[beagleboard] Re: Azure IoT Edge Runtime on BeagleBone Green
Hi! On Monday, April 5, 2021 at 4:25:10 PM UTC+3 Stony Kong wrote: > *The following packages have unmet dependencies:* > * moby-engine : Depends: moby-containerd (>= 1.4.3) but it is not going to > be installed* > * Depends: moby-runc (>= 1.0.0~rc93) but it is not going to > be installed* > *E: Unable to correct problems, you have held broken packages.* > This tries to install some packages which fail to resolve their dependencies. I'd guess that either the packaging is done poorly, or the Debian your BBB is running is simply not compatible. Since the packages are emitted by Microsoft, there's not much productive you can do. You can try to build the things from scratch from source code, if feeling adventurous. -- Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/3ee283f8-a04d-47a9-8510-c18d84237edan%40googlegroups.com.
[beagleboard] Re: 4G dongle keeps debouncing
On Thursday, October 29, 2020 at 1:16:13 AM UTC+2 atiqu...@gmail.com wrote: > I have a 4G dongle which keep on denouncing between two interfaces > How do you supply power to the 4G dongle? It's not connected directly to the Beagle's USB port, is it? -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/e2b9e165-96c2-456d-b221-77f529ebf7b5n%40googlegroups.com.
[beagleboard] Re: Date/time update of second BBB
On Friday, October 23, 2020 at 8:29:20 AM UTC+3 biggu...@gmail.com wrote: > I have two BeagleBone Black boards connected together with an Ethernet > cable. One BBB has a battery backed RTC the other does not. Is there a > simple way for the second BBB to update its date/time from the first BBB? If they were both connected to Internet, you wouldn't have to do anything as the second BBB would update its time via NTP as soon as it's online. If they're not connected to Internet, it should be a matter of running an NTP server on the BBB with battery packed RTC. The other one can run an NTP client. First tutorial I came upon: https://vitux.com/how-to-setup-ntp-server-and-client-on-debian-10/ I don't know how the NTP server will behave if there's no Internet in the first BBB. Configuration may require some tweaks to account for this. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/fd7a2741-aca4-408e-80aa-b3f8a43a3386n%40googlegroups.com.
Re: [beagleboard] Re: Remote access for the beagle bone black (BBB) to modify /upgrade file from cloud server
On 21.10.20 06:27, Niresh wrote: Thank you Tamro. I hope a trigger mechanism is not a problem because continuous data exchange is happening between the AWS server and controller through websocket. I can create a separate data packet to initiate this trigger command to a controller to which I wish to upgrade. In my case, only two executable binary files are required for firmware update and which I must keep in the AWS server because data privacy is very important. It's not terribly relevant where the files are stored - it's hosting them that requires some thinking. E.g. googling for "amazon aws host apt repository" seems to come up with a few solutions - e.g. https://www.aptly.info. I can't vouch for any of them. I'm a newbie to the Linux environment. So, It would be very grateful if you can explain to me with an example for better understanding or a link. I'm sorry to say that automatic upgrades of running services on a Debian box is quite the deep end to get thrown into as a newbie :) Mark's suggestion is not an improper one in this case. If you go the route of deb package and apt repository, I'd suggest reading chapters 5, 6 and Debian Administrator's handbook https://debian-handbook.info/browse/stable/ - this briefly explains how they are used. https://wiki.debian.org/DebianRepository/Setup has some info on how repositories are set up. AWS-friendly Aptly is in there somewhwere. Depending on your circumstances, you might want to look into creating a systemd service out of your application: https://linuxconfig.org/how-to-create-systemd-service-unit-in-linux -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/2337d951-9226-703e-0e9f-bdeaa2554c82%40gmail.com.
Re: [beagleboard] Re: Remote access for the beagle bone black (BBB) to modify /upgrade file from cloud server
On 20.10.20 15:24, Niresh wrote: I have a binary file that is basically compiled C application code. If I send an upgrade command to the controller from the server, the controller should take that binary file from the server, replace the old one and run with the updated binary file. I've distributed custom software as deb packages hosted by a private apt repository (Sonatype Nexus OSS). Mass-updates got manually triggered via ansible (which required the aforementioned SSH jumphost). You can do something similar - package your binaries as a deb package, host an apt repository and figure out how to trigger updates. Or you can roll your own update scripts and repo. Or you can search for third-party solutions. I assume most of the underlying problems are the same - a trigger mechanism, a hosted package repository, authenticated data pipe, service restarting, recovery from botched upgrades etc. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/0552ff7a-cca0-b8fe-de2b-85aeee5d3251%40gmail.com.
Re: [beagleboard] Re: Remote access for the beagle bone black (BBB) to modify /upgrade file from cloud server
Hi Niresh, On Tuesday, October 20, 2020 at 7:18:07 AM UTC+3 Niresh wrote: > Thank you for the information. I remember my ex-colleague used to download > a file from a server using curl utility and run those files. So, I would > like to know any similar simple procedure to download and upgrade BBB > binary files which are stored in the server. > You're looking to automatically upgrade your custom software? There's a bunch of ways to do it, depending on how your thing is distributed (e.g. binary executable, Java bytecode, Python, ...). The obvious solution would be to package your software as a Debian .deb package, and run "apt update && apt -y install yourpackage" from a cron script. This would attempt to upgrade to a fresh version of "yourpackage" with whatever frequency you set up the cron script. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/8847ee90-49b4-4ecf-986c-d1d383162cc2n%40googlegroups.com.
Re: [beagleboard] Re: Remote access for the beagle bone black (BBB) to modify /upgrade file from cloud server
On Thursday, October 15, 2020 at 8:43:46 AM UTC+3 Niresh wrote: > Is there a way to access the beagle-bone using SSH or something like that > without registering the device to another server? > This is a network setup issue. Any remote access solution requires that a network route exist from your computer to the BBB. If you and the BBB are in the same local network, it's routable and you can connect to it. If the BBB has a public IP address (or port forwarding is set up to it), it's routable and you can connect to it. If the BBB is behind a NAT and no port forwarding is set up, it's not routable and you cannot connect to it. Here's where the various tunnelling and VPN solutions come in. You can choose one which is best for your requirements, but you cannot get remote access to a non-routable device without them. I've used a simple jumpbox in the form of an SSH server where all the different BBB-s connect to. Each opens a remote tunnel on a specific port which leads back to it. Eg. port 20013 would have a tunnel to device nr 13, port 20014 to nr 14 etc. There's a Debian package called autossh with a script which maintains a persistent SSH connection to the jumpbox. This is a fairly simple solution to set up and use initially, but not a very good one - mainly because it doesn't scale beyond a few dozen devices. I'd go for a VPN next time. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/43e35925-e214-450b-8f64-2b0f99b31491n%40googlegroups.com.
[beagleboard] Re: Hide the scrollbar from the window manager
Hi! On Tuesday, October 13, 2020 at 3:53:03 PM UTC+3 hanswu...@googlemail.com wrote: > Hello @ all, > > My system: > Beaglebone Black, > Debian Stretch V9.9, > 5" Display (Matrix Orbital) > > > browser: Midori V.0.5.11 > > additionally I have installed the following > openbox, > xserver-xorg-video-fbdev, > x11-xserver-utils, > lightdm > > > I wrote a web app in html5 and node express. > > I am using the window manager to display the web app in kiosk mode. > *$ DISPLAY=:0 midori -e Fullscreen -a http://localhost:8082 > <http://localhost:8082>* > > > *My problem:* > The vertical scrollbar is probably displayed very thin by the window > manager. I want to make this scrollbar wider or hide it and use CSS to > create my own scrollbar. > > If I create a custom scrollbar with css I get 2 scrollbars side by side. > > > /* width */::-webkit-scrollbar { > width: 20px; > } > /* Track */::-webkit-scrollbar-track { > box-shadow: inset 0 0 5px grey; > border radius: 10px; > } > /* Handle */::-webkit-scrollbar-thumb { > background: red; > border radius: 10px; > } > /* Handle on hover */::-webkit-scrollbar-thumb:hover { > background: #b3; > } > > > Who can help me ? > > Unfortunately I cannot upload any photos here. > > I always get the message that it is rejected by the server. > Seems like Midori uses the GTK toolkit. You can install the GTK theme switcher (gtk-theme-switch) and a few different themes (apt-cache search gtk-theme) and play around with them - maybe one of those is more suitable. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/5813b329-620e-4d7d-8a53-196c9a20419cn%40googlegroups.com.
[beagleboard] Re: Boot time optimization
On Thursday, October 8, 2020 at 11:28:49 PM UTC+3 mfar...@gmail.com wrote: > kernel:[5.4.47-bone30] > Ooh, kernel 5.4 - how experimental is this? > 1min 34.446s dev-mmcblk0p2.device > 1min 2.259s generic-board-startup.service > 50.664s dev-loop8.device > 49.671s dev-loop7.device > 49.351s dev-loop6.device > 48.843s dev-loop4.device > 48.804s dev-loop5.device > 48.462s dev-loop3.device > 48.420s dev-loop1.device > 48.392s dev-loop0.device > 48.197s dev-loop2.device > I haven't seen those dev-loop devices before (perhaps they come with kernel 5.4). They look rather suspicious with the 50 second duration. Have a look at what their logs say, e.g.: $ journalctl -u dev-loop0 > 24.250s snapd.service > The snap package system is a bit of a resource hog and it primarily serves as a convenience for some minority use cases. Are you sure you need it on a BBB? Final thought - maybe your SD card simply has poor performance? I doubt it'll make a significant difference, but you can try flashing your image into eMMC or using a higher-end SD card. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/12a87597-62b6-4869-8b01-a97b850854ffn%40googlegroups.com.
[beagleboard] Re: Beaglebone black connect to Web
On Thursday, October 8, 2020 at 3:16:24 PM UTC+3 dysona...@gmail.com wrote: > I have a beaglebone black and I'm attempting to connect it to the outside > world. I have the latest debian image installed. > > Here is what I have done: > > - installed latest windows 64 bit drivers > - Configured my wifi network adapter to be shared with the network adapter > created through the USB cable connection. > > When I attempt to ping www.google.com I get the error "Temporary failure > in name resolution" > > Any help would be of benefit. > IIRC the BB-s virtual Ethernet port is not configured to be an uplink to the Internet - the Ethernet port is. Changing this is going to be tricky (not impossible, but requires in-depth understanding of what you're doing). The simplest way to solve your problem is to connect the BB to Internet using an Ethernet cable and your nearest router. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/2f1af32f-1814-4d5d-837b-4aea0dc59984n%40googlegroups.com.
[beagleboard] Re: rcpy for beaglebone blue
Try installing libgpiod from the debian package repository - with luck you might find an older version which supports the available kernel version. Run "apt search libgpiod" and see if the packages are there; then install them as usual ("sudo apt install libgpiod libgpiod-dev" or something similar). For compiling other source code against it, you'll also need the development package which ends with "-dev". -- Kind regards, Tarmo On Tuesday, September 22, 2020 at 4:46:44 AM UTC+3 ryansel...@gmail.com wrote: > Hello, > > I'm just getting into beaglebone and working with linux in general. I have > the robotics controller library loaded and up to date. I would like to use > rcpy. The first step on the rcpy documentation is to install libgpiod. I > can git that fine, but when I go to do the autogen step, it gives me an > error saying "libgpiod needs linux headers version >= v5.5.0". The kernel I > have is 4.19 from the most recent flashed image. Is there anyway around > this? > > Thanks in advance > -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/d38ff3af-04a4-424d-81d6-28649b85d3ebn%40googlegroups.com.
Re: [beagleboard] How to automatically set system time of BBB by external DS3231
I'd recommend a new thread for a new topic. -- Tarmo On Thu, 3 Sep 2020, 06:34 Niresh, wrote: > Hi Tarmo, > Thank you for your suggestion. > > I have a few queries related to USB dongle connection to BBB and mutex > operation. Shall I ask that query in this post or separate? > > Regards, > NK > > On Wed, Sep 2, 2020 at 6:06 PM Robert Nelson > wrote: > >> On Wed, Sep 2, 2020 at 12:41 AM Niresh wrote: >> > >> > Hi, >> > You are right, I've done a little experiment to confirm the same. >> > >> > Test #1: Made both system and RTC time synchronized with 2020-09-02 >> 09:00:30 and then rebooted , both time syncs. >> > >> > Test #2: Made RTC time delay with an hour by using command hwclock >> --set --date "09/02/20 08:00:30 "and system time 2020-09-02 09:00:30 and >> then rebooted. Now, the system time is not sync with RTC,it remains >> unchanged 2020-09-02 09:00:30 >> > >> > Test #3: Made RTC time advance with an hour by using command hwclock >> --set --date "09/02/20 10:00:30 "and system time 2020-09-02 09:00:30 and >> then rebooted. Now, the system time is sync with RTC, that's with >> 2020-09-02 10:00:30 >> > >> > Could you please tell me starts from which version of kernel >> "BB-I2C2-RTC-DS3231.dtbo" device tree works. I will tell you why, I was >> using arm-linux-gnueabihf-gcc-4.9 to compile my application code in the >> 4.4.30-ti-r64 kernel, and it compiled and ran perfectly. Now the same >> application code gives me an error message with compiler >> arm-linux-gnueabihf-gcc-8 in the new image 4.19.94-ti-r42. >> > >> > Like I said before, I am not a linux guy and the developer, who coded >> this and has left the organisation, cautioned me to stick with >> 4.4.30-ti-r64 kernel for stable operation of application code. Please >> advise which version is good to go and how to overcome this issue. >> >> Can you share the actual "error message"? >> >> Regards, >> >> -- >> Robert Nelson >> https://rcn-ee.com/ >> >> -- >> For more options, visit http://beagleboard.org/discuss >> --- >> You received this message because you are subscribed to a topic in the >> Google Groups "BeagleBoard" group. >> To unsubscribe from this topic, visit >> https://groups.google.com/d/topic/beagleboard/m4hZYsA-d8M/unsubscribe. >> To unsubscribe from this group and all its topics, send an email to >> beagleboard+unsubscr...@googlegroups.com. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/beagleboard/CAOCHtYgMqVpXgg6qJ1fgtkCbJCY2f8OtcDfVEYbYW3CWbFA0rw%40mail.gmail.com >> . >> > -- > For more options, visit http://beagleboard.org/discuss > --- > You received this message because you are subscribed to a topic in the > Google Groups "BeagleBoard" group. > To unsubscribe from this topic, visit > https://groups.google.com/d/topic/beagleboard/m4hZYsA-d8M/unsubscribe. > To unsubscribe from this group and all its topics, send an email to > beagleboard+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/beagleboard/CAF7-PaQSOLy9q65Fth%3DjMkAEL5rQSaxGchrDM4%2Bzvtuj8%3D7jaA%40mail.gmail.com > <https://groups.google.com/d/msgid/beagleboard/CAF7-PaQSOLy9q65Fth%3DjMkAEL5rQSaxGchrDM4%2Bzvtuj8%3D7jaA%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/CAPW7rrJHwEoP7%2B1wLK4LWZb0UCapyw%3D_0HrP3%2BMExgF%2B1d%3D96g%40mail.gmail.com.
Re: [beagleboard] How to automatically set system time of BBB by external DS3231
On 02.09.20 11:55, Tarmo Kuuse wrote: "apt-cache search 'kernel-image.*4.4.30*'") but I have no idea what Sorry, the search term is probably 'linux-image*4.4.30*', I don't remember what the BB-s kernel packages were named. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/09d31846-6702-2093-80ee-c292579d7767%40gmail.com.
Re: [beagleboard] How to automatically set system time of BBB by external DS3231
Hi Niresh, On 02.09.20 08:41, Niresh wrote: You are right, I've done a little experiment to confirm the same. Good to hear! Could you please tell me starts from which version of kernel "BB-I2C*2*-RTC-DS3231.dtbo" device tree works. I will tell you why, I was using arm-linux-gnueabihf-gcc-4.9 to compile my application code in the 4.4.30-ti-r64 kernel, and it compiled and ran perfectly. Now the same application code gives me an error message with compiler arm-linux-gnueabihf-gcc-8 in the new image 4.19.94-ti-r42. Like I said before, I am not a linux guy and the developer, who coded this and has left the organisation, cautioned me to stick with 4.4.30-ti-r64 kernel for stable operation of application code. Please advise which version is good to go and how to overcome this issue. I haven't a clue. Looking at the the larger picture, the Debian distribution installed on your BeagleBone (version 10 a.k.a. "buster") is happy to work with a few different versions of the gcc compiler and the kernel. The problem here is that you're looking to use some really old stuff. Regarding gcc - Debian 10 has out of box support for gcc-7 and gcc-8, but not gcc-4. You can search for older versions (e.g. "apt-cache search gcc-7") and install them (e.g. "sudo apt install gcc-7"). If gcc-4.9 is really required, you might try the official packages from Debian 8 "jessie" (when it was last supported), but it's going to be tricky to install and leave the package system in a weird state. You can also compile gcc-4.9 yourself from source :). As an alternative, you can try to keep your code up to date with newer compiler releases and try fixing the code so it compiles with gcc-8. Regarding the kernel - there are probably old kernel releases which are OK to use on Debian 10 and might even support the recent device trees that you need. You can search for them in the package system (e.g. "apt-cache search 'kernel-image.*4.4.30*'") but I have no idea what happens if you install them. Maybe Robert can comment on that. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/6f471f52-c12a-aa7a-eee3-00f88c4d97fb%40gmail.com.
Re: [beagleboard] How to automatically set system time of BBB by external DS3231
On 01.09.20 14:55, Niresh wrote: I purposefully disconnected the battery to confirm system time updates with external RTC time. But the below message shows not updated right. Local time: Tue 2020-09-01 10:32:27 UTC Universal time: Tue 2020-09-01 10:32:27 UTC RTC time: Sat 2000-01-01 00:15:42_ Ah, you're expecting the system time to be set to 2000-01-01? Won't happen. RTC knows it's been reset and because of that the system refuses to sync with it. The reason why your system time is close to accurate (but still off by several minutes, if you check) after everything has been powered off is due to a fallback method in timesyncd. It will occasionally (e.g. on NTP sync and on shutdown) store the current timestamp in a magic file on disk. If there is no source of time available during boot (no RTC and no NTP) then timesyncd will declare the value of that timestamp as current time. At least it guarantees a monotonously increasing time. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/338861fc-3973-12a5-e0e9-6fc979c9dfc3%40gmail.com.
