**** BEGIN LOGGING AT Fri Mar 06 03:00:04 2020 Mar 06 07:32:05 No one with official https://debian.beagleboard.org/images/bone-debian-9.9-lxqt-armhf-2019-08-03-4gb.img.xz touchscreen not working ?? This is the differece in loaded modules between this and a working one: https://pastebin.com/Wc0CXYEg Mar 06 07:42:46 ti_am335x_tsc and ti_am335x_tscadc modules not loaded, cannot load them manually, no such file or directory.... Mar 06 09:15:12 pretty sure the tsc driver is compiled into the kernel Mar 06 09:15:48 CONFIG_TOUCHSCREEN_TI_AM335X_TSC=y Mar 06 09:15:51 indeed Mar 06 09:17:31 fred__tv: most likely it's an issue with the DT overlay... have you checked the kernel log for errors? Mar 06 09:19:16 where to check ? Mar 06 09:19:23 (shamed) Mar 06 09:19:42 journalctl -k Mar 06 09:21:55 Mar 06 08:41:08 beaglebone kernel: ti_am3359-tscadc: probe of 44e0d000.tscadc failed with error -22 Mar 06 09:22:15 if your bbb has internet access you can share your kernel log using: journalctl -k | pastebinit Mar 06 09:22:52 any lines before that that are a bit more informative about why it failed? Mar 06 09:23:01 amazing..... http://paste.debian.net/1133691/ Mar 06 09:23:54 ti_am3359-tscadc 44e0d000.tscadc: Number of i/p channels more than 8 Mar 06 09:24:41 I'm guessing "i/p" is an asinine abbreviation for input Mar 06 09:25:22 the error indicates the number of touchscreen channels + number of generic adc channels is greater than the total number of adc inputs (8) Mar 06 09:25:34 which is indeed a DT issue Mar 06 09:26:12 are you using a standard cape overlay or a custom one? if the former, which? if the latter, please share its source code Mar 06 09:28:56 Hardware: always the same old 648x480 lcd whose eprom claims to be a LCD7.... Mar 06 09:29:29 (anyway using te same touchscreen pins of an LCD7) Mar 06 09:29:36 that's not what I asked though :) Mar 06 09:30:20 I'm guessing custom Mar 06 09:30:41 ah the overlay ? the one the image come with Mar 06 09:31:05 hmm Mar 06 09:31:50 can you share the output of sudo /opt/scripts/tools/version.sh Mar 06 09:32:55 http://paste.debian.net/1133693/ Mar 06 09:34:23 oh I thought version.sh included the overlays that are loaded Mar 06 09:34:58 in that case... ls /proc/device-tree/chosen/overlays/ Mar 06 09:35:37 BB-ADC-00A0 BB-BONE-LCD7-01-00A2 name Mar 06 09:36:42 ah, bingo... the problem is the ADC overlay being loaded... and evidently being loaded on top of the LCD7 overlay Mar 06 09:36:52 that's definitely a bug in u-boot Mar 06 09:37:14 easy workaround: uncomment disable_uboot_overlay_adc=1 Mar 06 09:37:17 in /boot/uEnv.txt Mar 06 09:44:35 what a stupid... Mar 06 09:44:44 I'm writing an email to report this Mar 06 09:45:20 I always set them as disabled.. Mar 06 09:45:58 (adc audio video wireless emmc, as I'm using white) Mar 06 09:46:19 the audio/video/wireless/emmc ones won't do anything on the white obviously Mar 06 09:46:31 :) Mar 06 09:48:05 Anyway LXQT is actually killing the white..... Mar 06 09:49:05 lol Mar 06 09:53:55 half ram size probably matter.... Mar 06 09:54:32 fred__tv: did it fix the issue btw? Mar 06 09:54:48 surely ! Mar 06 09:55:34 now I have XY axes working negative but it's an issue I already solved in the past... Mar 06 10:00:49 since your cape isn't actually a BB-BONE-LCD7-01-00A2 maybe consider fixing its eeprom and making a correct overlay Mar 06 10:01:42 Once you told me at least to erase eprom content and force an appropriate overlay Mar 06 10:02:10 if it's a custom cape yep Mar 06 10:03:30 unless you want to be able to take it off and have the BBB autodetect that fact... then program a custom identifier into EEPROM to allow to appropriate overlay to be autoloaded Mar 06 10:04:08 but if you're not in the habit of swapping capes around then autodetection isn't needed and forcing a suitable overlay is fine Mar 06 10:05:01 ok , but the inverted XY touchscreen is dued to a wrong adc pin wirinc on the cape ? Mar 06 10:06:28 not sure if you're allowed to swap the wires in the config, but I'm pretty sure you can explicitly specify the axes are inverted Mar 06 10:08:47 or maybe you can swap them around, not sure Mar 06 10:09:20 in debian 8 i simpli added an "option calibration" to /usr/share/X11/xorg.conf.d/10-evdev.conf but with this 9 it works no more....:( Mar 06 10:09:30 with "XY axes working negative" you mean both the X-axis and the Y-axis are individually inverted right, i.e. as if it's rotated 180 degrees? Mar 06 10:09:40 you should never modify stuff in /usr/share Mar 06 10:09:55 the place for customizing X11 configuration is /etc/X11/xorg.conf.d/ Mar 06 10:10:03 ah Mar 06 10:10:17 /usr/share is "owned" by the package management system Mar 06 10:10:36 moving pointer up it goes down (and left/right) Mar 06 10:11:45 and no xorg.conf.d directory under /etc/X11 .... Mar 06 10:11:55 you can just create it Mar 06 10:12:22 any files there will override files in /usr/share/X11/xorg.conf.d/ if they have the same name Mar 06 10:13:03 though typically you wouldn't override the generic evdev.conf but instead just add a custom snippet that matches your touchscreen Mar 06 10:13:45 anyway I don't think configuration of evdev has itself changed... but it's possible that the libinput driver is being by default instead of the evdev driver Mar 06 10:14:10 so mods to evdev are