:52 So I am very hesitant to go around noodling into kernels and all that and like to just buy and deploy... these hardware component shortages are killing me though and forcing me to look for different ways to make hardware compatible Sep 12 01:08:01 MM is Million Sep 12 01:08:14 Oh. Sep 12 01:08:19 It is awkward. Sep 12 01:08:45 No worries :-) Sep 12 01:08:51 How can you not own an ARM server w/ bulging characteristics while in this million cha-ching game? Sep 12 01:09:18 I quit. Sep 12 01:09:20 You win Sep 12 01:09:24 ... Sep 12 01:10:03 Research LKM Loadable Kernel Modules and tell me I am wrong...please. Just let me know what I gave up! Sep 12 01:10:16 ha ha, I know... the projects that we provide service for are valued at $3-5B usually... Sep 12 01:10:45 I am going through the forum to see where the page is for the images Sep 12 01:11:20 Here: https://rcn-ee.com/rootfs/bb.org/testing/2022-09-02/bullseye-minimal-armhf/am335x-debian-11.4-minimal-armhf-2022-09-02-2gb.img.xz Sep 12 01:11:20 https://forum.beagleboard.org/t/debian-11-x-bullseye-monthly-snapshots-arm64/32318 is this it? Sep 12 01:11:33 That is arm64. Sep 12 01:11:38 No. Sep 12 01:11:52 thanks! Sep 12 01:11:57 It used to be called console. Now, it is called minimal. Sep 12 01:12:08 Minimal works! Sep 12 01:12:12 I like the ring to it! Sep 12 01:13:56 So I flash this using Balena etcher and hold down the S2 button to boot from the card? Then activate the eMMC flash option to flash it to eMMC on next boot? Sep 12 01:14:21 Forget the S2 button. That is old news. Sep 12 01:14:29 Oh. Sep 12 01:14:32 Hold on. Sep 12 01:15:40 Here: sudo enable-beagle-flasher is the command (I think). I do not use it. Sep 12 01:15:43 I use SD Cards. Sep 12 01:15:56 ok Sep 12 01:16:38 Is this a stable image though? Sep 12 01:17:06 So, yes. flash w/ etcher, boot, sudo enable-beagle-flasher, shutdown -h now or poweroff, then take out the sd card, and boot! Sep 12 01:17:07 Oh. Sep 12 01:17:10 Stable. Sep 12 01:17:15 No clue. It works. Sep 12 01:17:29 Not to the enxtent of the law! Sep 12 01:17:35 Ha. Sep 12 01:17:36 Typo. Sep 12 01:17:41 extent! Sep 12 01:17:58 root@BeagleBone:/home/debian# uname -r Sep 12 01:17:59 5.10.131-ti-r49 Sep 12 01:18:00 root@BeagleBone:/home/debian# cat /etc/debian_version Sep 12 01:18:00 11.4 Sep 12 01:18:10 Looks like it worked... now to check the GPS..... Sep 12 01:18:15 Yes! Sep 12 01:18:36 the 'bullseye' images haven't become "stable" except for AI-64 because we are still in the Cloud9-IDE to Visual Studio Code transition. More testing on VSX is desired, but I think we are about to call it done and switch all our docs from Cloud9 to VSX. Sep 12 01:18:41 ....and DAMN it works! Sep 12 01:18:45 Oh. Sep 12 01:19:10 jkridner: Thank you for jumping in. Sep 12 01:19:15 Now, I know. Sep 12 01:19:28 still, the bullseye images aren't changing very much. just the VSX stuff really. Sep 12 01:20:11 The VSX stuff works for me already! I went to :3000 and it booted right into the source. Sep 12 01:20:26 So am335x-debian-11.4-minimal-armhf-2022-09-02-2gb.img just came out a few days back Sep 12 01:20:51 It is constantly updated until a set one, i am guessing. Sep 12 01:21:27 I would guess that to be true. I have updated images before today. Sep 12 01:23:31 Maybe I not phrasing this question right so apologies in advance... there is no way to upgrade the kernel of Buster IoT to 5.10? Sep 12 01:25:45 Looks like there is no connmanctl on this version Sep 12 01:26:21 BB-Flash: the bullseye images use systemd-networkd as network manager instead of connman Sep 12 01:26:33 and yes you can run kernel 5.10 on buster too Sep 12 01:27:02 Is there any documentation to upgrade the kernel? Sep 12 01:27:42 