Re: [beagleboard] How to automatically set system time of BBB by external DS3231
On 01.09.20 13:59, Niresh wrote: I am a bit confused. Please correct me if I am wrong. I have removed the internet connection, power to the board and battery from the external RTC DS3231 module too. Well, removing the battery will reset your DS3231. This defeats the purpose of having a battery powered RTC. Its ok, the device tree swaps rtc0 and rtc1 but the system time is not updated with external RTC time after boot. Powering up the board without internet connection and external RTC resetted to default time. Yes, because you disconnected the battery from the RTC. Don't. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/342c4a25-ccc0-f69b-3039-028c3cc59485%40gmail.com.
Re: [beagleboard] How to automatically set system time of BBB by external DS3231
Hi Niresh, On 01.09.20 07:43, Niresh wrote: I've tested BB-I2C2-RTC-DS3231.dtbo in the new image 4.19.94-ti-r42, they mount rtc1 in /dev but system time is not updated with rtc1. Attached dmesg with and without DS3231 overlay in uEnv.txt . Following message with DS3231 overlay enabled. root@beaglebone:/dev# hwclock -r -f /dev/rtc1 2000-01-01 00:49:21.771707+00:00 That's OK. The device tree swaps rtc0 and rtc1. Now your DS3231 is /dev/rtc0 - which is what you need. The AM335x-s RTC is /dev/rtc1 - it goes back to year 2000 on every boot. You don't need it. root@beaglebone:/dev# timedatectl Local time: Tue 2020-09-01 04:34:29 UTC Universal time: Tue 2020-09-01 04:34:29 UTC RTC time: Tue 2020-09-01 04:34:29 Time zone: Etc/UTC (UTC, +) System clock synchronized: no NTP service: active RTC in local TZ: no Seems to be correct. "RTC time" is your DS3231. debian@beaglebone:/dev$ dmesg | grep rtc [ 1.280996] rtc-ds1307 2-0068: registered as rtc0 [ 1.283636] omap_rtc 44e3e000.rtc: already running [ 1.284312] omap_rtc 44e3e000.rtc: registered as rtc1 [ 1.574739] [drm] Cannot find any crtc or sizes [ 1.604280] rtc-ds1307 2-0068: setting system clock to 2020-09-01 04:30:36 UTC (1598934636) That's your DS3231 being assigned /dev/rtc0 by the kernel and the system time correctly set from it. Looks good. The last step is to enable NTP synchronization: $ sudo timedatectl set-ntp true Then you're done. When network comes online, systemd-timesyncd will synchronize both your DS3231 and the system time from NTP. Feel free to fine-tune ("man timedatectl" and man "timesyncd.conf"). -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/a8f8f997-41a5-8032-4c21-e8cc425b49ae%40gmail.com.
Re: [beagleboard] How to automatically set system time of BBB by external DS3231
Hi Niresh, On Monday, 31 August 2020 14:14:05 UTC+3, Niresh wrote: > > Hi, > > debian@beaglebone:/dev$ i2cdetect -r -y 2 > > 0 1 2 3 4 5 6 7 8 9 a b c d e f > > 00: -- -- -- -- -- -- -- -- -- -- -- -- -- > > 10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > > 20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > > 30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > > 40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- > > 50: -- -- -- -- -- -- -- UU -- -- -- -- -- -- -- -- > > 60: -- -- -- -- -- -- -- -- 68 -- -- -- -- -- -- -- > > 70: -- -- -- -- -- -- -- -- > Looks like your RTC is connected to I2C bus number 2 (counting starts from 0). I'm a bit confused. How could the command in your initial post pass and create a working RTC device? You were addressing the wrong bus (number 1) here: echo ds3231 0x68 > /sys/class/i2c-adapter/i2c-1/new_device Anyway, if you're really connecting the RTC to bus 2 then "BB-I2C*2*-RTC-DS3231.dtbo" should be your device tree. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/ed5f7cb9-873a-44d0-8715-c5c624f046d4o%40googlegroups.com.
Re: [beagleboard] How to automatically set system time of BBB by external DS3231
Hi Niresh, On Monday, 31 August 2020 06:45:48 UTC+3, Niresh wrote: > > Attached output from "sudo /opt/scripts/tools/version.sh" and the content > of your "/boot/uEnv.txt" file along with boot log message. > They look good. I suspect you're mostly there :) debian@beaglebone:/boot$ dmesg | grep rtc > > [2.283660] rtc-ds1307: probe of 1-0068 failed with error -16 > This indicates that the device tree was loaded correctly, but the kernel can't talk to your RTC chip. If you can figure out why, you've enabled systemd-timesyncd. Hmm, can you double-check that the I2C bus number is correct and the RTC chip is online? There is a utility called i2cdetect which should reveal your RTC on bus 1, address 68. https://linux.die.net/man/8/i2cdetect > Please help me with bash scripts for external RTC and its procedure. > Undo your changes to uEnv.txt and follow this guide instead. It's not ideal, but probably covers the relevant requirements. https://learn.adafruit.com/adding-a-real-time-clock-to-beaglebone-black/set-rtc-time -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/500b2fb8-177f-4a6e-885a-1d0c612f2e24o%40googlegroups.com.
Re: [beagleboard] How to automatically set system time of BBB by external DS3231
On Thursday, 27 August 2020 08:07:53 UTC+3, Niresh Kumar wrote: > > I've have tested both BB-I2C1-RTC-DS3231.dtbo and BB-I2C2-RTC-DS3231.dtbo >in the new image 4.19.94-ti-r42, they mount rtc1 in /dev but system > time is not updated with rtc1. Please let me know if any other procedure is > required and also procedure to update external rtc1 time when the > internet is connected. > That's unfortunate. Still, logs from the uboot bootloader (observed from the onboard TTL level UART) or the Linux kernel (from dmesg or syslog) might indicate what's wrong. To get effective help, please post output from "sudo /opt/scripts/tools/version.sh" and the content of your "/boot/uEnv.txt" file. > One thing i have observed, with BB-I2C1-RTC-DS3231.dtbo the timedatectl > command displayed "RTC time: n/a" not system time as in normal condition. > Weird. I would interpret it as the AM335x internal RTC having been pushed aside from /dev/rtc0 successfully, but the external RTC does not pick the device up. > Please clarify the following points > 1. How do you say the rtc chip is in i2c bus one. Using i2cdetect command > "i2cdetect -r 2" , 0x68 external device showed in i2c bus two. > There might be differences in how different systems count buses. One might start from 0 and other from 1. TBH it's always been confusing as hell for me. > 2. I understand overlay is an out of box approach. In 4.4.30-ti-r64, > system time updated with "echo ds3231 0x68 > > /sys/class/i2c-adapter/i2c-1/new_device; hwclock -r -f /dev/rtc1" in > /etc/rc.local. > Is there any problem if I use this approach in older images? > No, go right ahead. You're left with the task of figuring out a solution for syncing time between the RTC, system clock and NTP - as you wrote out in the very first message. My suggestion was to have systemd-timesyncd do it for you, but you can, as an alternative, do it yourself with a few bash scripts. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/37bcd7f5-7e35-4c1f-9507-0716056b2c26o%40googlegroups.com.
Re: [beagleboard] How to automatically set system time of BBB by external DS3231
On Monday, 24 August 2020 00:12:39 UTC+3, Alexander Zangerl wrote: > > On Thu, 20 Aug 2020 00:40:38 -0700, Niresh Kumar writes: > >1). I have placed the following lines in the /etc/rc.local, and system > >time gets updated with external RTC DS3231 automatically at every reboot > > > >*echo ds3231 0x68 > /sys/class/i2c-adapter/i2c-1/new_device* > ... > > > >2) I have put the line *dtb_overlay = > >/lib/firmware/BB-I2C2-RTC-DS3231.dtbo * in /boot/uEnv.txt but it does > not > > > the overlay you're using won't work because your rtc chip is > on i2c bus one: as its name (...I2C2...) indicates it's for a > clock chip being on i2c bus two. > > use dtc to decompile that overlay, edit for i2c1 and dtc it back into > binary form. > > Good catch, although it's probably not necessary to decompile binary trees as the source is available here together with a nice makefile to build a single overlay: https://github.com/beagleboard/bb.org-overlays Without having tested it, I might imagine a process similar to this one: $ git clone https://github.com/beagleboard/bb.org-overlays.git $ cd bb.org-overlays $ cp src/arm/BB-I2C2-RTC-DS3231.dts src/arm/BB-I2C1-RTC-DS3231.dts $ vim src/arm/BB-I2C1-RTC-DS3231.dts ... update I2C bus number ... $ make src/arm/BB-I2C1-RTC-DS3231.dtbo $ cp src/arm/BB-I2C1-RTC-DS3231.dtbo /lib/firmware -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/ffd21cc3-37ea-4526-80da-b4e1ae428d66o%40googlegroups.com.
Re: [beagleboard] How to automatically set system time of BBB by external DS3231
On Friday, 21 August 2020 06:13:24 UTC+3, Niresh Kumar wrote: > > We did not make any changes to the .dts file. uEnv.txt file attached for > your reference. > > There is no script "verison.sh" in the path /opt/scripts/tools/ in kernel > version 4.4.30-ti-r64. > Wow, you're running a rather old Debian release. The stuff that I suggested was likely not implemented back then. I would upgrade to a recent Debian release: http://beagleboard.org/getting-started Note to whoever maintains beagleboard.org: the Letsencrypt SSL cert expired 3 days ago. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/42bb5db5-cc5a-4104-a2f4-b781ed889bbfo%40googlegroups.com.
Re: [beagleboard] How to automatically set system time of BBB by external DS3231
On Tuesday, 18 August 2020 17:47:00 UTC+3, Tarmo wrote: > > On Tuesday, 18 August 2020 14:45:23 UTC+3, Robert Heller wrote: >> >> At Tue, 18 Aug 2020 02:40:41 -0700 (PDT) beagl...@googlegroups.com >> wrote: >> > I am using beagle Bone Black with debian image. Could any please >> suggest >> > how to automatically set the system time (on every reboot) by external >> RTC >> > module DS3231. Once internet is connected, i want to update both system >> > time and DS3231 with the network time. >> > >> > I am controlling a relay based on system time, my application should >> take >> > DS3231 time and run perfectly even if internet is disconnected and not >> > available for so many days. >> > >> > Few things i have tried, the following content has been kept in the >> script >> > and running @reboot script in crontab. The system time has been changed >> if >> > we enter this command manually after boot but not in auto start script. >> > Please advice. >> > >> > *echo ds3231 0x68 > /sys/class/i2c-adapter/i2c-1/new_device* >> > >> > *hwclock -r -f /dev/rtc1* >> >> Well, the simple / dumb option would be to put the above two lines in >> /etc/rc.local, which will restore the system clock from the RTC. >> >> The other option is to create a proper RTC service. A properly setup RTC >> service will restore the system clock from the RTC early in the boot >> process >> and save the system clock late in the shutdown process. >> > > A proper RTC service is already implemented by the systemd-timesyncd > service, installed by default on the debian images: > > https://www.freedesktop.org/software/systemd/man/systemd-timesyncd.service.html > > It does everything Niresh needs out of box. There's a caveat: > systemd-timesyncd only works with */dev/rtc0* device. By default > */dev/rtc0* is grabbed by the AM335x CPU-s internal RTC (which has no > battery backup) and the external RTC is left with */dev/rtc1* which gets > ignored. I did not find any way to configure systemd-timesyncd to use the > other device. > > Robert pointed me to a device tree which very conveniently swapped the > device nodes for those RTCs so the external one got */dev/rtc0*. I just > had to load the device tree via /boot/uEnv.txt and time syncing worked > beautifully. For Niresh, I suggest trying this one: > > > https://github.com/beagleboard/bb.org-overlays/blob/master/src/arm/BB-I2C2-RTC-DS3231.dts > > <https://www.google.com/url?q=https%3A%2F%2Fgithub.com%2Fbeagleboard%2Fbb.org-overlays%2Fblob%2Fmaster%2Fsrc%2Farm%2FBB-I2C2-RTC-DS3231.dts&sa=D&sntz=1&usg=AFQjCNEt87UyBS-inpep25aZNVAfhEhfXw> > To clarify the "trying" step: you'll probably find the compiled device tree in the Beagle's file system: "/lib/firmware/BB-I2C2-RTC-DS3231.dtb". Just add this to /boot/uEnv.txt using standard method and reboot. Monitoring the boot log via dmesg should tell you which RTC device was awarded /dev/rtc0. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/340040ae-d9d1-4f28-9518-ec4f5ab8ac1eo%40googlegroups.com.
Re: [beagleboard] How to automatically set system time of BBB by external DS3231
On Tuesday, 18 August 2020 14:45:23 UTC+3, Robert Heller wrote: > > At Tue, 18 Aug 2020 02:40:41 -0700 (PDT) beagl...@googlegroups.com > wrote: > > I am using beagle Bone Black with debian image. Could any please suggest > > how to automatically set the system time (on every reboot) by external > RTC > > module DS3231. Once internet is connected, i want to update both system > > time and DS3231 with the network time. > > > > I am controlling a relay based on system time, my application should > take > > DS3231 time and run perfectly even if internet is disconnected and not > > available for so many days. > > > > Few things i have tried, the following content has been kept in the > script > > and running @reboot script in crontab. The system time has been changed > if > > we enter this command manually after boot but not in auto start script. > > Please advice. > > > > *echo ds3231 0x68 > /sys/class/i2c-adapter/i2c-1/new_device* > > > > *hwclock -r -f /dev/rtc1* > > Well, the simple / dumb option would be to put the above two lines in > /etc/rc.local, which will restore the system clock from the RTC. > > The other option is to create a proper RTC service. A properly setup RTC > service will restore the system clock from the RTC early in the boot > process > and save the system clock late in the shutdown process. > A proper RTC service is already implemented by the systemd-timesyncd service, installed by default on the debian images: https://www.freedesktop.org/software/systemd/man/systemd-timesyncd.service.html It does everything Niresh needs out of box. There's a caveat: systemd-timesyncd only works with */dev/rtc0* device. By default */dev/rtc0* is grabbed by the AM335x CPU-s internal RTC (which has no battery backup) and the external RTC is left with */dev/rtc1* which gets ignored. I did not find any way to configure systemd-timesyncd to use the other device. Robert pointed me to a device tree which very conveniently swapped the device nodes for those RTCs so the external one got */dev/rtc0*. I just had to load the device tree via /boot/uEnv.txt and time syncing worked beautifully. For Niresh, I suggest trying this one: https://github.com/beagleboard/bb.org-overlays/blob/master/src/arm/BB-I2C2-RTC-DS3231.dts -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/f0fdc385-9663-4dd4-a35b-b8844bdb92f3o%40googlegroups.com.
[beagleboard] Re: disable logging in beaglebone black
On Thursday, 13 August 2020 14:15:24 UTC+3, jennifer Dsilva wrote: > > how can we disable logging in bbb? my bbb writes log files into /var/log/ > directory and consumes a lot of space > As Jon mentioned, if you're getting anomalously verbose logs then have a look at what is the source of it and why. Generally I'd expect less then a megabyte of new logs every day from the system, plus any services which you're running. I don't know how verbose are the high-level services on official images (like nodejs, cloud9 etc) - I remove those before doing anything else. If you're worried about running out of disk space, then find the source of verbose logs and fix it. If you're worried about killing the Flash chip with frequent writes, the standard mitigation procedure is to divert the entire /var/log directory to a ramdisk. Some people use tmpfs, some people overlayfs. Latter is probably easier to set up. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/d86c8163-9e05-434d-b9b6-bc4c29c3c7ffo%40googlegroups.com.
Re: [beagleboard] Re: Beagleboard X15 Microsoft Visual Studio GUI example project
On Thursday, 16 July 2020 20:10:48 UTC+3, jonnymo wrote: > > Actually, Visual Studio is a good platform for cross compiling especially > with the newer versions of VS. There is a built in Linux feature that lets > you create Linux based projects and then you can just reference your ARM > based toolchain in the VS config. I use this with the RasPi but it should > be the same for the BB products. > Interesting. I guess it's a good time to be a cross-platform developer. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/8f27a87a-88ce-4b95-b249-a37d12a37627o%40googlegroups.com.
[beagleboard] Re: Beagleboard X15 Microsoft Visual Studio GUI example project
Hi Kevin, On Thursday, 16 July 2020 04:31:40 UTC+3, Kevin Hudson wrote: > > I am working on a project that uses the Beagleboard X15 as the hardware > platform. My task is to write a GUI that runs on the X15 using Microsoft > Visual Studio 2017. Do you know of any example project code that is > available anywhere that I can use to get started and as a reference? Thank > you for your help. > Note that from an application developer's perspective the wildly popular Raspberry Pi and the Beagles are mostly the same platform - a Debian GNU/Linux system on armhf architecture. You can use this to your advantage to google for something like "cross develop raspberry on windows" which will bring up quite a few more articles than searching for beagle-specific stuff. Visual Studio is a great tool for developing against x86/Windows targets. If you're developing with VS against a ARM/Linux target using third-party GUI libraries, it's probably possible (according to some articles Google spit out) but setting up cross-development GUI libraries is going to be a bit of a pain, I suspect. You could also take a more traditional cross platform approach, which involves setting up a GNU gcc toolchain from linaro with your favourite editor/IDE and required GUI dependencies. Haven't done either, though, so can't offer solid advice. An alternative with much less set-up effort - you can write, compile and run code of almost any language and GUI framework directly on the Beagleboard itself, especially if the project is small. Just install package 'build-essential' (assuming we're talking about a C/C++ application), the development packages for your chosen GUI framework (e.g. 'qtbase5-dev' for QT) and your favourite Linux editor/IDE. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/e41f096b-9eef-4ea2-a9a4-f65cc428e13do%40googlegroups.com.