ignored... Mar 06 10:14:45 which you could undo by 'sudo apt-get install xserver-xorg-input-evdev' and 'sudo apt-get purge xserver-xorg-input-libinput' Mar 06 10:15:58 or you could adapt your configuration to libinput, which would have the benefit of also working for other libinput applications (e.g. wayland), but I seem to recall configuring touchscreen calibration was not extremely clear, or at least lacked convenient tools Mar 06 10:16:32 ideally touchscreen configuration would be done in DT, but I don't know if that's possible Mar 06 10:17:50 trying to apply to libinput..... Mar 06 10:19:58 Do you have skill in intel i.mx51 ? https://ibin.co/5EfbwMdeah2h.jpg Mar 06 10:20:20 libinput touchscreen calibration can either be done in an udev rule (setting ENV{LIBINPUT_CALIBRATION_MATRIX}) or using Option "CalibrationMatrix" in an xorg.conf.d snippet Mar 06 10:20:33 I have zero experience with any i.mx Mar 06 10:21:10 ok Mar 06 10:23:25 no luck in modifying 40-libinput.conf (that contains the same as 10-evdev.conf) Mar 06 10:27:21 trying to work with transformation matrix Mar 06 10:35:17 transformation matrix affects behavior, but it's very tricky to understand..... Mar 06 10:37:48 try -1 0 0 0 -1 0 0 0 1 Mar 06 10:39:42 actually or -1 0 1 0 -1 1 0 0 1 Mar 06 10:40:50 is there a vay to apply without reboot ? Mar 06 10:41:33 yeah with xinput Mar 06 10:41:44 maybe this is of use: https://wiki.archlinux.org/index.php/Talk:Calibrating_Touchscreen#Wrapper_around_xinput_calibrator Mar 06 10:47:25 -1 0 1 0 -1 1 0 0 1 does the trick ! great! Mar 06 10:48:15 makes sense, so the transformation is applied to coordinates scaled to 0-1 Mar 06 10:52:22 a big thanks a s usual.... Mar 06 12:47:04 zmatt: how can I use a white with display as minimal webpage browser without install heavy graphic environment that chokes the board ? Mar 06 14:02:51 what is the "SMC" ARM (A9?)v7 instruction (Secure Monitor Call) all about? Mar 06 14:03:00 it looks like SMC generates a special type of exception which a) places the processor in a special privilege level (TBD), and b) invokes the routine pointed to by the MVBAR (Monitor Vector Base Address Register). it does this synchronously. Mar 06 14:03:06 is this correct? Mar 06 14:05:06 zmatt: hey man, how are you? Mar 06 14:05:43 (probably asleep...) Mar 06 15:11:37 it was only 3 pm his time probably working he's probably still working too. Mar 06 15:11:54 yates: it calls into the secure monitor Mar 06 15:12:32 yates: the secure monitor services available on GP TI SoCs are very limited, and they're documented in the TRM Mar 06 15:14:32 iirc on the am335x the only secure monitor calls are to write certain special registers that can only be written in secure privileged state Mar 06 15:26:29 zmatt: couldn't you write such registers in supervisor mode? i thought supervisor mode gave you access to everything Mar 06 15:26:52 do you mean certain p15 coprocessor registers? Mar 06 15:27:05 (which again are writable in supervisor mode, i think) Mar 06 15:27:07 yeah, smc #1 writes r0 to a cortex-a8 special-purpose register that depends on r12: r12=0x100 writes to the Auxiliary Control Register, r12=0x101 writes to the Non Secure Access Control Register, r12=0x102 writes to the L2 Cache Auxiliary Control Register Mar 06 15:28:25 no there's a privilege level above the "non-secure"/"public" supervisor state, in fact above hypervisor state (when applicable) Mar 06 15:28:38 secure privileged state Mar 06 15:28:47 here is the output from IDAPRO with the SMC - it doesn't appear to use r12 at all. http://paste.ubuntu.com/p/845dxm3kM9/ Mar 06 15:29:18 oh I thought you were talking about code for the beaglebone Mar 06 15:29:26 he he... Mar 06 15:29:53 what makes you think it's not? :) Mar 06 15:30:08 because this is code that sets up the secure monitor Mar 06 15:30:20 the second instruction can only be executed in secure privileged mode Mar 06 15:30:54 ditto that MCR Mar 06 15:32:28 you are right, it is for a Marvell SoC Mar 06 15:34:13 zmatt: but doesn't some such code also have to be executed for a beaglebone, in the early startup stages? Mar 06 15:34:23 yes, in secure rom Mar 06 15:35:15 i guess this Marvell SoC (which may be custom) allows the setup from FLASH. Mar 06 15:36:40 anyway, the SMC acts like an SVC, except it switches to secure monitor mode and jumps to MVBAR+0x08 Mar 06 15:37:08 what is "SVC"? Mar 06 15:37:15 supervisor call Mar 06 15:37:29 how userspace invokes the kernel Mar 06 15:37:48 you mean the linux kernel? Mar 06 15:38:13 the processor doesn't care what OS you're running :P Mar 06 15:39:44 is the am335x a cortex a8? and isn't it multicore? Mar 06 15:40:15 it's a cortex-a8, which doesn't exist in multicore version Mar 06 15:40:27 only uniprocessor Mar 06 15:41:40 i believe we're running an a9, 4-core. this code is one "thread", for core processor 0. it has processor affinity register in the p15 Mar 06 15:56:44 apparently the "secure monitor" paradigm is a component of arm's "TrustZone"? Mar 06 15:58:24 correct Mar 06 16:01:09 it's kind of a bit like virtualization, except with only 2 VMs (secure and non-secure), with many special registers simply having two copies, and things like cache lines and bus transactions are tagged with a single bit indicating whether the request is from secure world or public