I generally just use apt to install a specific kernel, but there's also a shell script... let me check Sep 12 01:29:54 thanks! Sep 12 01:30:11 any specific reason you need kernel 5.10 on debian buster? Sep 12 01:30:25 buster is getting kinda old itself Sep 12 01:30:43 I used to use a GPS module which always worked and provided a Serial port ttyUSB0 Sep 12 01:31:11 Now they are shipping with a different driver which is only working out of the box with kernel 5 and above Sep 12 01:32:30 the kernel log snippet you showed seems to suggest the device was being detected and a driver was loaded, but the usb device subsequently disconnected Sep 12 01:33:29 Here I am pasting it again Sep 12 01:33:30  215.075549] usb 1-1: new full-speed USB device number 2 using musb-hdrc Sep 12 01:33:31 [ 215.224779] usb 1-1: New USB device found, idVendor=067b, idProduct=23a3, bcdDevice= 1.05 Sep 12 01:33:31 [ 215.224798] usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3 Sep 12 01:33:32 [ 215.224807] usb 1-1: Product: USB-Serial Controller Sep 12 01:33:32 anyway, if you want to try kernel 5.10, sudo /opt/scripts/tools/update_kernel.sh --lts-5_10 Sep 12 01:33:32 [ 215.224814] usb 1-1: Manufacturer: Prolific Technology Inc. Sep 12 01:33:33 [ 215.224821] usb 1-1: SerialNumber: DZBVb106315 Sep 12 01:33:35 please don't Sep 12 01:33:41 don't paste multi-line stuff into the channel Sep 12 01:33:51 use a paste service like pastebin.com Sep 12 01:33:52 Sorry.. what is pastebin? Sep 12 01:33:59 I know another user mentioned it Sep 12 01:34:08 a website to share snippets of text Sep 12 01:34:25 (there's many of those, pastebin.com is just one example) Sep 12 01:34:36 got it Sep 12 01:36:06 if your beaglebone has internet access you can also use the pastebinit utility, e.g.: dmesg | pastebinit Sep 12 01:38:23 so pastebin generates a link to the wall of text just like imgur? Sep 12 01:38:36 yeah, it's like imgur for text Sep 12 01:38:52 so you can then share that link instead of trying to paste a wall of text into chat Sep 12 01:39:29 thanks! Sep 12 01:45:47 Matt, if I can ask you one thing, https://beagleboard.org/latest-images is where I assume the stable versions are (Recommended debian images). These are what I have always used for our production installations for 3+ years and have been very productive. Are the ongoing more prone to issues etc? Sep 12 01:46:06 Sorry I meant to say the images in forum.beagleboard.org Sep 12 01:47:03 I mean, there will be differences (you already discovered e.g. the different network manager used), and obviously any change comes with the risk of issues Sep 12 01:47:35 at the same time, it's not a good idea to just stick with something forever as it eventually decays into non-maintenance, nor to wait with a transition until the very last moment Sep 12 01:48:02 my suggestion would be to at least start testing with the new images when you have the time, even if you're not going to use them in production yet Sep 12 01:49:22 thanks! Sep 12 01:49:27 https://pastebin.com/9qm5KSBV Sep 12 01:50:16 I was running 4.19.94-ti-r42 I think, it just upgraded it to r73 Sep 12 01:53:42 looks like your update_kernel.sh script is old and doesn't understand the --lts-5_10 option Sep 12 01:54:37 I did flash bullseye to a microsd and can boot from that... will that have an updated script that I can copy back over to 10.3? Sep 12 01:55:30 you should be able to update the script by doing "git pull" in /opt/scripts/ Sep 12 01:55:41 ok Sep 12 01:59:03 yeah I think it's updating now! Sep 12 02:06:55 Matt, you are the best! I was able to update the kernel and the device is now working... really appreciate the help! **** ENDING LOGGING AT Mon Sep 12 02:59:57 2022