[beagleboard] Re: BBB RTC synchronization with NTP
On Wednesday, 15 July 2020 16:29:55 UTC+3, jaka.k...@gmail.com wrote: > > Current Beaglebone Black images offer support for high resolution timers > in Linux kernel. They are controlled by ktime.h library and depend on > system's clocks (CLOCK_MONOTONIC, CLOCK_REALTIME,...). The clocks can be > looked up from userspace programs with commands clock_gettime > <https://linux.die.net/man/2/clock_gettime>, clock_settime and > clock_getres from time.h library. > What I would like to do is make/write a protocol in c that would > synchronize realtime clocks between two boards over ethernet cable > connection or internet. My goal is testing how accurately two clocks can be > set to the same global time under different protocols. I'm planning to > start with ntp, would using an existing library or client make sense here? > I read BBB sometimes comes with daemons for ntp, in that case how would I > set one board to act as a server in protocol? > Note that the Beaglebone Debian images use a small systemd service called systemd-timesyncd to set the system RTC. This service queries current time via SNTP (on boot, and periodically while the system is running) and updates /dev/rtc0 with the received time. Perhaps you can get started with that simple, out-of-box feature? https://wiki.archlinux.org/index.php/systemd-timesyncd https://www.freedesktop.org/software/systemd/man/systemd-timesyncd.service.html -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/f6e9aae0-eff2-4b88-ba30-44f79d428a0do%40googlegroups.com.
Re: [beagleboard] Re: Graphical Desktop with keyboard & display
On Monday, 15 June 2020 16:57:30 UTC+3, evilwulfie wrote: > > I am fairly sure he has a display manager installed as VNC will not > function > without one. > AFAIK VNC can be configured to work either independently or with a display manager. Depends on how Bernard has configured the VNC server on his BB, which he didn't specify. If he connects with VNC directly to a display manager, e.g. GDM, then yes he should see the same GDM login window on native terminal. If it's not there, perhaps it helps to walk through the virtual terminals Ctrl-Alt-F1 to Ctrl-Alt-F7 - it might show up somewhere. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/a032cce6-37af-40f3-b6d6-8b0a924d8251o%40googlegroups.com.
[beagleboard] Re: Graphical Desktop with keyboard & display
Hi Bernard On Monday, 15 June 2020 00:13:12 UTC+3, Bernard Fischer wrote: > > I have vnc working fine with a GUI on my BB AI. But if it plug in a > display and keyboard/mouse i get a text terminal and can't start up a > graphical desktop. startx is not found. > > What packages do I need to install for that? If it works over VNC why > doesn't it work directly? > One of the easiest ways to get started with graphical desktops in Debian is to install a "task-XXX-desktop" metapackage for your favourite desktop environment XXX, which pulls in the whole kaboodle via dependencies. It will probably include some packages which you don't want, so feel free to prune later. E.g. for installing the light-weight LXDE environment you'd do this: $ sudo apt install task-lxde-desktop The reason you don't see a graphical environment when plugging in a monitor is you don't have a display manager installed ( https://wiki.debian.org/DisplayManager). The task you install will include a "friendly" display manager, e.g. in case of LXDE it'll be the lightdm. You might need to restart the board (or at least the display manager service) after you've plugged in a monitor post-boot. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/1abc43d9-631e-4360-ba5e-d39838bcab4bo%40googlegroups.com.
Re: [beagleboard] Re: File system switches to Read-Only randomly after booting.
On 09.06.20 12:18, m...@schulzd.me wrote: I tried a 12V 5A switch mode power supply - still no dice. Be careful, the BBB takes 5V not 12V. I even flashed it using this PSU, switched to USB to add WiFi credentials, and then put it back to 12V and SSH'd in via WiFi. Still has the same issue with read-only. I really think it must be a hardware problem, I might go have a look at the board gerbers/cad files and see if I can spot anything, perhaps it could be something as benign as the eMMC/Wifi chips having being on the same power plane (most likely I assume?), and when WiFi requires more current during a transmission, it's unable to source enough, which could cause voltage to drop or not enough current supplied to eMMC, which then causes the error? This would be why adding a beefier PSU would help I guess - but if for some reason the traces are really thin or something, or not enough vias or not not enough shielding... who knows :-) I guess the absolute very last thing you could try, would be clearing the eMMC and re-writing bootloader and everything from scratch - but what's the point? This issue *ALSO* happens when I'm booted to the SD card - so to me that seems like it's something with the WiFI chip causing havoc in other parts of the board (maybe RAM/CPU ??) Thanks for the help anyway guys! Strange, I've poked one BBB Wireless a few years ago and I don't recall such issues. I vaguely recall the eMMC flaking out on regular BBB-s when I powered them from a bad power supply and supplied a full 3G GSM modem directly from its USB (which was simply a bad idea and far exceeded the specs of the USB supply on BBB). -- Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/5058f70b-a837-e850-3fb7-1a5f205a5fe0%40gmail.com.
Re: [beagleboard] Re: File system switches to Read-Only randomly after booting.
Hi, On 08.06.20 03:21, m...@schulzd.me wrote: Alright seems like I spoke too soon again: during an upgrade I got this: Warning: Stopping cloud9.service, but it can still be activated by: cloud9.socket c9-core-installer:Stopping:cloud9.socket c9-core-installer:extracting:c9-core_3.1.3543+git20170407-v4.8.2-build.tar.xz groupadd: group 'cloud9ide' already exists c9-core-installer:Installed Setting up ti-pru-cgt-installer (2.1.5-0rcnee1~jessie+20180514) ... --2020-06-08 00:17:34-- http://downloads.ti.com/codegen/esd/cgt_public_sw/PRU/2.1.5/ti_cgt_pru_2.1.5_armlinuxa8hf_busybox_installer.sh Resolving downloads.ti.com (downloads.ti.com)... 23.37.140.249 Connecting to downloads.ti.com (downloads.ti.com)|23.37.140.249|:80... connected. HTTP request sent, awaiting response... 200 OK Length: 37320342 (36M) [application/x-sh] Saving to: ‘ti_cgt_pru_2.1.5_armlinuxa8hf_busybox_installer.sh’ ti_cgt_pru_2.1.5_armlinuxa8hf_busybox_insta 13%[> ] 4.98M 270KB/s in 12s Cannot write to ‘ti_cgt_pru_2.1.5_armlinuxa8hf_busybox_installer.sh’ (Input/output error). Bus error dpkg: error processing package ti-pru-cgt-installer (--configure): subprocess installed post-installation script returned error exit status 135 Connection to 192.168.7.2 closed by remote host. Connection to 192.168.7.2 closed. So looks like it's still the same problem I would assume. I will try again and with 256kb min memory, but to be honest this looks more and more like something really wrong with the hardware - very unreliable and certainly would not give me confidence if this was running a drone or anything in the air. Getting on some thin ice now, but the next best guess would be a flaky power supply. Unplug devices connected to BB-s USB ports, supply power to BB via the 5V barrel plug (and a quality power brick) or from a good powered USB hub. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/80007d53-7f49-c95b-9052-25eb91c66605%40gmail.com.
[beagleboard] Re: File system switches to Read-Only randomly after booting.
Hi Phil, On Wednesday, 13 May 2020 16:33:40 UTC+3, robotsrule...@gmail.com wrote: > > Any suggestions??? > If it's running off the internal eMMC, it should be quite reliable. The most obvious advice is to look at the system logs after the problem has occurred and try to find hints of what has caused the problem. Certainly the kernel will leave a line there when a file system is remounted read-only - something similar to "kernel: Remounting filesystem read-only". So have a look at /var/log/syslog and see what you can find. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/9248737f-0c83-49f8-ab03-2bc9a393%40googlegroups.com.
[beagleboard] Re: BBB Won't Shutdown
Hi Ken, On Friday, 3 April 2020 17:44:16 UTC+3, KenUnix wrote: > > > I recently added an external 5V supply to the BBB barrel jack because I am > going to add a relay cape. > > If I issue "sudo shutdown -h now" the BBB looks like it is going to > shutdown judging by the messages on the Jtag port but then re-boots? > See messages below. > > I am using USB connectivity. If I unplug the 5V adapter and issue "sudo > shuwdown -h now:" it works. > > [ OK ] Reached target Shutdown. > [ OK ] Reached target Final Step. > [ OK ] Started Power-Off. > [ OK ] Reached target Power-Off. > [44567.694668] reboot: Power down <=== This is where it should power off > > U-Boot SPL 2019.04-2-gbb4af0f50f (Jul 08 2019 - 11:44:39 -0500) <=== > This is what happens > Trying to boot from MMC2 > > U-Boot 2019.04-2-gbb4af0f50f (Jul 08 2019 - 11:44:39 -0500), Build: > jenkins-github_Bootloader-Builder-128 > > CPU : AM335X-GP rev 2.1 > I2C: ready > DRAM: 512 MiB > No match for driver 'omap_hsmmc' > No match for driver 'omap_hsmmc' > Some drivers were not found > Reset Source: Power-on reset has occurred. > RTC 32KCLK Source: External. > MMC: OMAP SD/MMC: 0, OMAP SD/MMC: 1 > > BeagleBoard.org Debian IoT Image 2020-03-26 > Linux beaglebone 4.19.94-ti-r41 #1buster SMP PREEMPT Sat Mar 21 03:13:40 > UTC 2020 armv7l GNU/Linux > I've seen similar behaviour on a subset of my BBB-s and BeagleCores - perhaps 1-3 out of ten just refused to shut down when powered via 5V supply input. Hasn't bothered me enough to dig in, but since it's specific to a few hardware specimens, it doesn't look like a software issue. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/051281a7-58b4-4748-8c6b-c3aaad8c8105%40googlegroups.com.
[beagleboard] Re: [Beaglebone Enhanced][VAYU2-4GEC25-E][Quectel] - Not work mobile connection if ethernet wire is plug
Hi Nicola, On Thursday, 30 January 2020 17:26:09 UTC+2, Nicola Russo wrote: > > I'm using the cape model: VAYU2-4GEC25-E ( > https://www.yantrr.com/product/vayu2-4g-ltecat4asia-eu-w-gps/) together > with a Beaglebone Enhanced that running on Debian 9.5 2018-10-07 4GB eMMC > IoT . > > I started ppp connection as described below with wvdial: > -- > debian@beaglebone:~$ sudo wvdial & > > [1] 1288 > debian@beaglebone:~$ --> WvDial: Internet dialer version 1.61 > --> Initializing modem. > --> Sending: ATZ > ATZ > OK > --> Sending: ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0 > ATQ0 V1 E1 S0=0 &C1 &D2 +FCLASS=0 > OK > --> Sending: at+cgdcont=1,"ip","[web.omnitel.it](http://web.omnitel.it/)" > at+cgdcont=1,"ip","[web.omnitel.it](http://web.omnitel.it/)" > OK > --> Modem initialized. > --> Sending: ATDT*99# > --> Waiting for carrier. > ATDT*99# > CONNECT 15000 > --> Carrier detected. Starting PPP immediately. > --> Starting pppd at Mon Jan 20 09:30:05 2020 > --> Pid of pppd: 1290 > --> Using interface ppp0 > --> pppd: ▒^▒(9t[01][18]9t[01] > --> pppd: ▒^▒(9t[01][18]9t[01] > --> pppd: ▒^▒(9t[01][18]9t[01] > --> pppd: ▒^▒(9t[01][18]9t[01] > --> pppd: ▒^▒(9t[01][18]9t[01] > --> pppd: ▒^▒(9t[01][18]9t[01] > --> local IP address 5.91.117.101 > --> pppd: ▒^▒(9t[01][18]9t[01] > --> remote IP address 10.64.64.64 > --> pppd: ▒^▒(9t[01][18]9t[01] > --> primary DNS address 10.133.18.210 > --> pppd: ▒^▒(9t[01][18]9t[01] > --> secondary DNS address 10.132.100.181 > --> pppd: ▒^▒(9t[01][18]9t[01] > ^C > debian@beaglebone:~$ ^C > debian@beaglebone:~$ ^C > debian@beaglebone:~$ ping [google.com](http://google.com/) > PING [google.com](http://google.com/) (216.58.208.142) 56(84) bytes of > data. > 64 bytes from [lhr25s08-in-f14.1e100.net]( > http://lhr25s08-in-f14.1e100.net/) (216.58.208.142): icmp_seq=1 ttl=50 > time=47.7 ms > 64 bytes from [lhr25s08-in-f14.1e100.net]( > http://lhr25s08-in-f14.1e100.net/) (216.58.208.142): icmp_seq=2 ttl=50 > time=56.0 ms > 64 bytes from [lhr25s08-in-f14.1e100.net]( > http://lhr25s08-in-f14.1e100.net/) (216.58.208.142): icmp_seq=3 ttl=50 > time=48.9 ms > > -- > > After I connected Ethernet wire of my subnet at beaglebone, so the mobile > connection doesn't work anymore. Only if I disconnect Ethernet wire, the > mobile connection works. > > In the Debian System Is It possible to set a connection priority between > 4G connection and internet connection by LAN wire? > > I need to use 4g connection and ethernet port for TCPIP communications > simultaneously. > In which configuration file can I set this priority? The word to google for is network manager. Without a network ethernet and dialup are implemented by a set of bash scripts which don't consider each other and step on each other's toes. Typically the most annoying thing that happens is the connection which is established last will blindly overwrite DNS configuration for the previous. It might also overwrite the default route (or worse - NOT overwrite the default route in which case its provider-private DNS servers are likely unreachable from the other connection which effectively kills the Internet connection). There are many ways for this to fail. You need a network manager which is aware of both the Ethernet and dialup connections and can juggle the DNS servers, default routes and other resources according to configuration that you specify. AFAIK that they operate according to your requirements in tandem with their "friendly" dialup manager, so ppp and wvdial cannot be used. Gnome project has NetworkManager with dialup companion ModemManager. This is intended for desktop usage and might be a wee bit demanding on resources or required libraries. I know very little about those two, so I might be wrong The other is connman with dialup companion ofono. This is intended for lightweight embedded systems. Connman is likely installed on your system already, you just need to set up ofono to do the dialup. The downside is that both (especially ofono) are rather poorly documented and getting the whole thing up and running in tandem is a bit frustrating for a new user. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/87c5b46b-d9cf-4ba1-b87d-b4882eba3339%40googlegroups.com.
[beagleboard] Re: Can't ping or ssh into BBB Wireless
Hi Zeph, On Wednesday, 15 January 2020 20:29:02 UTC+2, Zeph wrote: > > I have two boards a BeagleBone Black Wireless and a BeagleBone Black Rev > C. Recently, when booting my BeagleBone Black Wireless with an Ubuntu SD, > the board will light up and in my Windows 10 computer's network connections > the Remote NDIS Compatible Device shows up but pinging 192.168.7.2 does not > give any response and I can't SSH into the board anymore. If I put an SD > with debian into the BBB Wireless, I can SSH into it fine. Also, if I take > the Ubuntu SD and use it in my BeagleBone Black Rev C I can ping it and SSH > into it. The only combination that doesn't work is the Ubuntu SD with the > BBB Wireless. > > Can anyone give insight on what the issue might be? I think this error may > have started to appear when I was changing network settings to try and use > internet sharing through USB. > If you've changed the network settings and can't figure out how to change them back, it would be a rather long and unproductive thread to remotely debug the situation via this newsgroup :) I'd suggest starting from scratch with a new Ubuntu image. If you can't spare to lose the data on it, copy it somewhere else. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/35861bfd-434d-43d3-9c3e-8bdd26837c64%40googlegroups.com.
Re: [beagleboard] Ip forwarding
Hi Josiah! On Monday, 16 December 2019 16:00:20 UTC+2, Josiah Akinloye wrote: > > Thanks Gwen. > I really appreciate your effort. > I will get back to you > Again... a big thanks > > On Mon, 16 Dec 2019 at 14:56 Gwen Stouthuysen > wrote: > >> No, I haven't tried to remove connmanctl >> >> I have no problem with the application, as it does exactly what I want it >> to do. >> >> There is quite some configuration behind the curtains, dive deeper into >> the configuration files and you might find the settings you want to alter. >> >> Gwen >> >> Op ma 16 dec. 2019 om 14:17 schreef Josiah Akinloye > >: >> >>> Thanks Gwen, Do you have any idea how I can manage the WiFi network on >>> beagle bone wireless without using CONNMANCTL >>> >>> Kind regards, >>> Josiah >>> >>> On Mon, 16 Dec 2019 at 13:55 Gwen Stouthuysen >> > wrote: >>> >>>> Here is some explication/tutorial on bridging the two networks in your >>>> BBW >>>> >>>> https://wiki.debian.org/BridgeNetworkConnections >>>> >>>> Regards, >>>> >>>> Gwen >>>> >>>> Op ma 16 dec. 2019 om 13:49 schreef Josiah Akinloye < >>>> josiaha...@gmail.com >: >>>> >>>>> Thanks for your reply, I have been able to get internet connectivity >>>>> from the modem using MBIM and also I have created a WiFi gateway + DHCP >>>>> server using COMMANCTL but where I have problem is the bridging the >>>>> gateway >>>>> with the network port offered by the 4g modem. Thanks again >>>>> >>>>> On Mon, 16 Dec 2019 at 11:58 Gwen Stouthuysen >>>> > wrote: >>>>> >>>>>> Let me try to explain what you will have to do: >>>>>> >>>>>>- Connect the module to the BBBW. >>>>>> - It has USB drivers so it should be straight forward to >>>>>> enable it as a network node. >>>>>> - A small cape could be designed to do this, RPi hats do exist >>>>>> - >>>>>> >>>>>> https://www.elementzonline.com/blog/Accessing-internet-with-SIM7600-4G-Modem-using-MBIM-interface-in-Pocketbeagle-Beaglebone >>>>>> >>>>>> - Set up a WiFi gateway (Wifi access point credentials + DHCP >>>>>>server for this port) >>>>>>- Bridge the gateway with the network port offered by the >>>>>>SIM7600e driver >>>>>>- Set up a firewall to avoid malicious access from both WiFi and >>>>>>Cellular >>>>>> >>>>>> It is a bit more complex than forwarding the IP address, but it is a >>>>>> well documented setup. >>>>>> >>>>>> But you are lucky: the standard Debian OS prepares the WiFi as an >>>>>> access point, you just have to arrange the SIM7600e integration and the >>>>>> bridge & Firewall >>>>>> >>>>> When both your dialup connection and Wifi are managed by connman, you can use its built-in tethering functionality to share Internet from first to the other. If I'm interpreting connmanctl v1.36 documentation correctly (connmanctl --help), the command would look something like this: $ sudo connmanctl tether wifi on MY_SSID MY_PASS There might be some requirements, i.e. connman might assume it has established the dialup connection (as opposed to MBIM). I haven't experimented this much. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/0c404225-528a-4242-b887-a3cf795a7c0d%40googlegroups.com.