world Mar 06 16:01:52 that correlates with some stuff i've read Mar 06 16:02:21 smc calls seem to be very similar to hvc calls Mar 06 16:02:39 in a sense yeah Mar 06 16:02:48 and both are referenced in the doc SMC CALLING CONVENTION System Software on ARM ® Platforms Mar 06 16:08:44 zmatt: i'm confused about "secure privileged state" versus "supervisor mode". what can't you do in (non-secure) supervised mode that you can in secure privileged state? Mar 06 16:11:16 ...and how do you switch between "secure supervisor" and "non-secure supervisor" modes? Mar 06 16:18:53 switching between them is handled by the secure monitor (which is essentially the hypervisor for trustzone) Mar 06 16:20:07 for the most part, what secure state can do that non-secure state can't is access stuff in secure state :P in the same sense as what one VM can't do is access stuff of another VM Mar 06 16:20:29 so is it basically like this: if you have some code you want to secure, allow it to be run only via SMC? Mar 06 16:20:44 hokay.. Mar 06 16:21:22 but for special cpu registers it varies: some are banked between the two states, some are only accessible in secure privileged state.. it just varies, you need to check the docs Mar 06 16:22:21 and since the secure vs public world bit is included in bus transactions, the SoC can also impose restrictions... but that will vary per SoC Mar 06 16:23:15 oh shit.. Mar 06 18:49:44 is there a sorted list of arm opcodes (probably just the first byte would be good) and their associated assembly mnemonic? Mar 06 18:50:15 (not just first byte) Mar 06 18:50:49 for arm encodings, not thumb Mar 06 18:50:50 it's not that simple, there isn't a neat "opcode" field... but the ARM architecture reference manual includes a decent overview iirc Mar 06 18:51:11 ah, the legacy encoding Mar 06 18:51:21 yes Mar 06 18:51:38 (but yes, it has overview for both) Mar 06 18:57:15 what instruction is "E1 A0 F0 0E"? Mar 06 18:58:39 mov pc, lr Mar 06 18:58:51 thanks to https://onlinedisassembler.com/odaweb/ Mar 06 19:41:25 how is doing "preemt RT44" on beaglebone AI ? on raspberry RT44-->70us , the old one RT40-->350us ,what an increase... Mar 06 19:42:17 RTfre: what is "preemt RT44" ? Mar 06 19:42:45 also, I've heard the -rt kernel is kinda broken on bbai (due to the broadcom wifi driver) Mar 06 19:43:52 rt.wiki.kernel.org preemt RT Mar 06 19:44:03 realtime kernel hack Mar 06 19:44:40 I know what preempt rt is but what do you mean by "RT44" Mar 06 19:45:04 its a version Mar 06 19:45:40 no it's not. kernel versions have three components, e.g. 4.14.108 Mar 06 19:46:03 version of the patch. Mar 06 19:47:27 latency isn't going to depend on just the hardware and the version of the patch, it will also depend on kernel version, kernel config, and system workload Mar 06 19:47:56 and how processes and kernel threads have been prioritized Mar 06 19:49:33 the RT patch (try to)replaces parts of kernel where it takes to much time... Mar 06 19:49:53 I know what the rt patch does Mar 06 19:50:26 with target to minimize the time.. with RT44 there was a big step... Mar 06 19:51:14 that's a gross oversimplification of what the goal of the rt patch is... on average the rt patch will make things slower Mar 06 19:51:15 sure there is xenomai.. but this is other thing. Mar 06 19:52:08 i have done some tests.. but on rpi.. Mar 06 19:52:57 dependencies i found: Mar 06 19:53:17 MIN AVG MAX <--- linux kernel RT | graphics | cpu f | isolcpu | app start then close Mar 06 19:53:37 ___ ___ 351 <--- rpi 4.19.97 RT40 | 1920x1080 60Hz | 600…1500 | NO | chromium libre-calc Mar 06 19:53:40 stop Mar 06 19:53:46 don't paste multi-line stuff into chat Mar 06 19:53:57 where to? Mar 06 19:54:29 if you feel the need to share this information (although I'm not sure why you think this is useful to share), use a paste service like pastebin.com Mar 06 19:56:53 https://pastebin.com/wZ9bbdCB Mar 06 19:57:59 I have trouble imagining on what system it would make sense to have an rt kernel and then run chromium or libre-calc Mar 06 19:59:39 ist just some app load... you could change the apps for your propose Mar 06 19:59:51 "To get the most accurate latency measurements, Cyclictest should be executed while the evaluated system is running a load as similar as possible to the RT application for which the system is intended." Mar 06 20:00:11 Just to say this on my end, the BBAI cannot run the RT kernel as of now (from my experience). But... Mar 06 20:00:24 yeah I mentioned that Mar 06 20:00:27 I know little compared to seasoned knowledgers. Mar 06 20:00:33 I saw it. I was just reiterating. Mar 06 20:00:37 without these apps its nearly the same.. in RT40 its increased significant, but i dont log this.. Mar 06 20:01:01 please specify the full kernel version, just "RT40" or "RT44" is meaningless Mar 06 20:03:13 is this RT ?: http://gfnd.rcn-ee.org:8080/job/beagleboard_kernel_builder/job/4.19-rt/ Mar 06 20:04:23 pretty much non-rt kernel package built has a corresponding rt kernel package Mar 06 20:04:28 *pretty much every Mar 06 20:05:51 its the RT(version44) of the patch for the linuxkernel(version4.19.106) Mar 06 20:06:48 heres link to that: https://mirrors.edge.kernel.org/pub/linux/kernel/projects/rt/4.19/patch-4.19.106-rt44.patch.xz Mar 06 20:07:48 I don't know what you're trying to