Re: [beagleboard] Re: Bash throws errors after a week of uptime
On Tuesday, 10 December 2019 19:48:11 UTC+2, bryan...@owasp.org wrote: > > I have been monitoring the space. I thought it was /tmp but the usage is > super low: > > debian@radius:~$ df > Filesystem 1K-blocksUsed Available Use% Mounted on > udev 219552 0219552 0% /dev > tmpfs 494966156 43340 13% /run > /dev/mmcblk0p1 30714340 4244892 25170228 15% / > tmpfs 247476 0247476 0% /dev/shm > tmpfs 5120 4 5116 1% /run/lock > tmpfs 247476 0247476 0% /sys/fs/cgroup > tmpfs 49492 0 49492 0% /run/user/1000 > > >> The device /dev/mmcblk0 is usually the SD card. Also the BBB has 4 GB of eMMC, not 32 GB as available your root file system. I still suspect that you're running off the SD card, and it's failing :) -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/bf7403e1-1258-46b6-bcfb-b09d115cee8b%40googlegroups.com.
[beagleboard] Re: Bash throws errors after a week of uptime
On Tuesday, 10 December 2019 00:28:30 UTC+2, bryan...@owasp.org wrote: > > > I am running Debian 9.11 (4.14.108-ti-r124) on BBB and it runs fine for > about a week and then Bash no longer can find commands and it throws errors > upon every subsequent command. Commands result in one of two errors > depending upon the command. Here are two recent examples: > > $debian:/opt/tools$ ping google.com > > -bash: /bin/ping: Input/output error > > $debian:/opt/tools$ reboot > -bash: reboot: command not found > > Has anyone else seen this problem? I am unsure where to create issues for > Debian Stretch on ARM/BBB. > > Looks like a busted SD card or eMMC (depending on where your root file system resides). -- Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/bf83b06a-de3b-447f-92c9-0b061b6c11c1%40googlegroups.com.
Re: [beagleboard] dtb-rebuilder vs. BeagleBoard-DeviceTrees on BeagleBone Black
On 10.10.19 17:41, Robert Nelson wrote: On Thu, Oct 10, 2019 at 6:17 AM Tarmo Kuuse wrote: On 09.10.19 23:57, Robert Nelson wrote: On Wed, Oct 9, 2019 at 11:31 AM Robert Nelson wrote: Here's the final version.. https://github.com/beagleboard/linux/releases/tag/4.19.73-ti-r29 What else would be useful? ;) That's beautiful, thank you! So yeah, i test what i can, can't test everything especially if Ubuntu or Debian throws in a wrench with a broken gcc version or something else in their testing repos.. Ok, so you verify that the kernel builds, then package and publish the result and finally run some integration tests using real hardware. Sounds great. Question: when you run the weekly or daily integration tests on your little farm, how can I consume only the kernel packages that passed? This file get's updated.. https://github.com/beagleboard/image-builder/blob/master/configs/kernel.data#L7 Good to know. Thanks. Now, before we go more into the details, i don't' "guarantee" anything.. If your worried about something breaking, have a test node you test with before you push out to many units.. Whaddya mean, test? https://thedailywtf.com/articles/Flawless-Compilation -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/b9745bc8-9845-6f54-a8ab-8f053be4f909%40gmail.com.
Re: [beagleboard] dtb-rebuilder vs. BeagleBoard-DeviceTrees on BeagleBone Black
On 09.10.19 23:57, Robert Nelson wrote: On Wed, Oct 9, 2019 at 11:31 AM Robert Nelson wrote: Here's the final version.. https://github.com/beagleboard/linux/releases/tag/4.19.73-ti-r29 What else would be useful? ;) That's beautiful, thank you! So yeah, i test what i can, can't test everything especially if Ubuntu or Debian throws in a wrench with a broken gcc version or something else in their testing repos.. Ok, so you verify that the kernel builds, then package and publish the result and finally run some integration tests using real hardware. Sounds great. Question: when you run the weekly or daily integration tests on your little farm, how can I consume only the kernel packages that passed? -- Kind regards Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/34da0a8f-506f-b9e5-a5f2-8737ad33cba2%40gmail.com.
Re: [beagleboard] dtb-rebuilder vs. BeagleBoard-DeviceTrees on BeagleBone Black
Hi Robert, On 08.10.19 17:34, Robert Nelson wrote: Thank you. That's good to know. A kernel upgrade on our devices requires using the dtb-rebuilder to get our modified base DTB-s (am335x-boneblack.dtb). For the 4.19-ti series the most recent supported kernel version in your repo appears to be "4.19.25-ti-r17" from 6 months ago (when looking at commit comments). The beagleboard repo doesn't list any kernel versions anywhere. When using this repo, how would I recognize which 4.19-ti kernel version I'd be able to apply my modified DTB-s to? That shouldn't be an issue going forward.. (easy to say as 4.19-ti has quieted down, so that might backfire with 5.4-ti) With the old dtb-rebuilder repo, we'd always push a dts patch to the kernel build script first, tag it and then manually copy it back to the dtb-rebuilder. So things would get out of sync if i didn't copy it over.. With the new repo, everything gets pushed to the BeagleBoard-DeviceTrees repo first, and the files get sync'ed directly back to the kernel build script afterwards: https://github.com/RobertCNelson/ti-linux-kernel-dev/blob/ti-linux-4.19.y/patch.sh#L353-L409 It should be less error prone. I hope.. Ok, the patch flow process seems sound. But I have a rather simpler question. How would I match a specific kernel deb package to a BeagleBoard-DeviceTrees repo commit which went into _this_ package? I suspect I can't randomly pick a set of DTB files meant for one kernel, modify and then apply those to another kernel. E.g. I've chosen linux-image-4.19.50-ti-r24 as my kernel (from your apt repo). Now I wish to modify the DTB-s corresponding to _this_ kernel package. Not for 5.x, not for some newer or older release of 4.19-ti. How would I find the DTB-s in this package from the BeagleBoard-DeviceTrees repo? Off-topic: I'm also assuming that the kernel packages in repos.rcn-ee.com are all tested and good for production use, which may or may not be correct. -- Lugupidamisega/Kind regards Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/9c5f9a4c-5a63-2419-18ae-28e3ffd83656%40gmail.com.
Re: [beagleboard] dtb-rebuilder vs. BeagleBoard-DeviceTrees on BeagleBone Black
Hi Robert, On Tuesday, 8 October 2019 04:44:25 UTC+3, RobertCNelson wrote: > > On Mon, Oct 7, 2019 at 8:28 PM Mark A. Yoder > wrote: > > > > I've playing with device trees on the BeagleBone Black and have come > across two repos that appear to do similar things. > > https://github.com/RobertCNelson/dtb-rebuilder.git > > > > and > > https://github.com/beagleboard/BeagleBoard-DeviceTrees.git > > > > What's the difference between the two? I've been using the second to > experiment with different device trees. Should I be using the rebuilder? > > Hi Mark, > > I'm EOL'ing mine, the beagleboard one should be use for all current > and future dts patches. > > Regards, > > -- > Robert Nelson > https://rcn-ee.com/ > Thank you. That's good to know. A kernel upgrade on our devices requires using the dtb-rebuilder to get our modified base DTB-s (am335x-boneblack.dtb). For the 4.19-ti series the most recent supported kernel version in your repo appears to be "4.19.25-ti-r17" from 6 months ago (when looking at commit comments). The beagleboard repo doesn't list any kernel versions anywhere. When using this repo, how would I recognize which 4.19-ti kernel version I'd be able to apply my modified DTB-s to? -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/80a029a5-048a-457f-98f9-1efd59ee7269%40googlegroups.com.
[beagleboard] Re: Simcard or SMS module
Hi! On Friday, 27 September 2019 00:09:12 UTC+3, z.mome...@gmail.com wrote: > > Hello! i want to use of beagle bone black in my smart home project, i want > to send sms from an android application to a simcard or sms module that > connected to beagle bone, but i don't know sms module for this, is there > any simcard or sms module for this usage? > Thanks! > I've used Mikroelektronika Klick GSM modules, which require an additional adaptor board. E.g. those two: https://www.mikroe.com/beaglebone-mikrobus-cape https://www.mikroe.com/gsm-2-click It's kind of clumsy, but works. You'll want to supply a few amps of power through the barrel connector instead of the MicroUSB connector, as the GSM module will occasionally consume quite a bit of energy. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/cf988fcc-3d10-4bc6-87d6-adfa4fa0357a%40googlegroups.com.
[beagleboard] Re: Cannot ping google in BBB even after sharing internet
Hi Brian, On Tuesday, 3 September 2019 18:44:50 UTC+3, Brian Tremaine wrote: > > I realize this is an old post but I'm hoping to get some answers ;) : > Problems pinging www.google.com > > I have a BeagleBone Black Rev C that is running the latest Debian. I've > been going through Derek Malloys book and video to get things started > (excellent material BTW), > > I am connected to a Win 10 laptop (64-bit) using the USB interface and am > using putty to connect through SSH. My problem is I cannot ping a DNS name, > i.e. www.google.com. pinging 8.8.8.8 does work. When I try to edit /etc/ > resolv.conf it says file does not exist. On the terminal the file > resolv.conf is shown in red; What does this mean? > > Can anybody help with hints on how to set up name server on BB Black Rev C > with Debian? > On my BBB "/etc/resolv.conf" is a symbolic link which points to connman's /run/connman/resolv.conf: $ ls -lha /etc/resolv.conf lrwxrwxrwx 1 root root 24 Jun 30 13:12 /etc/resolv.conf -> /run/connman/resolv.conf In your case - if the terminal shows this file in red, it likely indicates a broken symlink, i.e. it points to file a file which doesn't exist. You can verify this by listing the file being pointed to and verifying its existance. In my case (my link is OK): $ ls -lha /run/connman/resolv.conf -rw-r--r-- 1 root root 85 Sep 4 08:24 /run/connman/resolv.conf As a temporary solution you can delete the broken symlink and add your own name server manually $ sudo rm /etc/resolv.conf $ sudo vim /etc/resolv.conf nameserver 8.8.8.8 As a permanent solution, try to figure out why connman hasn't managed to do its job :) -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/44a6bba7-a973-43fd-a1b5-0cbee4abb6c9%40googlegroups.com.
Re: [beagleboard] Re: Getting IP address of BBB on ubuntu terminal
On 12.07.19 07:50, Megha Bhirade wrote: I tried by connecting LAN cable from router to board Ethernet port, but for ping command i am getting : acufore@acufore-H110M-S2:~$ ping beaglebone.local ping: unknown host beaglebone.local acufore@acufore-H110M-S2:~$ avahi-resolve -n beaglebone.local Failed to resolve host name 'beaglebone.local': Timeout reached Your computer can't reach the beaglebone. I'm out of simple advice. Assuming your LAN setup is OK, it's rather suspicious that neither USB nor Ethernet gives you access to it. The USB connection also creates a virtual serial port through which a serial console is available (which is nearly is as good as an ssh connection). On my machine it appears as device /dev/ttyACM0. I use it by installing the package "screen" and running: $ screen /dev/ttyACM0 115200 Pressing Enter causes a login prompt to appear, where you can log in using the default credentials (debian/debian, IIRC). If this doesn't work, I would try re-flashing the board. Follow the instructions here: https://beagleboard.org/Support/BoneScript/updates/ Note that you're debugging it blindly which makes everything more difficult. If you attach a USB-Serial adapter to the debug header, you'll get the bootloader debug output and a serial console. If you don't have one at hand, search aliexpress for "FTDI USB to TTL" and get the first thing that pops up. -- Kind regards Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/e717d54a-4a91-c975-d7ce-794f66b00521%40gmail.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] Re: Getting IP address of BBB on ubuntu terminal
On 11.07.19 15:48, Megha Bhirade wrote: Thanks for reply. for second option -connecting LAN cable to BBB . i connected LAN cable from router with DHCP server, where i can see :beaglebone.local.. can u explain the basic connection : i followed like , 1.connected board(BBB) to PC using USB cable 2. router with DHCP server 3. one LAN cable from the router to board ethernet port Well, there's nothing to "see" as such. Hopefully the BBB is now reachable via your local network. Your next steps depend on what are you planning to do with it, and this only you know. I assumed you wanted to log into it via ssh and hack away. You can perform some basic network diagnostics using your PC (it must be connected to the same LAN, obviously). To check if name resolution works (returns the IPv4 or IPv6 address of the BBB): $ avahi-resolve -n beaglebone.local To check if the BBB is reachable from your computer $ ping beaglebone.local -- Kind regards Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/ce8c9bd6-2452-02f4-c18e-607785ebe181%40gmail.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: Getting IP address of BBB on ubuntu terminal
On Thursday, 11 July 2019 14:34:48 UTC+3, Megha Bhirade wrote: > > > >> Hi tarmo, >> >i am using a PC with Linux Ubuntu and version is 16.04LTS (64-bit) > Ubuntu as a native operating system. > > As u suggested i used *tail -n 100 /var/log/syslog *in to command > line(terminal). i got much data not able to understand and i stored the > data in to one file please check it i am attaching the same data.. > > one more thing u told about the local LAN can u explain it once?? > OK. I have no experience using Ubuntu 16.04LTS, but I'd assume it contains all necessary drivers for BBB out of the box. The 2 minute log snippet does not contain anything which I would recognize as a BBB being plugged into USB. Perhaps you did it earlier? Anyway, your computer appears to try getting an IP from the BBB-s DHCP server, but there's no response. No idea why. Usually my computer gets a 192.168.6.1 or 192.168.7.1 (actually both, on two separate virtual interfaces) address from the BBB (as described in https://beagleboard.org/static/beaglebone/latest/README.htm). The BBB has an Ethernet port. Connect it to your LAN (a router with a DHCP server and other standard services is required, obviously). Then you can connect to it from the same LAN. It should respond to name "beaglebone.local" by default. If this name doesn't resolve then install avahi packages in Ubuntu. Then use your favourite ssh client to connect. Not much more to explain there. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/a282ace5-5e32-4ec7-bd5c-dbe57921cf06%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: Cape GPIO usage -- are there reserved pins?
Hi Robert, On Wednesday, 10 July 2019 05:32:06 UTC+3, Robert Heller wrote: > > OK, I have designed a cape, a Quad Stallmotor controller with/sense, which > uses 8 GPIO lines: four as outputs to drive a "stallmotor" type of switch > motor for model railroad turnouts [switch tracks] and four inputs to sense > the current state of the stallmotor (the state of the switches). > > The I/O pins I am using are: > > // GPIO0_7: P9-42 > // GPIO1_6: P8-3 > // GPIO1_28: P9-12 > // GPIO1_7: P8-4 > > > // GPIO1_16: P9-15 > // GPIO1_17: P9-23 > // GPIO3_21: P9-25 > // GPIO3_19: P9-27 > > I am testing the cape and I having some issues and wanted to check if > there > are problems with using these pins, partitularly GPIO1_6 on P8-3 and > GPIO1_7 > on P8-4 and I am slightly confused about GPIO0_7 on P9-42 (P9-42 seems to > be > listed as P9-42a and P9-42b). > > I am getting kernel error messages relating to mmcblk1 (I am not using > mmcblk1 > and I have zeroed it). I *think* this might be because what is "wired" to > P8-3 and P8-4. > > Should I be looking at redesigning my cape to use other GPIO lines? > > -- > Robert Heller -- 978-544-6933 > Deepwoods Software-- Custom Software Services > http://www.deepsoft.com/ -- Linux Administration Services > hel...@deepsoft.com-- Webhosting Services > I recommend Derek Molloy's GPIO tables for figuring out which pin is used for what. Out of experience I would recommend not using for GPIO a pin which has anything written in the "Notes" column. https://github.com/derekmolloy/boneDeviceTree/blob/master/docs/BeagleboneBlackP8HeaderTable.pdf https://github.com/derekmolloy/boneDeviceTree/blob/master/docs/BeagleboneBlackP9HeaderTable.pdf -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/1b015b3d-e82a-44e6-a00f-5cf90677a84c%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: Getting IP address of BBB on ubuntu terminal
Hi Megha, On Thursday, 11 July 2019 10:09:25 UTC+3, Megha Bhirade wrote: > > Hi, > > i am new to beagle bone black board. just after getting new board i > connected that board to my PC using USB cable and checked it is showing > BEAGLE BONE like mass storage drive on system(linux ubuntu), i clicked and > entered in to drive and got one START.HMT file clicked on it, it went to > website and showed next step install drivers.. > > i downloaded the mkudevrule.sh file and run it by using chmod + x(for > permission) , sudo command , it completed... > > i got 73-beagle bone.rules file is created in /etc/udev/rules.d/... after > this i concluded drivers are installed properly.. > > i am trying to get ip address of BBB , but not able to get?? please > suggest me what is problem: > > > acufore@acufore-H110M-S2:~$ cd /etc/udev/rules.d/ > acufore@acufore-H110M-S2:/etc/udev/rules.d$ ls > 73-beaglebone.rules > acufore@acufore-H110M-S2:/etc/udev/rules.d$ cd > acufore@acufore-H110M-S2:~$ ifconfig > enp2s0Link encap:Ethernet HWaddr e0:d5:5e:11:5d:65 > inet addr:192.168.4.59 Bcast:192.168.4.255 Mask:255.255.255.0 > inet6 addr: fe80::4508:cbb4:d574:e0c1/64 Scope:Link > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 > RX packets:30922 errors:0 dropped:0 overruns:0 frame:0 > TX packets:13579 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:1000 > RX bytes:16593168 (16.5 MB) TX bytes:1669767 (1.6 MB) > > enx3403de92aa91 Link encap:Ethernet HWaddr 34:03:de:92:aa:91 > UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 > RX packets:41 errors:3 dropped:0 overruns:0 frame:3 > TX packets:2 errors:58 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:1000 > RX bytes:10411 (10.4 KB) TX bytes:520 (520.0 B) > > loLink encap:Local Loopback > inet addr:127.0.0.1 Mask:255.0.0.0 > inet6 addr: ::1/128 Scope:Host > UP LOOPBACK RUNNING MTU:65536 Metric:1 > RX packets:3169 errors:0 dropped:0 overruns:0 frame:0 > TX packets:3169 errors:0 dropped:0 overruns:0 carrier:0 > collisions:0 txqueuelen:1 > RX bytes:324659 (324.6 KB) TX bytes:324659 (324.6 KB) > > acufore@acufore-H110M-S2:~$ ^C > acufore@acufore-H110M-S2:~$ > > like this i am getting how can i proceed for next..? > So you have a PC with Ubuntu? Which version? Are you running Ubuntu as a native operating system or from a virtual machine? Connect the BBB using USB, dump logs (e.g. using "tail -n 100 /var/log/syslog") and post here for further analysis. In the mean time - as a quick workaround I would connect the BBB to your local LAN using Ethernet. Then you'd find it from your PC using the name beaglebone.local. E.g. "ssh debian@beaglebone.local". -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/844fa60a-2f2e-497c-ab6b-c33115aeee19%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: How to identify GPIO Chip and pin
Hi! On Wednesday, 3 July 2019 04:51:55 UTC+3, simonm...@gmail.com wrote: > > I am trying to use GPIO with the robot control library. > But I cant find the value to pass to chip and pin > > ex : > int rc_gpio_init > <http://www.strawsondesign.com/docs/librobotcontrol/group___g_p_i_o.html#ga8276c43b1cb0aad4f7b4d08deeb7e893> > > (int chip, int pin, int handle_flags) > I've used Derek Molloy's excellent GPIO tables for reference. Regarding your question column "Mode 7" would seem to indicate chip and pin numbers, e.g. gpio1[6] would be chip 1, pin 6. https://github.com/derekmolloy/boneDeviceTree/tree/master/docs -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/1423737c-e4d5-4a84-8d7b-ac1a1b7efe91%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: gdbserver install failed
Hi George, Firstly, I didn't see an "apt update" command in your listing. This updates the package lists and should be executed before installing any packages. > Unpacking libtiopencl1 > (01.01.17.01-git20181129.0-0rcnee0~stretch+20190103) ... > dpkg: error processing archive > /var/cache/apt/archives/libtiopencl1_01.01.17.01-git20181129.0-0rcnee0~stretch+20190103_armhf.deb > > (--unpack): > trying to overwrite '/usr/share/ti/opencl/clc.h', which is also in > package libopencl1 01.01.15.03-r0.0rcnee7~stretch+20180809 > Secondly, there's a conflict between packages "libopencl1" and "libtiopencl1". You need to choose one or the other. Remove the other with "apt purge ...". -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/4b2ec493-3dee-4b97-a380-16d1072ab74f%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: Stack 15.4 Linux gateway using Beagle Bone Black wireless
On Friday, 15 March 2019 16:07:57 UTC+2, Tarmo Kuuse wrote: > > Hi Neil, > > On Friday, 15 March 2019 03:31:48 UTC+2, neil.m...@gmail.com wrote: >> >> >> Hello all, >> >> I am currently working on implementing this project : >> >> http://www.ti.com/lit/ug/tiduci9c/tiduci9c.pdf >> >> I am using a Beagle Bone black wireless instead of the Beagle Bone >> black, and I am using a CC1310 Microcontroller from TI. >> >> Now , I am currently working on the following section in the above link: >> >> 3.1.2.1 BeagleBone Black SD Card. >> >> Here are the steps stated in that sections: >> >> 1. Download the prebuilt TI Linux processor SDK SD card image >> am335x-evm-linux-xx.xx.xx.xx.img.zip from >> http://software-dl.ti.com/processor-sdk-linux/esd/AM335X/latest/index_FDS.html >> (where >> xx.xx.xx.xx is the version number of the latest Linux Processor SDK). >> 2. To program the micro SD memory card, see the instructions in >> Processor SDK Linux Creating an SD Card with Windows >> >> When I navigate to link to try and find the correct image , I fould the >> one below to be the right one: >> >> am335x-evm-linux-05.02.00.10.img.zip >> <http://software-dl.ti.com/processor-sdk-linux/esd/AM335X/latest/exports/am335x-evm-linux-05.02.00.10.img.zip> >> Only >> used when creating an SD Card on Windows 1051208KHowever, the image is >> even over 16GB , and in the hardware requirements they say that 8gb MicroSD >> card is required. Is the above image the right one for this assignment? In >> case it is , then an 8GB MicroSD card is not big enough as specified in the >> assignment. If this is not the image I should be using , I would like some >> help please to find out which one it is. Thanks a lot. >> > > It seems to me that 1051208K is approximately 1 GB. > > Anyway - not knowing anything about images provided by TI - perhaps it > would be easiest to just try it out? > Apologies, the uncompressed image is 16GiB. It looks like you might need a 32 GB SD card for this. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/9461850b-1dcf-4046-99f0-e9a7784d7c7c%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: Stack 15.4 Linux gateway using Beagle Bone Black wireless
Hi Neil, On Friday, 15 March 2019 03:31:48 UTC+2, neil.m...@gmail.com wrote: > > > Hello all, > > I am currently working on implementing this project : > > http://www.ti.com/lit/ug/tiduci9c/tiduci9c.pdf > > I am using a Beagle Bone black wireless instead of the Beagle Bone black, > and I am using a CC1310 Microcontroller from TI. > > Now , I am currently working on the following section in the above link: > > 3.1.2.1 BeagleBone Black SD Card. > > Here are the steps stated in that sections: > > 1. Download the prebuilt TI Linux processor SDK SD card image > am335x-evm-linux-xx.xx.xx.xx.img.zip from > http://software-dl.ti.com/processor-sdk-linux/esd/AM335X/latest/index_FDS.html > (where > xx.xx.xx.xx is the version number of the latest Linux Processor SDK). > 2. To program the micro SD memory card, see the instructions in Processor > SDK Linux Creating an SD Card with Windows > > When I navigate to link to try and find the correct image , I fould the > one below to be the right one: > > am335x-evm-linux-05.02.00.10.img.zip > <http://software-dl.ti.com/processor-sdk-linux/esd/AM335X/latest/exports/am335x-evm-linux-05.02.00.10.img.zip> > Only > used when creating an SD Card on Windows 1051208KHowever, the image is > even over 16GB , and in the hardware requirements they say that 8gb MicroSD > card is required. Is the above image the right one for this assignment? In > case it is , then an 8GB MicroSD card is not big enough as specified in the > assignment. If this is not the image I should be using , I would like some > help please to find out which one it is. Thanks a lot. > It seems to me that 1051208K is approximately 1 GB. Anyway - not knowing anything about images provided by TI - perhaps it would be easiest to just try it out? -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/1c02c408-96c9-420d-9b48-66f5a79644b1%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: full disk space
Hi Erik, On Thursday, 28 February 2019 18:00:46 UTC+2, erikb...@gmail.com wrote: > > thanks for the reply, im a newby at linux. ill try to refer to it as root > partition from now on. > > I flashed the image found from the beaglebones website. > https://beagleboard.org/latest-images > > this one> Debian 9.5 2018-10-07 4GB SD IoT > <https://www.google.com/url?q=https%3A%2F%2Fdebian.beagleboard.org%2Fimages%2Fbone-debian-9.5-iot-armhf-2018-10-07-4gb.img.xz&sa=D&sntz=1&usg=AFQjCNGfdbNv101GQEIlwZ3Ty6_4JvzePg> > > Do you know any alternatives? more cleaner images? > This is the "iot" variant which comes with an assortment of development tools pre-installed. Some of those seem rather specific (don't know the details, sorry) so there's a good chance you wouldn't need them all. You could uninstall those one by one, or you could begin fresh with a "console" image which has almost nothing added the base OS and install whatever packages are relevant. The corresponding image from the 2018-10-07 release would be this one: https://rcn-ee.net/rootfs/bb.org/testing/2018-10-07/stretch-console/bone-debian-9.5-console-armhf-2018-10-07-1gb.img.xz You could also poke around here and pick out something a little fresher (some are more experimental than others, the "keepers" file is a good source of hints as to which release to choose) https://rcn-ee.net/rootfs/bb.org/testing/ -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/a1b2c636-c75e-4280-a328-34f798296fc2%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: full disk space
Hi Erik, On Thursday, 28 February 2019 14:37:22 UTC+2, erikb...@gmail.com wrote: > > My beaglebone black /dev partition is almost full when having a fresh > installation > Is 3.3/3.5GB full normal? > Most people call it the root partition. Anyway, a fresh installation of what, exactly? You're always free to remove stuff you don't need. For example - to have a look at installed packages and their disk usage, I have found this command to be useful: "dpkg-query -W -f '${Installed-Size} ${Package}\n' | sort -n" -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/f646f24c-aaae-4c40-9271-f79dc908c148%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: Flashing Debian Image on SD card Partition
Hi Sharvin, On Friday, 22 February 2019 14:38:05 UTC+2, Sharvin Shah wrote: > > I have an sd card with 2 Partitions. mmcblk0 is the disk and mmcblk0p1, > mmcblk0p2 are the 2 partitions I have created. > > I am trying to flash the Debian OS on the Partition using the dd command. > > dd if=os.img bs=10M of=/dev/mmcblk0p1 > > I am getting the following output > > 3565158400 bytes (3.6 GB, 3.3 GiB) copied, 297 s, 12.0 MB/s 106+1 records > in 106+1 records out 3565158400 bytes (3.6 GB, 3.3 GiB) copied, 296.587 s, > 12.0 MB/s > > Now when I try to mount the partition using mount -t auto /dev/mmcblk0p1 > mydir > > I am getting the following error: > > mount: wrong fs type, bad option, bad superblock on /dev/mmcblk0p1, > missing codepage or helper program, or other error In some cases useful > info is found in syslog - try dmesg | tail or so. > > My dmesg | tail Output: > > [ 856.292964] F2FS-fs (mmcblk0p1): Magic Mismatch, valid(0xf2f52010) - > read(0x0) [ 856.292983] F2FS-fs (mmcblk0p1): Can't find valid F2FS > filesystem in 1th superblock [ 856.305466] F2FS-fs (mmcblk0p1): Magic > Mismatch, valid(0xf2f52010) - read(0x0) [ 856.305483] F2FS-fs (mmcblk0p1): > Can't find valid F2FS filesystem in 2th superblock [ 856.315244] F2FS-fs > (mmcblk0p1): Magic Mismatch, valid(0xf2f52010) - read(0x0) [ 856.315256] > F2FS-fs (mmcblk0p1): Can't find valid F2FS filesystem in 1th superblock [ > 856.324437] F2FS-fs (mmcblk0p1): Magic Mismatch, valid(0xf2f52010) - > read(0x0)Units: sectors of 1 * 512 = 512 bytes > > Even I am not able to boot from the partition. > > But if I try the dd command on mmcblk0 disk it is working perfectly. I am > able to mount and boot. > > According to my understanding, Linux OS is designed to be installed on a > disk, not on a partition and dd is to be used for the flashing on the disk. > > I want to write a Debian image on a Partition mmcblk0p1 using a command > line tool. How can I achieve this? > > I tried dd, pv, cat, cp, ddsd, Dcfldd But ended up with the same result. > You haven't specified which Debian image you're using. It likely is built with an included partition table and therefore needs to be written to the disk raw. If the partition scheme on the pre-built image doesn't suit your needs, I'd try to find a way to re-build the image according to your partitioning requirements. I haven't done this myself, but perhaps this is a good place to start: https://elinux.org/BeagleBoneBlack_Rebuild_Debian_Image_Using_image-builder -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/deab004b-b8c9-46de-b49e-87d4babfa720%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] Re: Why dns is not working in this BBB -> laptop hotspot configuration
On Monday, 18 February 2019 14:35:05 UTC+2, ith...@gmail.com wrote: > > So it is like that: even if I put nameservers 10.42.0.1 or 8.8.8.8, > whatever, into connmanctl config and asking it then shows that exact IP as > the DNS server, what is put into resolv.conf is some three values which > look like IP6 adresses, I have no idea what they mean. Anyway, DNS is not > working with them and I need to replace everything in resolv.conf with > nameserver 10.42.0.1 after every rewrite of the file by connmanctl, then it > works fine. > Well, the contents of resolv.conf come from your DHCP server (unless you've arranged it differently). To fix that, you'd need to re-configure your DHCP server. -- Kind regards Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/853ef0ad-c000-48ce-a732-62efcd85944c%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: No space on left device
Hi Ichrak, On Friday, 10 August 2018 20:42:33 UTC+3, Ichrak Mansour wrote: > > Hello, > > I am newbie, I use the version buster of my BeagleBone and when I run my > application on it, I get "no space on left device" > > How can I fix that please ? > To diagnose the state of your disk space there are two basic tools. 1. To get an overview of what's going on with your storage, "df -h" tells how big is each mounted file system and how much of it is used. 2. To dig into specific directories, "sudo du -hs /dir" will tell you how much storage space is used by files in directory /dir. You can run snippets such as this to understand what exactly is consuming space in directory /dir "sudo find /dir -maxdepth 1 -exec du -hs {} \;" Anyway, do mind where your application writes data - some filesystems such as /tmp and /var/tmp are mounted as tmpfs (i.e. ramdisks) with rather small size. To have a look at installed packages and their disk usage, I have found this command to be useful: "dpkg-query -W -f '${Installed-Size} ${Package}\n' | sort -n" -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/f5b49abd-22da-4635-b41e-b1981765cb86%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: Eth0 Stopped Working
Hi Jim, On Wednesday, 13 February 2019 21:24:52 UTC+2, Jim Fell wrote: > > Yesterday, my BeagleBone Black (Rev C) was working fine. Today, when I > powered up the system eth0 is not coming up. > > $ ip addr show eth0 > 2: eth0: mtu 1500 qdisc noop state DOWN group > default qlen 1000 > link/ether c8:df:84:d7:07:a6 brd ff:ff:ff:ff:ff:ff > > > How do I get eth0 to come up again? I've tried rebooting the board with > `$ sudo reboot` and restarting the service with `$ sudo > /etc/init.d/networking restart` but neither helped. Yesterday, I installed > smbclient and cifs-utils, and deleted the /usr/src and /usr/include > directories. Does anyone know how to get eth0 working? > There is a known problem with the Ethernet MAC not coming up on boot. This is likely to happen when you're powering the BeagleBone through the barrel connector. If this is the case, a simple reboot won't help. You need to shut down, remove power for 10 seconds and reconnect power. IIRC the physical reset button also helps. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/16a25c7f-26aa-48fc-85c1-24270c0cd7b2%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: BBB/PB/BBBW set mac address
Hi Dave, On Sunday, 10 February 2019 11:15:10 UTC+2, Dave wrote: > > Is there a way to force the MAC addresses of net devices on the > BeagleBones ? > Yes - google for "linux set mac address" and you'll have plenty of instructions. This page pops up prominently: https://en.wikibooks.org/wiki/Changing_Your_MAC_Address/Linux -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/cb4e413a-4834-452c-8c37-88e861b088f0%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: Why dns is not working in this BBB -> laptop hotspot configuration
On Monday, 11 February 2019 18:03:01 UTC+2, ith...@gmail.com wrote: > > Hello Tarmo. Thanks for your help, host google.com 8.8.8.8 gives > ;; connection timed out; no servers could be reached > > I have lost too much time an attempt to have a working network connection > on BB. > > Firstly, via USB, a known problem. With Linux, it never worked like a > normal network. > Then, an adapter Ethernet-USB. BB recognized it in maybe 80%, then 20% of > times, then never. My laptop recognizes it always until now. > Then, connmanctl. I have never seen this application before, in any case, > no way of connecting it to a given enterprise Wifi, and its error reporting > was like "invalid parameter" to its internal command with no parameters. > Sure, a more typical Wifi and it would work. > So I decided to connect BB to my laptop working as a hotspot. It accepts > specific ports, SSH for example, DNS it seems no way. > > With RPi, it just works. Cable, Wifi, in different places and networks, it > just works. BB, no way. > I'm sorry to hear about your USB-Ethernet adapter dying. I have used a couple of different USB-Ethernet adapters on my Beaglebone-derived device - some are very stable and some drop out after a few hours or days, but a reset always brings them back online. Perhaps you should get a different one. BTW, I'm a little confused - if you have a BeagleBone Black (as the topic states), why can't you use the built-in Ethernet port? Anyway, Internet connection sharing is a fragile operation as there are many things which need to go right for it to work. If your ping and ssh go through but DNS doesn't, it might be that TCP and ICMP work, but UDP doesn't. You can do some experiments using simple tools like telnet and netcat, but this doesn't really concern the BeagleBone any more. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/eb00af82-3d89-432e-a49f-b17887017532%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: Why dns is not working in this BBB -> laptop hotspot configuration
On Monday, 11 February 2019 16:28:55 UTC+2, ith...@gmail.com wrote: > > In /etc/resolv.conf there is a line nameserver 8.8.8.8 > > I am sorry, there is Debian on BB and not Ubuntu, this may explain the > missing tools. > I'd recommend debugging DNS connectivity using the low-level "host" command. It ignores the system's DNS client configuration. For example, the following command queries the name "google.com" from DNS server 8.8.8.8: $ host google.com 8.8.8.8 I would firstly try the DNS servers provided by my Internet Service Provider. Anyway, if "host" resolves names and the system's DNS client doesn't, then you have a lead. If neither option resolves names, your Internet connection is broken. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/cc18d88c-9710-46ac-a002-8621dc162fef%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: boot from eMMC while sd card inserted
Hi Markus, On Saturday, 19 January 2019 20:53:56 UTC+2, maram@googlemail.com wrote: > > Hallo, > > I would like to use the sd card as extended harddrive to keep data on it. > But, whenever I reboot the BBB (Debian 9) it will try to boot from sd card, > even if this is not marked as bootable. How to solve the problem? > > Kind regards > Markus > You haven't described how "try to boot" does not work for you. Does the SD boot succeed or does it fail, are there any error messages, etc. Do you have a bootable image on the eMMC? In my experience U-Boot is configured to attempt to boot from SD card whenever it finds anything bootable on it. This happens regardless of the state of the Boot button. I also suspect partition flags (bootable, non-bootable) are not honored. Anyway, I've re-partitioned and formatted my SD cards - after that my BB-s happily boot from eMMC. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/eda29260-4da7-405b-ac15-172645226651%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] How to use PocketBeagle CAN Pins?