communicate or why Mar 06 20:07:53 I know what the rt patches are Mar 06 20:08:26 i downloaded kernel for the platform, then patched and installed it, then ive done cyclictest and was delighted by the increase.. Mar 06 20:09:23 my question is: if the increase is on BB too ? Mar 06 20:09:25 like I said, nearly all the kernel packages for beaglebones and beagleboards have corresponding -rt packages, so if you want an rt kernel all you have to do is apt-get install it Mar 06 20:10:18 ah ok also rt44? Mar 06 20:12:40 e.g. this is the current list of -ti kernel packages in buster: https://pastebin.com/EyKa6T75 .. it doesn't explicitly show the -rt version, but it looks like that's usually pretty much narrowed down by the kernel version Mar 06 20:13:47 hm seems to be RT36 @ 4.19.94. ok will be soon i think.. Mar 06 20:14:25 you really don't read what I said huh Mar 06 20:15:31 last one i found rt36 Mar 06 20:16:06 4.19.94-ti-rt-r36 uses the 4.19.94-rt39 patchset, which is the latest rt patch for that kernel version Mar 06 20:16:06 wheres the list from? search where? Mar 06 20:16:16 apt-cache search Mar 06 20:16:37 ok no 44... will be soon i think.. Mar 06 20:17:15 ok thanks. Mar 06 20:17:30 4.19-ti isn't the default kernel series anyway on the bbai, pretty sure that's still 4.14 Mar 06 20:18:18 (4.14.108-ti-rt-r127 uses the 4.14.106-rt56 patch) Mar 06 20:20:28 ti := texas instruments? Mar 06 20:20:33 yes Mar 06 20:21:09 -r127 version of ti? Mar 06 20:21:38 -r127 just indicates the version number of the kernel package build by rcn (including e.g. the patches and kernel config) Mar 06 20:25:20 ok so for 4.14 if there are the same improvements, today is patch-4.14.170-rt75.patch.xz 05-Mar-2020 Mar 06 20:31:15 are there scripts for self building BBAI linux × RTpatch ? Mar 06 20:33:43 https://pastebin.com/eLhrp1Hg notes on building customized kernel packages... it looks like the rt patch used is controlled by one of the variables in version.sh Mar 06 20:47:50 Hi, I'm new here. Does anyone know how to install and run websocket.io? Mar 06 20:48:21 I always get errors trying to run anything with it Mar 06 20:48:38 Hello! Mar 06 20:48:58 I am sort of new to websocket.io. Let me go check it out. Mar 06 20:49:08 it does sound familiar. Mar 06 20:50:01 is it this idea: https://socket.io/docs/? Mar 06 20:50:51 The websocket.io on npm is eight years old since the last update. Mar 06 20:50:54 It may be out of date. Mar 06 20:51:07 It may be out of date w/ dependencies and the like. Mar 06 20:51:22 I remember it from a book I once read. Mar 06 20:52:03 brb, give me 30 minutes. The lady sold me the incorrect ink. Back to the Store! Mar 06 21:00:44 for websockets in nodejs, use the "ws" package Mar 06 21:01:30 z4tech: websocket.io is thoroughly obsolete Mar 06 21:12:54 Sorry I made a mistake, I'm actually installing socket.io Mar 06 21:13:05 And it gives me a lot of errors Mar 06 21:13:30 no matter I use sudo or not Mar 06 21:15:04 using sudo for installing an npm package into your nodejs project makes absolutely no sense and probably means you now have a permissions mess Mar 06 21:16:49 @zmatt is right. He told me many times about permissions and using sudo for installing outside of the Debian apt package manager. Mar 06 21:16:57 I tried both npm install socket.io and sudo install socket.io -g Mar 06 21:17:07 Permissions, permissions, permissions. Mar 06 21:17:16 Let me see if I can figure out something. Mar 06 21:17:19 set_ no permissions for you! Mar 06 21:17:29 Fine. Can I get soup? Mar 06 21:17:35 I just installed it on a beaglebone and it installed without any problem... more importantly, neither it nor any of its dependencies were binary packages, so that means that there's nothing that could make it not install on a beaglebone Mar 06 21:17:42 set_ sure just no permissions. Mar 06 21:17:53 Fine. Mar 06 21:18:03 I just plugged in. I am booting. Mar 06 21:18:13 socket.io, right? Mar 06 21:18:20 Not websockets? Mar 06 21:18:40 Just triple checking. Mar 06 21:18:43 z4tech: "gives me a lot of errors" is not a useful problem description, noone can help you with that Mar 06 21:18:59 use a paste service like pastebin.com to share the actual errors Mar 06 21:22:25 also, attempting to globally install packages with npm install -g sounds like a great way to globally break nodejs :P Mar 06 21:22:33 *with sudo npm install -g Mar 06 21:22:38 Here: https://pastebin.com/bK2W0Tmw Mar 06 21:22:54 never put "sudo" before "npm" Mar 06 21:23:06 Can I say, NEVER! Mar 06 21:23:11 Just playin'. Mar 06 21:23:38 You may need .json. Mar 06 21:23:40 set_: this is just saying you have a package.json file in your homedir that doesn't contain valid json Mar 06 21:23:50 Oh. Mar 06 21:23:51 Okay. Mar 06 21:24:08 also doing this in your homedir is a bad idea, make a directory for it Mar 06 21:24:18 Right, got it. Mar 06 21:24:29 I never heard more "truer" words. Mar 06 21:24:42 with some minimal package.json containing e.g. { "name": "foo", "version": "0.0.1" } Mar 06 21:24:53 Some of the deps. are outdated. Mar 06 21:24:59 Oh. Okay. Mar 06 21:25:16 z4tech: Are you still trying? Mar 06 21:27:48 https://pastebin.com/2eQcSPC1 is my next step. Mar 06 21:27:54 Issues galore but a lot installed. Mar 06 21:28:53 ehm, the errors have nothing to do with socket.io and everything to do with all the other crap you have in the same directory Mar 06 21:29:06 Ha. Maybe so. Mar 06 21:29:20 not "maybe", it clearly shows it's giving an error on building the roboticscape module Mar 06 21:29:39 Oh. Okay. Mar 06 21:29:43 Let me try another route. Mar 06 21:30:12 just make a new directory, create a minimal package.json in it like I just said, then do "npm install socket.io" Mar 06 21:30:44 z4tech: Did you see what was just typed out? Mar 06 21:30:52 That may be a cure for now. Mar 06 21:31:05 I was talking to you, not to him Mar 06 21:31:15 I know but he can use it too. Mar 06 21:31:21 He/she/whomever. Mar 06 21:31:41 I have no help to offer him since he hasn't given any details of what problem he's encountering Mar 06 21:32:01 what I just said was specifically to you, for the problem you're having Mar 06 21:32:32 Oh. I just figured I would jump in to try to pitch in. Oh and thank you. I now know that running a ton of stuff in my home/debian dir. just creates more issues. Mar 06 21:34:18 That is what I am talking about. Mar 06 21:34:25 Sorry. Mar 06 21:34:31 Wrong timing. Mar 06 21:36:07 https://pastebin.com/cbUrekUc is what happened when I made a dir, adding the package.json, and then npm install socket.io. Mar 06 21:36:45 Capitalize .json to .JSON? Mar 06 21:37:44 no Mar 06 21:37:57 Okay? Mar 06 21:38:04 What then, you say? Mar 06 21:38:05 just copy-paste what I typed, without mistakes Mar 06 21:38:32 I do not know how to C & P in this chat. Mar 06 21:39:04 you typed "name:" instead of "name": Mar 06 21:39:11 Oh. Mar 06 21:39:12 Okay. Mar 06 21:39:26 You are right. Yikes. Mar 06 21:40:45 It is installing right now. Brb. Mar 06 21:42:34 https://pastebin.com/hfi5iimp is it updated w/ it in a new dir. and w/ npm install socket.io. Mar 06 21:43:33 should I install the unmet deps? Mar 06 21:43:44 what do you mean? Mar 06 21:43:50 npm will automatically install all necessary deps Mar 06 21:43:59 It did not. Mar 06 21:44:02 Look at my paste. Mar 06 21:44:13 oh it's lying, those aren't required, they're optional Mar 06 21:44:27 but you can install them if you want Mar 06 21:44:36 No issue. Mar 06 21:45:00 I do not even care. I just wanted to help some random person/bot that needed support or the opposite. Or something. Mar 06 21:45:11 I just want to pitch in if I can help. That is all. Mar 06 21:45:41 I thought you and n3lglkjt needed help and someone to supply the work. Mar 06 21:45:46 you did help... you showed that socket.io just installs perfectly fine out of the box with whatever node version you have, which I presume is still the default node version Mar 06 21:45:59 Now it is. Mar 06 21:46:05 I erased it and reinstalled it. Mar 06 21:46:14 Do you want to know what version? Mar 06 21:46:28 so whatever problem z4tech is having is entirely an issue with their setup Mar 06 21:46:29 v6.17.0 Mar 06 21:46:40 lol wow that's ancient Mar 06 21:46:44 Oh. You are right but why get me to do it? Mar 06 21:46:47 Oh? Mar 06 21:47:02 I had used the package manager apt to get it. Mar 06 21:47:09 I did not use the deb. source. Mar 06 21:47:10 ehm, I didn't, you chose to do this yourself Mar 06 21:47:16 You are right. Mar 06 21:47:19 You win. as usual. Mar 06 21:47:50 sudo apt install npm nodejs Mar 06 21:48:07 I can probably upgrade it if necessary. Mar 06 21:48:08 anyway, socket.io installs fine with v6.17.0 and it installs fine with v12.15.0 (which is what I'm using) so I think it's safe to say the package is just fine Mar 06 21:48:30 Yep. Now, we need to learn how to utilize it, right? Mar 06 21:48:41 and z4tech hasn't responded since I asked for details, so since he evidently stopped caring it's about time we do so too Mar 06 21:48:49 Fine. Mar 06 21:49:00 OH well. I thought this was a socket.io party for the BBB? Mar 06 21:49:03 I am too wrong. Mar 06 21:50:07 I have been finding a bunch of animation projects lately on github. Stuff for lettering, stuff for moving letting, and so on. Mar 06 21:50:26 letting = lettering Mar 06 21:52:53 z4tech: Make a directory, add the package.json version of your choice like @zmatt typed, and then npm install socket.io. Mar 06 21:52:56 There. I did it! Mar 06 21:53:04 Help, help, help. Mar 06 22:01:31 Well, @zmatt did it. I just lollied and then gagged. Mar 06 22:03:03 why are you telling this to thim? what makes you think it is relevant? Mar 06 22:03:25 when I just said that I said this to you to solve _your_ problem Mar 06 22:03:33 Oh. I just wanted you to have credit. Sorry. Mar 06 22:03:55 It is not my issue. I am not z4tech. Mar 06 22:04:11 it was your issue, the one you showed via pastebin Mar 06 22:04:43 I just wanted to supply z4tech w/ some works but you are right. I got help w/ knowing I should make dir. when attempting to install specific libraries. Mar 06 22:04:46 Now. I know. Mar 06 22:04:46 I have no idea what issue z4tech has since he's been silent for nearly an hour Mar 06 22:05:12 Fine. I thought this would help him/her since he/she decided to at least ask a question. Mar 06 22:05:14 if I thought it was a useful suggestion for him, I would have suggested it to him Mar 06 22:05:21 Sorry guys, I guess I just mess up with the could9 too Mar 06 22:05:22 oh. Mar 06 22:05:23 Okay. Mar 06 22:05:30 z4tech is back! Mar 06 22:05:31 I reflash the image Mar 06 22:05:44 anyway, I've