Hi Gwen, On Wednesday, 16 January 2019 12:35:09 UTC+2, Gwen Stouthuysen wrote: > > Is there a method of using these commands at boot so that the CAN > interface is available? > > config-pin P1_28 can# rx > config-pin P1_26 can# tx > > I have trouble to find the necessary information on how to create a CANbus > related overlay. But in fact I only need these two commands to execute at > boot. > > Gwen > > > Op maandag 8 januari 2018 03:17:50 UTC+1 schreef RobertCNelson: >> >> On Sun, Jan 7, 2018 at 5:36 AM, wrote: >> > Hello Folks! >> > >> > We are currently experimenting with the PocketBeagle CAN Connection but >> > haven't got much success yet. The PocketBeagle will be connected to a >> car. >> > To accomplish this we use a CAN Board >> > (https://www.waveshare.com/sn65hvd230-can-board.htm) in connection >> with a >> > step down converter and a OBD2 Male Connector. >> > >> > Our first thought was that we can use the dcan1 Pins (P2.09 for RX, >> P2.11 >> > for TX) to connect the hardware and just test everything with the >> socketcan >> > tool - without success. >> > >> > We then tried different Tutorials: >> > http://www.thomas-wedemeyer.de/beaglebone-canbus-python.html >> > http://www.instructables.com/id/DIY-Beaglebone-CAN-Bus-Cape/ >> > They didn't work either. >> > >> > Another problem is that there is almost no documentation on how to use >> the >> > PocketBealge CAN or maybe we just can't find it. >> > >> > Has someone got a good tutorial or hints how to use the CAN Pins >> correct? Do >> > we need the Device Tree Overlays? How can we test that a Device Tree >> Overlay >> > is working properly or something is connected to the CAN pins? >> > We are really thankful for any advise or solution! >> >> https://github.com/beagleboard/pocketbeagle/wiki/Peripherals >> >> If you see any mistakes feel free to help improve ^ it.. >> >> Regards, >> >> -- >> Robert Nelson >> https://rcn-ee.com/ >> > Googling for "beaglebone config-pin on boot" matches a good example to get you going: https://gist.github.com/pdp7/d4770a6ba17e666848796bf5cfd0caee -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/85e83983-c4bf-44a8-b208-356551ac3a87%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: AT modem networking on PocketBeagle
Hi Gwen, On Thursday, 10 January 2019 11:47:10 UTC+2, Gwen Stouthuysen wrote: > > I would like to connect a Quectel BG96 Click board to a PocketBeagle > The hardware side is rather straight forward: install the expansion > headers and click in the board. > LEDs light up and off you go. > > The modem is not recognised by the operating system: how would he know > there is this specific device connected to the UART? > No EEPROM installed, no Cape overlay available. > > This can be solved using traditional techniques, if required as the UART I > would use is /dev/ttyS4, which is available anyhow. > > How I can I explain the operating system that this specific device is > installed and listening on /dev/ttyS4/? > > If I understand correctly the device is known by the kernel but is > expected to connect over USB. > The operating system does not concern itself with modems other than creating the virtual serial port when you connect it via USB. To do anything useful you start the relevant application and point it towards your modem. E.g. for dial-up you might use pppd, which might work with your BG96 out of the box using the distro-default chatscripts. You must configure pppd first - IIRC then file "/etc/ppp/peers/provider" is the default pppd configuration and a good starting point. File "/etc/chatscripts/gprs" is probably a good choice for a chatscript (which holds the AT commands executed by pppd). The dial-up process is started by command "sudo pon provider". pppd has a supposedly user-friendly wrapper or helper or something called wvdial - I haven't use this, so I can't advise you. For a more holistic networking solution you might also want to look at alternatives like connman/ofono, NetworkManager/ModemManager which may or may not have a working driver for your modem. I'm pretty sure ofono does not have a driver for Quectel BG96, but with a bit of luck and hacking you might be able to re-use the driver for Quectel UC15. Still, I'd suggest starting with pppd as it's a simpler, self-contained system. -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/f3758140-61ce-411b-8e7e-a422d8141dbe%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: How to connect Beaglebone blue to *any* network, except a simple Wifi
Hi! On Monday, 7 January 2019 18:17:01 UTC+2, ith...@gmail.com wrote: > > Firstly, via USB. I set a gateway on Beaglebone, but the desktop (host > Ubuntu) does not care. There are some instructions on the internet, > followed by "if does not work add use echo 1 > > /proc/sys/net/ipv4/ip_forward, a thing with iptables, whatever", it all did > not work, ping 8.8.8.8 on the bone does not work. Someone told that you > need to put the respective interface to a "trusted zone". I do not know > where I can found that zone. To add a VPN connection. "Add a VPN > connection" is grayed on my system. > > I did not want to touch wifi, because I would prefer beaglebone to be a > hotspot, but if it did not work with the cable... > > Firstly, there is "connman", to make it easier. I tried connman, but the > wifi network is enterprise tunneled double-authentication etc, connman just > prints "invalid arguments" when I try to connect. There is some > wpa-supplicant-like configuration file, added it as instructed, but no > effect, all I see is still "invalid arguments". No detailed error, no > problematic file, no problematic line, no list of these arguments, nothing, > just "invalid arguments". > > Lastly, there is wpa-supplicant itself. I made a respective file, but it > all starts and stops with > > $ wpa_supplicant -c /etc/wpa_supplicant/wpa_supplicant.conf -i wlan0 > Successfully initialized wpa_supplicant > Could not read interface p2p-dev-wlan0 flags: No such device > nl80211: Could not set interface 'p2p-dev-wlan0' UP > nl80211: deinit ifname=p2p-dev-wlan0 disabled_11b_rates=0 > p2p-dev-wlan0: Failed to initialize driver interface > P2P: Failed to enable P2P Device interface > > There is nothing about p2p-dev-wlan0, whatever it is, in > wpa_supplicant.conf > > Could you please help me? I had never problems with RPi, with BBone Black, > but Blue, no idea. I am not a network expert and I do not want to be one. I > just want to apt update etc. Thanks. > Are you trying to connect a Beaglebone to Internet using an USB connection to your PC? It's going to be a struggle, especially if you don't want to dive into networking details. I suggest buying a cheap USB-Ethernet adapter and connecting your Beaglebone to local LAN using that. It usually works just like that. Something like this: http://www.logilink.eu/Produkte_LogiLink/Aktive_Netzwerkkomponenten/Fast_Ethernet/USB_2_0_to_Fast_Ethernet_RJ45_Adapter_UA0144B.htm -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/9f32e6d9-e00f-4252-8262-1d114f3e0757%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] /sys/class/gpio/export not idempotent, unexports gpios
On Thursday, 20 December 2018 20:17:45 UTC+2, Tarmo Kuuse wrote: > On Thursday, 20 December 2018 20:04:17 UTC+2, Tarmo Kuuse wrote: >> >> On Monday, 3 December 2018 08:35:22 UTC+2, Ken Shirriff wrote: >>> >>> libgpiod looks like a nice API. I tried it out, but I'm rather baffled >>> by the behavior. If I try to set a pin with gpioset, it turns the pin off >>> and unexports it. >>> >>> $ config-pin P8_44 1 # This turns P8_44 on as expected >>> $ gpioset gpiochip2 9=1 # Should turn on gpio2[9], i.e. P8_44, but turns >>> it off. >>> $ config-pin P8_44 1# Now config-pin can't access the pin??? >>> WARNING: GPIO pin not exported, cannot set direction or value! >>> >>> So why does gpioset turn the pin off instead of on? This isn't specific >>> to P8_44; for instance "gpioset gpiochip2 7=1" will mess up P8_46 >>> (gpio2[7]). This also happens even if I don't use config-pin, so it's not >>> config-pin messing things up. >>> >> >> I'm also having problems with libgpiod - it simply does not work on a BBB >> with kernel 4.14.71-ti-r80. Cape universal is disabled, relevant pins are >> not exported via sysfs gpio interface. >> >> Example: I'm trying to set P9.23 a.k.a. gpio number 49 (gpiochip 1, pin >> 17) and absolutely nothing happens. The pin stays in some high-z state, >> measuring 0.6 volts: >> $ sudo gpioset gpiochip1 17=0 >> $ sudo gpioset gpiochip1 17=1 >> >> Running "gpioinfo gpiochip1" shows my pin as output, while it's >> definitely not that: >> ... >> line 17:"GPMC_A1" unused output active-high >> ... >> >> Also tried to control each of the user LEDs on pins 21-24 on the same >> port with very strange results. I managed to switch on usr1 or usr2 and no >> further changes are possible - the one LED won't turn off and the others >> won't turn on. Also wrote a small C program which uses the "linux/gpio.h" >> API - doesn't do anything useful to any of the pins. >> >> Am I missing something here? >> > > And the usual sysfs interface on the same works just fine. Verified with a > multimeter. > # echo 49 > /sys/class/gpio/export > # echo low > /sys/class/gpio/gpio49/direction > # echo high > /sys/class/gpio/gpio49/direction > > For the interested - the explanation to libgpiod "misbehaving". https://github.com/brgl/libgpiod/issues/32#issuecomment-444802563 This is according to design - a program can control the GPIO pin as long as it keeps the ioctl socket open. The kernel kindly resets the GPIO configuration to default value when the socket is closed. Apparently there is a daemon planned which would persist and handle requests for GPIO operations. Hmm. While libgpiod may be a good solution in many cases, I can see a few where I'd really prefer my GPIO state to persist without keeping the ioctl socket open or running a (yet to be written) daemon dedicated to it. This makes me really sad to hear of the old sysfs interface being deprecated - I'd have to revert to poking GPIO registers directly. -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/a2baa192-3999-4c5e-b5af-6a1ae1f3f8fa%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] /sys/class/gpio/export not idempotent, unexports gpios
On Thursday, 20 December 2018 20:04:17 UTC+2, Tarmo Kuuse wrote: > > On Monday, 3 December 2018 08:35:22 UTC+2, Ken Shirriff wrote: >> >> libgpiod looks like a nice API. I tried it out, but I'm rather baffled by >> the behavior. If I try to set a pin with gpioset, it turns the pin off and >> unexports it. >> >> $ config-pin P8_44 1 # This turns P8_44 on as expected >> $ gpioset gpiochip2 9=1 # Should turn on gpio2[9], i.e. P8_44, but turns >> it off. >> $ config-pin P8_44 1# Now config-pin can't access the pin??? >> WARNING: GPIO pin not exported, cannot set direction or value! >> >> So why does gpioset turn the pin off instead of on? This isn't specific >> to P8_44; for instance "gpioset gpiochip2 7=1" will mess up P8_46 >> (gpio2[7]). This also happens even if I don't use config-pin, so it's not >> config-pin messing things up. >> > > I'm also having problems with libgpiod - it simply does not work on a BBB > with kernel 4.14.71-ti-r80. Cape universal is disabled, relevant pins are > not exported via sysfs gpio interface. > > Example: I'm trying to set P9.23 a.k.a. gpio number 49 (gpiochip 1, pin > 17) and absolutely nothing happens. The pin stays in some high-z state, > measuring 0.6 volts: > $ sudo gpioset gpiochip1 17=0 > $ sudo gpioset gpiochip1 17=1 > > Running "gpioinfo gpiochip1" shows my pin as output, while it's definitely > not that: > ... > line 17:"GPMC_A1" unused output active-high > ... > > Also tried to control each of the user LEDs on pins 21-24 on the same port > with very strange results. I managed to switch on usr1 or usr2 and no > further changes are possible - the one LED won't turn off and the others > won't turn on. Also wrote a small C program which uses the "linux/gpio.h" > API - doesn't do anything useful to any of the pins. > > Am I missing something here? > And the usual sysfs interface on the same works just fine. Verified with a multimeter. # echo 49 > /sys/class/gpio/export # echo low > /sys/class/gpio/gpio49/direction # echo high > /sys/class/gpio/gpio49/direction -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/81ef87e8-ea04-4d4c-87be-bdbefc2a273e%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] /sys/class/gpio/export not idempotent, unexports gpios
On Monday, 3 December 2018 08:35:22 UTC+2, Ken Shirriff wrote: > > libgpiod looks like a nice API. I tried it out, but I'm rather baffled by > the behavior. If I try to set a pin with gpioset, it turns the pin off and > unexports it. > > $ config-pin P8_44 1 # This turns P8_44 on as expected > $ gpioset gpiochip2 9=1 # Should turn on gpio2[9], i.e. P8_44, but turns > it off. > $ config-pin P8_44 1# Now config-pin can't access the pin??? > WARNING: GPIO pin not exported, cannot set direction or value! > > So why does gpioset turn the pin off instead of on? This isn't specific to > P8_44; for instance "gpioset gpiochip2 7=1" will mess up P8_46 (gpio2[7]). > This also happens even if I don't use config-pin, so it's not config-pin > messing things up. > I'm also having problems with libgpiod - it simply does not work on a BBB with kernel 4.14.71-ti-r80. Cape universal is disabled, relevant pins are not exported via sysfs gpio interface. Example: I'm trying to set P9.23 a.k.a. gpio number 49 (gpiochip 1, pin 17) and absolutely nothing happens. The pin stays in some high-z state, measuring 0.6 volts: $ sudo gpioset gpiochip1 17=0 $ sudo gpioset gpiochip1 17=1 Running "gpioinfo gpiochip1" shows my pin as output, while it's definitely not that: ... line 17:"GPMC_A1" unused output active-high ... Also tried to control each of the user LEDs on pins 21-24 on the same port with very strange results. I managed to switch on usr1 or usr2 and no further changes are possible - the one LED won't turn off and the others won't turn on. Also wrote a small C program which uses the "linux/gpio.h" API - doesn't do anything useful to any of the pins. Am I missing something here? -- Kind regars, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/7baee28e-eb9b-4282-ba23-fc98bdf2f8c3%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: Unable to locate package in beaglebone black Rev C with Debian 8
On Thursday, 20 December 2018 06:08:03 UTC+2, aleka...@gmail.com wrote: > > I am a newbie for beaglebone black. I got my beaglebone black few days > before and trying to install matplotlib. I am getting E:Unable to locate > package python-matplotlib. > > I got my beaglebone black Rev C with debian 8 jessie and python 2.7 > installed > > I tried following commands to install matplotlib: > > sudo apt-get install python-matplotlib > pip install python-matplotlib > > then I tried update and upgrade and again tried above commands but no > success. > > Any help would be very useful. As I am stuck. > According to the Debian package documentation jessie should contain "python-matplotlib" with version 1.4.2: https://packages.debian.org/search?suite=default§ion=all&arch=any&searchon=names&keywords=python-matplotlib The commands to install the Debian package (omit the dollar sign): $ sudo apt-get update $ sudo apt-get install python-matplotlib You may also look for a matching package in the repository: $ apt-cache search matplotlib -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/eeaeb6ff-9938-4cf3-b50f-8708001e75f0%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: Problems with SSH over the USB
Hi Andy, On Sunday, 16 December 2018 23:36:06 UTC+2, Andy Bushnell wrote: > > I used all I could find on line. > > 192.168.7.2 > 192.168.8.2 > I suspect you're missing something here. I recall the default addresses for Ethernet emulation over USB being 192.168.6.2 and 192.168.7.2. One works if connecting the Beaglebone to a Windows computer, other with a Mac and both are usable on Linux. The 192.168.8.2 might be used by the WiFi access point mode. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/b68bfdc8-728a-45b5-85e7-8a32a1082f50%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: Using PTP with Beaglebone black
On Friday, 7 December 2018 02:06:46 UTC+2, BBB _user wrote: > > I am a student trying to implement PTP using BBB. I have studied that > there are BBB images that already have HW timestamps enabled. > Is there any easy way that I can follow to implement PTP without having to > compile a Linux kernel by myself? > Any help in this regard is much appreciated. > Without knowing anything about PTP, Debian has a package ptpd - perhaps that is what you're looking for. -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/10b6b5f7-b051-4a5c-90ae-a9a27548ac89%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] Re: Debian 9: autorun on boot with systemd does not work
Hi Harke, > No problem, I can do that, in fact I have done that many times with the >>> same results. I can enter: cd /home/debian/eme/ and then: python aceme.py >>> or I can enter: python /home/debian/eme/aceme.py. That works in both ways >>> exactly the same. When run under the bash it gives an error saying that >>> Tkinter does not work because of a $Display error. When run under QTerminal >>> in LXQT it works fine. This is the reason that I must be sure in the >>> service file that the graphics modules are loaded first. In fact this is >>> the (only) way to start the application. >>> Conclusion: the command line: python /home/debian/eme/aceme.py works >>> perfectly. However, this is not what I need to put after: ExecStart (as far >>> as I understand). Now I have: ExecStart=/home/debian/eme/aceme.py. And >>> I made aceme.py executable. >>> I do not know bash scripts so if necessary please help me here. >>> I hope this helps in the analysis. >>> >> Do I understand correctly that your application, which you're trying to run on boot, displays a GUI? I strongly suspect systemd cannot launch GUI programs. It's meant for daemons which work in the background and don't display anything. Not having done this personally on a Beagle, I'd try to go for something like this: 1. Ensure Beagle starts a graphical environment (LXQT or which ever you prefer) 2. LXQT automatically logs in with your user and brings up the desktop 3. LXQT autostart mechanism triggers your application. Somebody has written instructions for step 3 here: https://wiki.archlinux.org/index.php/LXQt#Autostart -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/1b9386b7-9f25-4ed7-b5ac-373a1287bdd6%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Any upcoming Debian 10 buster rework? How about console images?