got other things to do, afk Mar 06 22:05:57 Ciao good night too Mar 06 22:05:59 z4tech. Read the last hour and you will figure it out. Mar 06 22:06:02 Thank you for all the advice! Mar 06 22:06:05 Or... Mar 06 22:06:09 oh. Okay. Mar 06 22:06:23 Mars, here I come! Mar 06 22:07:28 you're talking to the Roman God of war now? Mar 06 22:09:06 No. Mar 06 22:09:15 I am moving to Mars to get more candy bars. Mar 06 22:10:06 I think I learned something nice today. Thank you all! Mar 06 22:23:00 I just looked in my node_modules dir. at /home/debian/node_modules and... Mar 06 22:23:09 There are over 75 dir. Mar 06 22:23:14 Yikes. Mar 06 22:23:33 I see why orderliness in communication is necessary. Mar 06 22:32:06 Sorry guys, problem still not solved. Mar 06 22:32:29 I tried to install socket.io in a folder: https://pastebin.com/RZ8wRTdb Mar 06 22:33:11 And run a very simple app: https://pastebin.com/Macmsihf Mar 06 22:33:56 It gives me the same error: https://pastebin.com/FWzZ1RRS Mar 06 22:34:04 Hello. Mar 06 22:34:33 When you need to supply the full/absolute dir, add it, i.e. not just ./index.html. Mar 06 22:34:38 That might help when you read files. Mar 06 22:35:13 Let me have a try Mar 06 22:36:02 It's still the same error... Mar 06 22:36:33 I think it might be the version of node and npm are outdated with the Flash image Mar 06 22:36:34 So, /home/debian/var/lib/cloud9/Test/index.html? Mar 06 22:36:40 Oh. Maybe. Mar 06 22:36:47 node v6.14.4 Mar 06 22:36:51 cloud9 is buggy. Mar 06 22:36:56 That is my version too. Mar 06 22:37:05 npm 3.10.10 Mar 06 22:37:23 Did you try right outside your /home// dir? Mar 06 22:37:44 Let me give it a shot Mar 06 22:37:59 So, try /home/debian/Test/ instead of /home/debian/var/lib/cloud9/Test/ Mar 06 22:39:08 But... Mar 06 22:39:17 You may need to remove it and then reinstall it. Mar 06 22:39:26 In the /home/debian/Test/ dir. Mar 06 22:40:03 Same error.. Mar 06 22:40:24 Let me try and get back to you. Is that okay? Mar 06 22:41:03 z4tech: Did you try your package.json file too? Mar 06 22:41:06 Thx! Mar 06 22:41:33 say tanx isn't a bad idea. :D Mar 06 22:41:49 { "name": "test", "version": "1.0.0", "description": "", "main": "index.js", "scripts": { "test": "echo \"Error: no test specified\" && exit 1" }, "author": "", "license": "ISC", "dependencies": { "socket.io": "^2.3.0" }} Mar 06 22:42:00 Nothing special Mar 06 22:42:29 Okay. If that will work, go for it. Let me try your source and ideas and then I can get a better idea on what is happening. Mar 06 22:42:30 I am trying to update my node and npm, but it ruins the cloud9 everytime Mar 06 22:42:39 Clou9 is buggy. Mar 06 22:42:57 As of now, Cloud9, is buggy. It is having trouble w/ what people want to do w/ it. Mar 06 22:43:29 I'll try update the node and npm and use ssh to see whether it works Mar 06 22:45:46 Okay. Mar 06 22:46:09 I am trying to update too. I will let you know what happens. I will show you the tree soon. Mar 06 22:48:06 yes, cloud9 is bluggy. Kind of frustrating as it's supposed to be able to help people. instead it hinders. I've had trouble with it because it's pythonic interface rhymes with succotash. Mar 06 22:48:52 Ooh...la la. Mar 06 22:49:53 https://github.com/socketio/socket.io/tree/master/examples/chat, look here for the basic example of the tree and how it is put together. Mar 06 22:50:01 I will try another way in the meantime. Mar 06 22:52:43 z4tech: You need to supply a port and the 3000 port is used by Cloud9. Mar 06 22:52:46 That is a major issue. Mar 06 22:53:05 Unless you have two boards. Mar 06 22:53:14 and cancel cloud9 on one of them. Mar 06 22:55:28 I use 8080 port Mar 06 22:56:20 The clould9 on board might be obsolete so it doesn't support newer version of node Mar 06 22:56:39 Yea. Maybe? But... Mar 06 22:56:43 Oh. Okay. Mar 06 22:56:59 8080 works but it will take some time for me to figure out exactly what is the issue here. Mar 06 23:04:27 Okay. Mar 06 23:04:59 You are going to have to find, literally, where ...options is located, and change it. Mar 06 23:05:22 Right/ Mar 06 23:05:42 It is in websocket.js for me. Mar 06 23:05:51 Line 347 Mar 06 23:06:14 I cannot tell you that is a cure but it is a start from what socket.io people are saying. Mar 06 23:06:22 but...who knows? Mar 06 23:06:43 does it say "options here" ? :D Mar 06 23:06:50 Nope. Mar 06 23:06:52 Sure does not. Mar 06 23:07:17 ah forget javascript obviously self documents (cough) Mar 06 23:07:28 I know. It is difficult. Mar 06 23:08:49 Actually it's not, the people who use it are difficult. It's a very serious problem with bad programming practice. A lot of the "script" kiddies grew up learning terrible habits. So .js code tends to borderline consistently on disastrous to fix. Mar 06 23:10:07 so I suggest you learn from others mistakes and not repeat your code being difficult to read? Mar 06 23:10:35 You can get rid of all the ...options and find where ...options are located. Mar 06 23:10:36 But... Mar 06 23:11:02 I do not know as of now exactly what the ...options description does or how it will affect the rest of the source. Mar 06 23:11:12 Off to github! Mar 06 23:13:16 I could not find it in the API. Mar 06 23:17:37 maybe it is just those three ".,s", i.e. ...? Mar 06 23:17:59 Cough. Those three dots. Mar 06 23:18:00 No? Mar 06 23:19:47 Every error I get is those three dots, i.e. which basically means etcetera or whoever you spell that word. Mar 06 23:19:50 Right? Mar 06 23:20:50 it's called an ellipsis meaning the same as "more" :D Mar 06 23:24:13 Ok, I tried to install node v13 and npm 6 and everything's good now Mar 06 23:24:26 What? Mar 06 23:24:28 Nice! Mar 06 23:24:34 Thank you very much! set_ Mar 06 23:24:37 Okay. Mar 06 23:24:50 I barely believe I did anything but yes, you are welcome. Mar 06 23:29:34 The dots were getting me in the source. I changed everything. Now, time to change it back. Mar 06 23:43:48 Erasing the npm and node stuff now. Yikes. Mar 06 23:47:13 Man, you guys solve issues easily. Sheesh. Mar 06 23:47:30 I just dig in their source and change things unless I give up. Mar 06 23:48:18 set_ Got my mini wifi adapter working. Mar 06 23:50:25 Strange thing. I initially setup a service then changed it to another. Now they are both listed in connmanctl one as *A and one as *AO. How do I get rid of the wrong one? Mar 06 23:53:40 Nice. Mar 06 23:53:53 No clue. Mar 06 23:54:01 I could not even get it up and running yet. Mar 06 23:54:11 After tons of years, I have failed at that idea. Mar 07 00:06:21 But...congratulations! Mar 07 00:13:28 Well it won't come up now Mar 07 00:22:58 set_ I wrote up a cheat sheet for setup. Want a copy? Mar 07 00:26:57 Yes. Mar 07 00:27:00 Please. Mar 07 00:32:40 Check it here : https://pastebin.com/LDQXBZwb Mar 07 00:33:57 Yep. Mar 07 00:34:17 That is what i know too. I usually do not need the technologies command under connmanctl, though. Mar 07 00:35:45 If you use tech... after setup it will look different. Mar 07 00:36:15 Okay. Mar 07 00:36:28 I will attemp it soon. Mar 07 00:36:35 attempt it. Sheesh. Mar 07 00:36:47 One or two entries will go from talse to true Mar 07 00:36:59 Hmm. Mar 07 00:37:10 Does it hold after a reboot? Mar 07 00:37:35 I have to redo whole mess I managed to get BBB in Emergency boot mode! Mar 07 00:38:02 Dang! Mar 07 00:38:21 Okay. I am about to install a new image. I want to see what i keep and what goes. Mar 07 00:41:28 I have been goofing up my system for years. I am finally starting to understand what I have been doing versus what needs to be done. Mar 07 00:41:37 Says, adminitration tasks. Ouch! Mar 07 00:55:05 set_, ever hear of a "Beaglebone Enhanced" ? Mar 07 00:57:44 Yep. Mar 07 00:57:48 I saw it. Mar 07 01:13:01 I just erased a ton of stuff and hopefully am going to learn my lesson (this time). Boy! Mar 07 01:13:27 I took out the Yea b/c of Flava Flav getting fired. I will miss him. Mar 07 01:16:30 Lost the connection again Mar 07 01:17:29 The "Enhanced" has 1GB RAM, wifi,built in sensors temp/humid Mar 07 01:17:51 Oh and 3 USB ports Mar 07 01:20:19 I thought it has four. Mar 07 01:21:03 I never did get one of those for some reason. Mar 07 01:21:03 But a different serial console pins Mar 07 01:21:03 Hmm. Mar 07 01:21:28 I think it's 3 real and 1 on the header or pad Mar 07 01:21:39 Oh. Mar 07 01:22:16 I just acquired this really expensive printer, i.e. an HP something or other. Linux, from HP's site, has a driver for it. Mar 07 01:22:28 But, it sounds like people are churning rocks right now as it tries to print. Mar 07 01:22:42 Has anyone made a printer from a BBB? Mar 07 01:23:22 The Beaglebone download site the images will work on the Enhanced. Postscript printer? Haven't seen one Mar 07 01:23:44 Okay. Mar 07 01:23:49 Yea. Mar 07 01:23:57 They will, at least that is what is claimed. Mar 07 01:24:10 I like the 1GB ram Mar 07 01:25:45 Golly, now your talking "lpr" "lpstat" USB drivers for linux. Oh, golly. What model? Mar 07 01:25:46 Yea. That is nice. I like how BBB.io peoples just make their images work out of the box. No issues. Mar 07 01:25:56 8000 Mar 07 01:26:11 inkjet Pro! Mar 07 01:26:24 I have hplip drivers. Mar 07 01:27:31 I think it needs printheads. It is having a difficult time right now. Mar 07 01:27:41 Did U look here : https://support.hp.com/us-en/document/c01174360 Mar 07 01:27:48 Yea. Mar 07 01:27:53 I have read a ton on it so far. Mar 07 01:28:34 I can just run their source w/ python test.py and so forth. Mar 07 01:28:53 They have a ton of scripts for testing the printer functions. Mar 07 01:29:12 Get new cartridges. Does it work on a PC? Mar 07 01:29:55 Well, I have not tested it on Win, yet. I got some new cartridges today. A cheap knock off brand is in town. I tried them. Mar 07 01:30:01 Probably the reason. Maybe? Mar 07 01:30:46 Don't get any ink on your hands. Hard to remove Mar 07 01:31:08 Cartridge World was the place to be and cheap. The HP co. was asking a lot just for one. I found a ton for the price of one. Mar 07 01:31:23 Yea, you use knock-offs and printer complains Mar 07 01:31:45 Right, right. It sounds like rain coming from the printer. Mar 07 01:32:04 I think it is making up its mind. AI! Mar 07 01:32:10 $600 dollar printer! Mar 07 01:32:17 I know. I know. Mar 07 01:32:28 I am losing it. Mar 07 01:32:46 It has absolutely three buttons, a ethernet port, and a funny power adapter. Mar 07 01:33:02 I do not understand why people paid this much for this thing. Mar 07 01:33:43 Humm on Amazon it says USB 2.0 Mar 07 01:33:49 I guess while one person prints, queues are in order for the rest of the people on the intranet, and then some. Mar 