Hi! For context - I'm preparing a batch of some BBB compatible devices and I'd love to install the same Debian version on them over the next few years. Buster seems like a forward-looking choice, as the state of Debian 10 itself seems good enough to me. Is there a development plan for Debian 10 (buster) on beaglebone? Specifically, is there any significant rework planned (apart from the upstream package updates) to kernel 4.14, uboot or devicetrees which would shake compatibility with the current state of buster? If not, I'd merrily plop down buster on my devices starting now. Second question - I'm looking for a buster "console" image without the various iot, cloud9, nodejs packages and couldn't find any in Robert's repository. Is that on purpose? I have been apt-get purging the "iot" image, but I'd feel warm and fuzzy with an image where they are not present in the first place. -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/2a9fe12f-a3ed-4fe9-a3eb-03e89a097a02%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: Unexporting Device Tree Overlays
Hi Garrett, On Thursday, 23 August 2018 01:17:40 UTC+3, Garrett Higgins wrote: > > hi, I am trying to activate all PWM pins so that I have a total of 8 to > use. When I check my slots file, I have a device tree overlay that I am > trying to export, but permission is denied. Can anyone help me? see below > > > debian@beaglebone:/lib/firmware$ cat $SLOTS > 0: PF -1 > 1: PF -1 > 2: PF -1 > 3: PF -1 > 4: P-O-L- 0 Override Board Name,00A0,Override Manuf,cape-universaln > debian@beaglebone:/lib/firmware$ echo -4 > /slots > -bash: /slots: Permission denied > debian@beaglebone:/lib/firmware$ > You'll want to do any reconfiguration of your system under permissions of the root user. Usually that involves prepending a command with "sudo", but it doesn't work with commands that write stuff to files using output forwarding (e.g. "$ echo -4 > /slots"). In this case I'd suggest switching to root shell with the following command. You'll likely be asked for the password of user "debian": $ sudo su - After this all commands will execute with root permissions. As for the specific problem you're trying to solve, you're currently on universal capes: https://github.com/cdsteinkuehler/beaglebone-universal-io Would it perhaps be easier to exploit those and enable your PWM-s with the config-pin script? Unfortunately, I don't know if enabling all of your desired PWM-s are supported there. Remember, you need root permissions. # config-pin P8.07 -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/ae7eabf0-0fcf-45ec-880e-7084ce0af502%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] Kernel 4.9.58-ti-r72 disappeared from stretch package list
On 2018-04-10 18:20, Robert Nelson wrote: On Tue, Apr 10, 2018 at 9:54 AM, Tarmo Kuuse wrote: Hi! I have been naïvely relying on Stretch kernel package "linux-image-4.9.58-ti-r72" and some custom devicetree modifications on Beaglebone Black for the past few months. Yesterday I noticed that the apt package lists no longer serve this specific version of the kernel. Is this drop expected? Can this kernel be restored? If not, are there any "LTS" kernel packages and how would I recognise those? Sorry, added back.. and tagged as special to not get auto-removed. ;) Wow, that was quick :) Thanks a bunch! -- Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/6f53f313-ef69-1da3-814e-a2b33cba054b%40gmail.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Kernel 4.9.58-ti-r72 disappeared from stretch package list
Hi! I have been naïvely relying on Stretch kernel package "linux-image-4.9.58-ti-r72" and some custom devicetree modifications on Beaglebone Black for the past few months. Yesterday I noticed that the apt package lists no longer serve this specific version of the kernel. Is this drop expected? Can this kernel be restored? If not, are there any "LTS" kernel packages and how would I recognise those? $ sudo /opt/scripts/tools/version.sh git:/opt/scripts/:[a5d5f437baeaa7f8e14d5ae037268abf7a67e88e] eeprom:[A335BNLTBBG1BBG217042451] model:[TI_AM335x_BeagleBone_Black] dogtag:[BeagleBoard.org Debian Image 2017-11-05] bootloader:[eMMC-(default)]:[/dev/mmcblk1]:[U-Boot 2017.09-2-g0f3f1c7907] kernel:[4.9.59-ti-r74] uboot_overlay_options:[enable_uboot_overlays=1] uboot_overlay_options:[uboot_overlay_pru=/lib/firmware/AM335X-PRU-UIO-00A0.dtbo] uboot_overlay_options:[enable_uboot_cape_universal=1] pkg:[bb-cape-overlays]:[4.4.20180408.0-0rcnee0~stretch+20180409] pkg:[bb-wl18xx-firmware]:[1.20180328-0rcnee2~stretch+20180328] pkg:[firmware-ti-connectivity]:[20170823-1rcnee1~stretch+20180328] -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/6fea0673-d889-4549-9607-5fdd7ac3e17b%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: SD card errors on hard reboot
On Monday, 19 March 2018 23:55:04 UTC+2, Troy Weber wrote: > > Hello! > > I have an application running on the Beaglebone Black. It runs from the > eMMC because we need to mount a blank SD card and occasionally store data > or video on the external SD card. Here are some details to jog ideas: > >- using a Bealgebone black >- connected to a powered hub and running 2 led strips and 1 camera >- made a custom cape/shield that allows us to interface with > - an RGB button > - a 16x2 char lcd > - an optically isolated relay module >- the relay module runs a solenoid circuit (in which I'm currently >debugging the need for a flyback diode) >- the SD card is mounted using */etc/fstab* with the line */dev/mmcblk0p1 >/media/sd auto auto,nofail,user,dmask=022,fmask=111 0 0* >- SD cards are inexpensive and found here: > > https://www.amazon.com/gp/product/B00WUCNGVQ/ref=oh_aui_search_detailpage?ie=UTF8&psc=1 > > My application is writing files to the SD card only on certain occasions > during which the Beaglebone will not be hard-rebooted. However, it will be > hard-rebooted (and the application started by systemd on next boot) as a > means of shutting it down. > > The problem is that I've recently began to discover issues with the blank > SD card on next boot after a hard reboot. For a short time I can *touch* > and write *vim* files on the card just fine. Once I start up the node.js > program, part way through it complains about the SD card being read-only. > From then on I cannot *touch* or *vim* files anymore. So I reformat the > card and copy the configuration files back on. I've noticed that one of my > configuration files is chopped short. > > Here are some things I've tried and intend to try: > >- have tried > - removing the power to the hub which seemed to be back-powering > the BBB (still fails on reboot) > - stopping the program that writes files entirely (still fails on > reboot) >- intend to try > - writing the configuration files elsewhere (on the eMMC for > example) > - better SD cards > > In the meantime, any ideas? > What's the file system on the SD card? FAT? ext4? -- Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/ea6a9f55-220d-40c0-8781-5dfffee87178%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] Re: PocketBeagal Debian internet over macOS 10.13.2 USB host? (internet over USB)
On Tuesday, 19 December 2017 17:35:07 UTC+2, Tarmo Kuuse wrote: > > On Tuesday, 19 December 2017 17:05:11 UTC+2, Jon Pellant wrote: >> >> What you say sounds good. If that is the case, there should be a way to >> tell the BB to accept the DHCP config. On Ångstrom, that would be with >> udhcpc. I am trying to figure out how to do that with Debian. I am >> currently exploring ‘connman’ but I just have not found the time. If that >> works, I will definitely post the solution here. >> > > Assuming you're using Robert's Debian images - the actual Ethernet > interface requires zero effort to accept the DHCP config. It's configured > as a DHCP client out of box. Just connect it to anything - a router, your > Mac with Internet Sharing enabled, ... - and it'll be online immediately. > > Only the network emulation on USB requires hacking boot scripts > (/opt/scripts/boot/autoconfigure_usb(0|1).sh, around the last 20 lines) to > disable hardcoded network configuration and switch from DHCP server to > client. Frankly, I wouldn't bother. > Oh, right. PocketBeagle. Sorry, I missed that. Well, then happy hunting! -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/a97583a1-0a52-436c-8033-3c1496cba7cf%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] Re: PocketBeagal Debian internet over macOS 10.13.2 USB host? (internet over USB)
On Tuesday, 19 December 2017 17:05:11 UTC+2, Jon Pellant wrote: > > What you say sounds good. If that is the case, there should be a way to > tell the BB to accept the DHCP config. On Ångstrom, that would be with > udhcpc. I am trying to figure out how to do that with Debian. I am > currently exploring ‘connman’ but I just have not found the time. If that > works, I will definitely post the solution here. > Assuming you're using Robert's Debian images - the actual Ethernet interface requires zero effort to accept the DHCP config. It's configured as a DHCP client out of box. Just connect it to anything - a router, your Mac with Internet Sharing enabled, ... - and it'll be online immediately. Only the network emulation on USB requires hacking boot scripts (/opt/scripts/boot/autoconfigure_usb(0|1).sh, around the last 20 lines) to disable hardcoded network configuration and switch from DHCP server to client. Frankly, I wouldn't bother. -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/cfd98d21-57c3-45dc-a011-fb247d983c48%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: [beaglebone] Unable to use WP8548 with ConnMan
On Monday, 18 December 2017 18:15:08 UTC+2, RobertCNelson wrote: > > On Mon, Dec 18, 2017 at 9:52 AM, Tracy Smith > wrote: > > ok thanks Robert. Did you configure the kernel and wpa_supplicant for > > WiFi independent of cellular? > > For that project with cellular, i used ppp for the cellular uplink. > > connman was used to create a wifi access point. > > Such that devices could connect to it and get network access. > > Regards, > > -- > Robert Nelson > https://rcn-ee.com/ > Lately I've been pondering on how to make both 3G and Ethernet work via connman. pppd and connman tend to bicker and overwrite each other's resolv.conf entries, which is a bit of a pain if one grabs the default route and the other installs its local DNS servers :) Anyway, as far as I can tell connman relies on a component called ofono to manage modems and do dialup. This, in turn, requires support for your 3G modem in the form of some driver. Unfortunately I can find zero practical documentation on ofono and - judging by updates in recent years - it seems to be abandonware. Hmm. -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/853374f4-2d9c-4816-bc7d-f5e47bc385de%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] Re: PocketBeagal Debian internet over macOS 10.13.2 USB host? (internet over USB)
On Tuesday, 19 December 2017 16:18:39 UTC+2, Jon Pellant wrote: > > @Yoda: System Preferences >> Sharing >> Internet Sharing and share with > the Beaglebone.local is what I am doing. This has worked for me in past > releases; however, not working for me in this configuration. > > > On Dec 18, 2017, at 20:49, Yoda > wrote: > > I have been seeing the same thing with MacOS 10.13.2. I am able to ssh > into the pocket beagle over the USB, but when I try to turn on Internet > Sharing to be able to forward to the internet then the ssh connection hangs > and I can no longer work on the ssh connection I had established. I > vaguely remember something about Connection sharing on Mac but I can seem > to find it. > > Hi Jon, I'm not an expert in anything Apple, however I strongly suspect that enabling Internet Sharing means your Mac will do the following: 1. create a private subnetwork on the selected interface with a semi-random address space (e.g. 192.168.XXX.0/24) 2. start a DHCP service to serve addresses from selected address space to client hosts 3. set the Mac up as a router doing NAT for the client hosts Unfortunately the BeagleBone's USB network emulation has already done steps 1 and 2 on _its_ side with pre-determined address spaces 192.168.6.0/24 and 192.168.7.0/24. This creates an address conflict between the BB and Mac which makes connectivity impossible. I'd suggest connecting your BB and Mac via Ethernet (buy a cheap USB-Ethernet adapter if needed) and enabling Internet Sharing on that interface. In this case the sharing feature should work as intended. Plus you'll always have the same USB network emulation to ssh into the BB in a tight spot. -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/30549d61-697e-468c-ba12-a440a6602e4d%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: fsck fails to start on writable partition at boot/startup
On Wednesday, December 6, 2017 at 10:40:11 PM UTC+2, Dave Barndt wrote: > > Does anything there look strange to you? > Can't say it does. Your /var/ file system is writable and gets corrupted - this is expected. fsck failing to run is unexpected. Looks like systemd has taken over the responsibility of boot-time fsck-ing. A look at the service which fails (/lib/systemd/system/systemd-fsck@.service) tells me that there is some scant documentation available in "man systemd-fsck". However, it looks like you might have to dig into systemd source to analyze how exactly boot-time fsck-s are triggered and what could go wrong there. Alternatively, perhaps you'd want to have a quick glance at changelog for systemd, to see if they've fixed any bugs in this area since your installed version. My BBB with Debian 8 is running systemd version "230-7~bpo8+2" (dpkg -l systemd) and they've made 5 releases since. -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/09bb67af-fd7e-461d-abaa-242b8f225fb2%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: fsck fails to start on writable partition at boot/startup
ound > [2.514227] bone_capemgr bone_capemgr: slot #3: No cape found > Loading, please wait... > rootfs: clean, 78785/98304 files, 384993/393216 blocks > [ 12.471231] pinctrl-single 44e10800.pinmux: pin 44e108c0.0 already > requested by 481aa000.serial; cannot claim for 0-0039 > [ 12.482304] pinctrl-single 44e10800.pinmux: pin-48 (0-0039) status -22 > [ 12.488911] pinctrl-single 44e10800.pinmux: could not request pin 48 > (44e108c0.0) from group adi_hdmi_bbbi_pins on device pinctrl-single > [ 12.501341] adv7511 0-0039: Error applying setting, reverse things back > [ 12.928105] pinctrl-single 44e10800.pinmux: pin 44e108c0.0 already > requested by 481aa000.serial; cannot claim for 0-0039 > [ 12.939198] pinctrl-single 44e10800.pinmux: pin-48 (0-0039) status -22 > [ 12.945814] pinctrl-single 44e10800.pinmux: could not request pin 48 > (44e108c0.0) from group adi_hdmi_bbbi_pins on device pinctrl-single > [ 12.958243] adihdmi 0-0039: Error applying setting, reverse things back > [ 14.022183] EDID block is all zeroes > [ 14.038732] adihdmi_encoder_get_modes - 788 - No EDID > [ 14.279735] adihdmi_encoder_get_modes - 788 - No EDID > [FAILED] Failed to start File System Check on /dev/mmcblk1p2. > See 'systemctl status systemd-fsck@dev-mmcblk1p2.service' for details. > [DEPEND] Dependency failed for /var. > [DEPEND] Dependency failed for Load/Save Random Seed. > [DEPEND] Dependency failed for Update UTMP about System Runlevel Changes. > [DEPEND] Dependency failed for /root. > [DEPEND] Dependency failed for Local File Systems. > [DEPEND] Dependency failed for netfilter persistent configuration. > [DEPEND] Dependency failed for Emergency Shell. > [DEPEND] Dependency failed for Emergency Mode. > [DEPEND] Dependency failed for Update UTMP about System Boot/Shutdown. > [DEPEND] Dependency failed for Flush Journal to Persistent Storage. > [DEPEND] Dependency failed for /srv. > [DEPEND] Dependency failed for /media. > [DEPEND] Dependency failed for /home. > [DEPEND] Dependency failed for /var/local/swapfile. > [DEPEND] Dependency failed for Swap. > Starting Raise network interfaces... > Starting File System Check on /dev/mmcblk1p2... > [ OK ] Reached target Timers. > [ OK ] Closed Syslog Socket. > [ OK ] Reached target Login Prompts. > Starting Create Volatile Files and Directories... > [ OK ] Reached target Sockets. > [FAILED] Failed to start Create Volatile Files and Directories. > See 'systemctl status systemd-tmpfiles-setup.service' for details. > [FAILED] Failed to start File System Check on /dev/mmcblk1p2. > See 'systemctl status systemd-fsck@dev-mmcblk1p2.service' for details. > [DEPEND] Dependency failed for /var. > [DEPEND] Dependency failed for /root. > [ OK ] Started Raise network interfaces. > [ OK ] Reached target Network. > [ OK ] Reached target Network is Online. > > (nothing appears after this; unable to break to a console prompt, etc.) > Hi Dave, Although I have seen a few corrupt file systems, I haven't stumbled on a case where fsck fails to run. With vanilla config fsck usually is started on boot and, in case of errors, patiently waits for you to manually confirm each fix by pressing "y". This can be worked around by adding to the kernel command line "fsck.mode=force fsck.repair=yes". What's in your /etc/fstab and /boot/uEnv.txt? -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/9491271a-33c0-4088-8b6d-feef12984d1d%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: MySQL on Beaglebone Black
On Thursday, 16 November 2017 16:27:47 UTC+2, Paul Van den Bergh wrote: > > My prevous post was shortened while posting > > 4. I see no firewall roules: > > root@beaglebone:~# iptables -L -n -v > > Chain INPUT (policy ACCEPT 2959 packets, 304K bytes) > > pkts bytes target prot opt in out source > destination > > > Chain FORWARD (policy ACCEPT 0 packets, 0 bytes) > > pkts bytes target prot opt in out source > destination > > > Chain OUTPUT (policy ACCEPT 478 packets, 42627 bytes) > > pkts bytes target prot opt in out source > destination > > I'll examine the logs in the hope to find the problem. Thanks for your > valuable input ! > > Regards, > Paul > > > Op donderdag 16 november 2017 15:22:57 UTC+1 schreef Paul Van den Bergh: >> >> Hi Tarmo, >> >> 1. since I can ssh into the BBB, it's the exact IP address: >> >> paul@debian-9-2-1-64:~$ ssh root@192.168.1.52 >> Last login: Tue Oct 10 14:07:21 2017 from 192.168.1.13 >> root@beaglebone:~# ifconfig >> eth0: flags=-28605 mtu 1500 >> inet *192.168.1.52* netmask 255.255.255.0 broadcast 192.168. >> 1.255 >> inet6 fe80::d239:72ff:fe17:eaae prefixlen 64 scopeid 0x20 >> ether d0:39:72:17:ea:ae txqueuelen 1000 (Ethernet) >> RX packets 7335 bytes 663521 (647.9 KiB) >> RX errors 0 dropped 1751 overruns 0 frame 0 >> TX packets 582 bytes 54176 (52.9 KiB) >> TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0 >> device interrupt 174 >> >> >> lo: flags=73 mtu 65536 >> >> inet 127.0.0.1 netmask 255.0.0.0 >> >> inet6 ::1 prefixlen 128 scopeid 0x10 >> >> loop txqueuelen 1 (Local Loopback) >> >> RX packets 160 bytes 12960 (12.6 KiB) >> >> RX errors 0 dropped 0 overruns 0 frame 0 >> >> TX packets 160 bytes 12960 (12.6 KiB) >> >> TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0 >> >> >> >> usb0: flags=4099 mtu 1500 >> >> inet 192.168.7.2 netmask 255.255.255.252 broadcast 192.168.7.3 >> >> ether d0:39:72:17:ea:b0 txqueuelen 1000 (Ethernet) >> >> RX packets 0 bytes 0 (0.0 B) >> >> RX errors 0 dropped 0 overruns 0 frame 0 >> >> TX packets 0 bytes 0 (0.0 B) >> >> TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0 >> >> >> >> usb1: flags=4099 mtu 1500 >> >> inet 192.168.6.2 netmask 255.255.255.252 broadcast 192.168.6.3 >> >> ether d0:39:72:17:ea:b3 txqueuelen 1000 (Ethernet) >> >> RX packets 0 bytes 0 (0.0 B) >> >> RX errors 0 dropped 0 overruns 0 frame 0 >> >> TX packets 0 bytes 0 (0.0 B) >> >> TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0 >> >> >> >> >> 2. mysql is running: (another problem is setting up the time server on >> the BBB
[beagleboard] Re: MySQL on Beaglebone Black
On Thursday, 16 November 2017 11:25:13 UTC+2, Paul Van den Bergh wrote: > > Same with telnet (from Debian in VirtualBox) : > > paul@debian-9-2-1-64:~$ telnet 192.168.1.52 3306 > Trying 192.168.1.52... > telnet: Unable to connect to remote host: Connection refused > > > > Then you can proceed to further debugging your setup. Some suggestions: 1. Verify that the IP you're connecting to is the IP of the Beaglebone ("ifconfig" on BB) 2. Verify that mysql service is running on the Beaglebone ("systemctl status mysql" on BB) 3. Verify that mysql is accepting connections on port 3306 from external hosts (no idea, refer to the docs) 4. Verify that there aren't any firewall rules to block traffic ("sudo iptables -L -n -v" on BB) 5. Go through the mysql logs for any failures (/var/log/mysql.*) -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/9626234a-b650-4f97-9644-2ca7b2266ad3%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: MySQL on Beaglebone Black
On Wednesday, November 15, 2017 at 12:38:49 PM UTC+2, Paul Van den Bergh wrote: > > > Hi all, > > I have installed MySQL-server and MySQL-client on a BBB : > > root@beaglebone:~# mysql --version > mysql Ver 15.1 Distrib 10.1.26-MariaDB, for debian-linux-gnueabihf ( > armv7l) using readline 5.2 > > I like to connect to it from a Windows machine using MySQL Workbench, but > I'm not able to setup a connection : > > > > root@beaglebone:~# cat /etc/mysql/my.cnf > # The MariaDB configuration file > # The MariaDB/MySQL tools read configuration files in the following > order:# 1. "/etc/mysql/mariadb.cnf" (this file) to set global defaults, > # 2. "/etc/mysql/conf.d/*.cnf" to set global options. > # 3. "/etc/mysql/mariadb.conf.d/*.cnf" to set MariaDB-only options. > # 4. "~/.my.cnf" to set user-specific options. > # > # If the same option is defined multiple times, the last one will apply. > # > # One can use all long options that the program supports. > # Run program with --help to get a list of available options and with > # --print-defaults to see which it would actually understand and use. > > # > # This group is read both both by the client and the server > # use it for options that affect everything > # > [client-server] > > # Import all .cnf files from configuration directory > !includedir /etc/mysql/conf.d/ > !includedir /etc/mysql/mariadb.conf.d/ > root@beaglebone:~# mysql > Welcome to the MariaDB monitor. Commands end with ; or \g. > Your MariaDB connection id is 3 > Server version: 10.1.26-MariaDB-0+deb9u1 Debian 9.1 > > Copyright (c) 2000, 2017, Oracle, MariaDB Corporation Ab and others. > > Type 'help;' or '\h' for help. Type '\c' to clear the current input > statement. > > MariaDB [(none)]> SELECT User, Host FROM mysql.user; > +--+-+ > | User | Host| > +--+-+ > | root | % | > | root | 192.168.1.% | > +--+-+ > 2 rows in set (0.01 sec) > > MariaDB [(none)]> > > Any help is very welcome. > > Thanks in advance, > > Paul. > My mistake, the command would be: telnet 192.168.1.52 3306 -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/3f7eae9e-fa2b-4d73-97e5-52806b5ed984%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: How to protect my BeagleBone Black from unpredictable 5V electric power interruption?