07 01:33:55 Yep. I think it is olderr. Mar 07 01:34:05 Or for the BBBE? Mar 07 01:34:34 Now that wifi is working can I access BBB over wifi? Mar 07 01:34:47 Yes! Mar 07 01:34:49 Wifi! Mar 07 01:35:43 You can plug it in and watch it go. Mar 07 01:36:35 OK, now showing up on network scan but wifi works Mar 07 01:36:44 Gargling rocks is what this thing sounds like. Sheesh. sudo apt update! Mar 07 01:36:51 OK, not showing up on network scan but wifi works Mar 07 01:37:06 Hmm. Mar 07 01:37:07 It just spit something out. Neat. Mar 07 01:37:24 >>>>>> Great <<<<<< Mar 07 01:37:28 Blank page. Mar 07 01:37:29 Dang it. Mar 07 01:37:39 It is printing now. Mar 07 01:37:51 Awesome! Mar 07 01:38:07 Could the BBB be hiding from the network? Mar 07 01:38:34 Maybe. So, what do you see when you type connmanctl, then services? Mar 07 01:38:38 Just nothing? Mar 07 01:39:10 Ethernet A0 or *A? Mar 07 01:39:44 This dang thing is just printing blanks now. Argh. Mar 07 01:39:46 No the BBB can access wifi OK. What I an trying to do is access BBB from PC to BBB Mar 07 01:39:54 oh. Mar 07 01:39:57 ip addr? Mar 07 01:40:31 USB to BBB for development and no ethernet cord for now. THat is all you can do until you make something. Mar 07 01:40:41 Since I am using DHC Mar 07 01:40:58 Since I am using DHCP where is the IP hiding? Mar 07 01:41:09 ip addr Mar 07 01:41:12 TYpe that. Mar 07 01:41:14 See... Mar 07 01:41:39 ifconfig is old school. ip addr says they are taking over. Mar 07 01:41:45 Try it. Mar 07 01:42:15 Then, look for wlan0. Mar 07 01:44:11 Yep, ifconfig showed it under wlan0. Oh no I can attach to it without a password!! Mar 07 01:44:29 Cloud9 Mar 07 01:44:33 Yep. Mar 07 01:44:38 But to run things, it may ask. Mar 07 01:44:42 No password!!! Mar 07 01:44:53 If you run something, it may ask you. Mar 07 01:45:36 It went rite in as user debian. whoami Mar 07 01:45:44 Yep. Mar 07 01:45:51 rite = right! Mar 07 01:45:58 Not secure at all Mar 07 01:46:09 Come on, man. Think of Flava Flav. Mar 07 01:48:13 I think if I have an app access should have atleast a password Mar 07 01:53:42 Sure. Mar 07 01:53:45 YOu can make one. Mar 07 01:57:29 Yea, but going to IP brings up Cloud 9 and disen't need a password Mar 07 01:59:39 I changed password for debian but Cloud 9 blows right in Mar 07 01:59:48 OH. Mar 07 01:59:52 Press cd Mar 07 02:00:03 It brings up cloud9 but you can get to your home/debian dir. Mar 07 02:00:14 YOu can erase that if necessary. Mar 07 02:00:24 But, you may get errors down the line. Mar 07 02:00:46 You know, think of the children learning and how they can jump in, immersively, and get to coding. Mar 07 02:01:20 That is a quick set up for them w/ out all the "intuitive" commands and ideas. Just install, go to the server, and code. Mar 07 02:01:27 They even have a thing called node red. Mar 07 02:01:43 It is more of a block style coding representation. Mar 07 02:03:45 Literally, blocks of visuals w/ words in b/t the blocks to connect them w/ colors and things. Mar 07 02:03:54 Like puzzles. Mar 07 02:09:12 Well, I think it is the print heads. The printer stopped making grunting noises. Mar 07 02:10:38 Sounds like it's trying to clean the jets Mar 07 02:11:22 Can you tell it to clean the print heads or wun a print test? Mar 07 02:14:01 Yea. I am doing it now. Mar 07 02:14:21 I am cleaning the printheads and printing a test page. Mar 07 02:14:35 I mean, that was the python script. Mar 07 02:14:43 Who knows what it will actually do. Mar 07 02:15:31 Printer should have built in ability to run test Mar 07 02:15:41 I know. That is what i think too. Mar 07 02:15:51 Not on this one. Mar 07 02:16:17 It has three buttons. One is Ethernet on, move print head, and the other is a red x. Mar 07 02:20:20 Old printer with built in web server. Mar 07 02:21:03 Nice, heh? Mar 07 02:21:26 It may have 3500 old prints on it. I have not a clue. Mar 07 02:22:38 I have an old multi-function HP 4620. print,copy,scn,fax Mar 07 02:25:52 I am off to look it up. Mar 07 02:26:54 same idea...the print heads are old and busted. I might need to replace them, e.g. not from HP. Sheesh. Mar 07 02:29:15 Get a new HP printer and start fresh Mar 07 02:29:47 Nope. I am in it to win it! Mar 07 02:30:21 It is like an old truck. Sometimes, they get discarded and other times they get fixed. Mar 07 02:36:25 KenUnix: I am transferring data right now onto the BBGW. Mar 07 02:36:33 How is your WiFi access coming along? Mar 07 02:44:18 Working fine now. So I know the usb 2.0 powered hub & wifi adapter work out of the box. Mar 07 02:48:15 Nice! Mar 07 02:55:01 Do you want to know which i got? Mar 07 02:55:53 I will tell you with the wifi adapter a poered usb hub is required Mar 07 02:56:28 plugging in he wifi directly it cut in and out Mar 07 02:56:38 plugging in the wifi directly it cut in and out Mar 07 02:57:15 you know up on beaglebone there is only 1 image for the 'ai' Mar 07 02:57:48 and it's lxqt Mar 07 02:59:42 Right. Mar 07 02:59:46 Right. Mar 07 02:59:48 Sure, tell me. Mar 07 03:01:20 WiFi https://www.amazon.com/gp/product/B008IFXQFU/ Mar 07 03:02:02 Hub https://www.amazon.com/gp/product/B005P2BY5I/ Mar 07 03:02:38 So, the wifi goes into the powered USB hub? Mar 07 03:02:44 SD cards https://www.amazon.com/gp/product/B06XWN9Q99/ **** ENDING LOGGING AT Sat Mar 07 03:03:04 2020