On Wednesday, November 15, 2017 at 2:19:09 PM UTC+2, c...@mail.ru wrote: > > I've heard about it if occur power interruption then my BBB is damage at > once. Is it really necessary only UPS or supercapacitor for this? > I personally haven't witnessed any damage to hardware. The ext4 file system, however, can become corrupt if the power loss occurs at an unfortunate moment. I've seen that happen 6-7 times on this year on some 10-s of devices which get their power disconnected every day. So far I've implemented following software workarounds: 1. Add nodelalloc as mount option. This should reduce probability of FS corruption according to http://www.pointsoftware.ch/en/4-ext4-vs-ext3-filesystem-and-why-delayed-allocation-is-bad/ 2. Add "fsck.mode=force fsck.repair=yes" to the kernel boot arguments. This will enforce FS checks on every boot and fix any problems found. I have no evidence on the effectiveness of workaround nr 1, but nr 2 seems to alleviate the problem significantly. I've had no dead devices since. -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/abebe697-a913-4652-b1f6-690b3bfaaae0%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: MySQL on Beaglebone Black
On Wednesday, November 15, 2017 at 12:38:49 PM UTC+2, Paul Van den Bergh wrote: > > > Hi all, > > I have installed MySQL-server and MySQL-client on a BBB : > > root@beaglebone:~# mysql --version > mysql Ver 15.1 Distrib 10.1.26-MariaDB, for debian-linux-gnueabihf ( > armv7l) using readline 5.2 > > I like to connect to it from a Windows machine using MySQL Workbench, but > I'm not able to setup a connection : > > > > root@beaglebone:~# cat /etc/mysql/my.cnf > # The MariaDB configuration file > # The MariaDB/MySQL tools read configuration files in the following > order:# 1. "/etc/mysql/mariadb.cnf" (this file) to set global defaults, > # 2. "/etc/mysql/conf.d/*.cnf" to set global options. > # 3. "/etc/mysql/mariadb.conf.d/*.cnf" to set MariaDB-only options. > # 4. "~/.my.cnf" to set user-specific options. > # > # If the same option is defined multiple times, the last one will apply. > # > # One can use all long options that the program supports. > # Run program with --help to get a list of available options and with > # --print-defaults to see which it would actually understand and use. > > # > # This group is read both both by the client and the server > # use it for options that affect everything > # > [client-server] > > # Import all .cnf files from configuration directory > !includedir /etc/mysql/conf.d/ > !includedir /etc/mysql/mariadb.conf.d/ > root@beaglebone:~# mysql > Welcome to the MariaDB monitor. Commands end with ; or \g. > Your MariaDB connection id is 3 > Server version: 10.1.26-MariaDB-0+deb9u1 Debian 9.1 > > Copyright (c) 2000, 2017, Oracle, MariaDB Corporation Ab and others. > > Type 'help;' or '\h' for help. Type '\c' to clear the current input > statement. > > MariaDB [(none)]> SELECT User, Host FROM mysql.user; > +--+-+ > | User | Host| > +--+-+ > | root | % | > | root | 192.168.1.% | > +--+-+ > 2 rows in set (0.01 sec) > > MariaDB [(none)]> > > Any help is very welcome. > > Thanks in advance, > > Paul. > The simplest debugging you could do is to attempt to create a TCP connection to the intended host and port. If this is accepted, then you have the TCP part working and only need to check MySQL configuration. tcp 192.168.1.52 3306 -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/12550297-b9e6-4c74-b909-765cb30d%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] Why two Ethernet over USB interfaces, usb0 and usb1?
On Sunday, 24 September 2017 21:19:02 UTC+3, qie...@gmail.com wrote: > > Okay, this answers my question in another thread. So now how do I > disable one or the other? > > On Thursday, August 31, 2017 at 4:11:49 PM UTC+2, RobertCNelson wrote: >> >> On Thu, Aug 31, 2017 at 9:07 AM, Tarmo Kuuse wrote: >> > >> > Hi! >> > >> > I've been reading the am335x_evm.sh & friends for a while and don't get >> it. >> > Why are there two separate Ethernet over USB interfaces - eth0 and eth1 >> - >> > doing the same exact thing? They both show up when I connect a BB to my >> > Debian PC. >> > >> > Having no experience with gadget and libcomposite, I'd guess that RNDIS >> and >> > CDC ECM are both required simultaneously to somehow support >> > Windows/Mac/Linux users out of the box? >> >> So this is Windows and Apple's fault.. >> >> Windows = CDC ECM = needs signed driver.. >> Apple = RNDIS = needs signed driver... >> >> Windows will use the RNDIS with a built-in driver, thus user doesn't >> have to install anything... >> Apple will use the CDC ECM with a built-in driver, thus user doesn't >> have to install anything... >> >> So in linux, we end with two interfaces, because Windows and Apple >> don't have a unified built-in driver... >> >> Regards, >> >> -- >> Robert Nelson >> https://rcn-ee.com/ >> > I'd go for hacking out either the rndis or ecm config from usb-gadget setup in am335x_evm.sh (see Robert's last post with highlighted lines). -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/42bfb1f7-f8c5-495e-86dd-a8c0839d33cc%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] Re: how to get read-only files system these days?
On Tuesday, 19 September 2017 20:03:46 UTC+3, Britton Kerin wrote: > > On Mon, Sep 18, 2017 at 2:21 AM, Tarmo Kuuse > wrote: > > On Sunday, 17 September 2017 10:45:16 UTC+3, Britton Kerin wrote: > >> > >> On Wed, Sep 6, 2017 at 7:59 AM, Tarmo Kuuse > wrote: > >> > On Tuesday, September 5, 2017 at 1:45:05 AM UTC+3, Britton Kerin > wrote: > >> >> > >> >> Any help modifying the above recipe or pointer to how to do it these > >> >> days would be appreciated. > >> > > >> > > >> > I just declare the root file system as read-only in fstab and create > >> > some > >> > ramdisks for stuff that needs to be writable. > >> > > >> > LABEL=rootfs / ext4 ro,noatime,errors=remount-ro 0 1 > >> > debugfs /sys/kernel/debug debugfs defaults 0 0 > >> > tmpfs /var/log tmpfs defaults,noatime,nosuid,mode=0777,size=128M > 0 > >> > 0 > >> > tmpfs /var/tmp tmpfs defaults,noatime,nosuid,size=16M 0 0 > >> > tmpfs /tmp tmpfs defaults,noatime,nosuid,size=128M 0 0 > >> > > >> > Service generic-board-startup (a.k.a. > >> > /opt/scripts/boot/generic-startup.sh) > >> > must run once with a writable root to set up a bunch of stuff (on > later > >> > boots it complains, but it doesn't seem to matter). After that you're > >> > stuck > >> > hacking all the software which expects to be able to write to disk. > For > >> > a > >> > working network, "/etc/resolv.conf" needs symlinking to a writable > >> > location. > >> > Many services expect to persistently store their stuff under > "/var/lib". > >> > > >> > I've used tmpfiles.d to create required symlinks, e.g. > >> > > >> > $ cat /etc/tmpfiles.d/10-ro-symlinks.conf > >> > L+ /etc/resolv.conf - - - - /var/run/symlinks/etc/resolv.conf > >> > L+ /etc/ppp/resolv.conf - - - - /var/run/symlinks/etc/ppp/resolv.conf > >> > >> Hmm having tried this I don't get it. This fragment seem to be > assuming > >> that: > >> > >> * the dirs /var/run/symlinks/etc etc. are created somehow, maybe > >> with a d line in the same file? > >> > >> * the link target files /var/run/symlinks/etc/resolv.conf exist, > >> I guess they can just be empty as connection manager is going to fix > >> them up for me anyway? > >> > >> Thanks, > >> Britton > > > > > > Right. I wanted to emphasize only the important lines, but prerequisites > are > > important too :) > > > > Here's the full file > > > > #Type Path Mode UID GID Age Argument > > d /var/run/symlinks 0755 - - - > > d /var/run/symlinks/etc 0755 - - - > > d /var/run/symlinks/etc/ppp 0755 - - - > > f /var/run/symlinks/etc/resolv.conf 0755 - - - > > f /var/run/symlinks/etc/ppp/resolv.conf 0755 - - - > > > > # NB! Connman might override this by linking > > "/usr/lib/tmpfiles.d/connman_resolvconf.conf" to > "/run/connman/resolv.conf" > > L+ /etc/resolv.conf - - - - /var/run/symlinks/etc/resolv.conf > > L+ /etc/ppp/resolv.conf - - - - /var/run/symlinks/etc/ppp/resolv.conf > > Ok this work, thanks so much. There are still a couple things I don't > get though: > > * I notice in your original replay you don't have a tmpfs for /run > (which is sym linked to from /var/run). So I wonder how the dirs > and files you create from /etc/tmpfiles.d/10-ro-symlinks.conf are > getting created, since I would think they would be on the (read-only) > root file system. Is there something missing from the fstab you > originally showed? I included this line as well: > > tmpfs /run tmpfs defaults,noatime,nosuid,size=16M 0 0 > > I haven't gone back and tested without it, so maybe it works as you > show and I just don't understand how. > The Debian images from Robert already have that set up. "/run" is a tmpfs and "/var/run" is symlink pointing to it. > * I end up with a rw tmpfs for /root (root home dir), as shown by > cat /etc/mtab (last line): > > root@bbone217slave:/etc# cat /etc/mtab > sysfs /sys sysfs rw,nosuid,nodev,noexec,relatime 0 0 > proc /proc proc rw,nosuid,nodev,noexec,relatime 0 0 > udev /dev devtmpfs rw,relatime,size=10240k,nr_inodes=55575,mode=755 0 > 0 &g
Re: [beagleboard] RTC reset after Beaglebone boot
On Thursday, 14 September 2017 20:03:50 UTC+3, RobertCNelson wrote: > > On Thu, Sep 14, 2017 at 11:50 AM, Robert Nelson > wrote: > > On Thu, Sep 14, 2017 at 12:59 AM, > > wrote: > >> Hello, > >> I have intefaced RTC DS1307 with Beaglbone. However, after the > beaglebone > >> boots up, its time is reset and RTC takes this reset time. Use of NTP > would > >> not be feasible since Beaglebone would be used in a remote location > with no > >> internet connection. What would be the solution to this issue ? > > > > As of this week, we have an easy solution. (using u-boot overlays of > > course), the trick is the rtc "aliases", so that the ds1307 gets rtc0 > > > > Seee either : > > > > > https://github.com/beagleboard/bb.org-overlays/blob/master/src/arm/BB-I2C1-MCP7940X-00A0.dts > > > > > or: > > > > > https://github.com/beagleboard/bb.org-overlays/blob/master/src/arm/BB-I2C2-PCF8523-00A0.dts > > > and here's it working: (eth is disconnected) > > debian@beaglebone:~$ dmesg | grep rtc > [1.594096] omap_rtc 44e3e000.rtc: already running > [1.594335] omap_rtc 44e3e000.rtc: rtc core: registered 44e3e000.rtc as > rtc1 > [1.776976] rtc-ds1307 1-0068: registered as rtc0 > [1.795900] rtc-ds1307 1-0068: setting system clock to 2017-09-14 > 17:01:12 UTC (1505408472) > > Regards, > > -- > Robert Nelson > https://rcn-ee.com/ > Very interesting. Which version of the kernel implements this? -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/8ab1c314-73f7-4f24-a2ed-81a0cac640cc%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] Re: how to get read-only files system these days?
On Sunday, 17 September 2017 10:45:16 UTC+3, Britton Kerin wrote: > > On Wed, Sep 6, 2017 at 7:59 AM, Tarmo Kuuse > wrote: > > On Tuesday, September 5, 2017 at 1:45:05 AM UTC+3, Britton Kerin wrote: > >> > >> Any help modifying the above recipe or pointer to how to do it these > >> days would be appreciated. > > > > > > I just declare the root file system as read-only in fstab and create > some > > ramdisks for stuff that needs to be writable. > > > > LABEL=rootfs / ext4 ro,noatime,errors=remount-ro 0 1 > > debugfs /sys/kernel/debug debugfs defaults 0 0 > > tmpfs /var/log tmpfs defaults,noatime,nosuid,mode=0777,size=128M 0 > 0 > > tmpfs /var/tmp tmpfs defaults,noatime,nosuid,size=16M 0 0 > > tmpfs /tmp tmpfs defaults,noatime,nosuid,size=128M 0 0 > > > > Service generic-board-startup (a.k.a. > /opt/scripts/boot/generic-startup.sh) > > must run once with a writable root to set up a bunch of stuff (on later > > boots it complains, but it doesn't seem to matter). After that you're > stuck > > hacking all the software which expects to be able to write to disk. For > a > > working network, "/etc/resolv.conf" needs symlinking to a writable > location. > > Many services expect to persistently store their stuff under "/var/lib". > > > > I've used tmpfiles.d to create required symlinks, e.g. > > > > $ cat /etc/tmpfiles.d/10-ro-symlinks.conf > > L+ /etc/resolv.conf - - - - /var/run/symlinks/etc/resolv.conf > > L+ /etc/ppp/resolv.conf - - - - /var/run/symlinks/etc/ppp/resolv.conf > > Hmm having tried this I don't get it. This fragment seem to be assuming > that: > > * the dirs /var/run/symlinks/etc etc. are created somehow, maybe > with a d line in the same file? > > * the link target files /var/run/symlinks/etc/resolv.conf exist, > I guess they can just be empty as connection manager is going to fix > them up for me anyway? > > Thanks, > Britton > Right. I wanted to emphasize only the important lines, but prerequisites are important too :) Here's the full file #Type Path Mode UID GID Age Argument d /var/run/symlinks 0755 - - - d /var/run/symlinks/etc 0755 - - - d /var/run/symlinks/etc/ppp 0755 - - - f /var/run/symlinks/etc/resolv.conf 0755 - - - f /var/run/symlinks/etc/ppp/resolv.conf 0755 - - - # NB! Connman might override this by linking "/usr/lib/tmpfiles.d/connman_resolvconf.conf" to "/run/connman/resolv.conf" L+ /etc/resolv.conf - - - - /var/run/symlinks/etc/resolv.conf L+ /etc/ppp/resolv.conf - - - - /var/run/symlinks/etc/ppp/resolv.conf -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/8a8ae157-b65f-402e-a2dd-ba0910c97e9d%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] Re: how to get read-only files system these days?
On Sunday, 17 September 2017 10:45:16 UTC+3, Britton Kerin wrote: > > On Wed, Sep 6, 2017 at 7:59 AM, Tarmo Kuuse > wrote: > > On Tuesday, September 5, 2017 at 1:45:05 AM UTC+3, Britton Kerin wrote: > >> > >> Any help modifying the above recipe or pointer to how to do it these > >> days would be appreciated. > > > > > > I just declare the root file system as read-only in fstab and create > some > > ramdisks for stuff that needs to be writable. > > > > LABEL=rootfs / ext4 ro,noatime,errors=remount-ro 0 1 > > debugfs /sys/kernel/debug debugfs defaults 0 0 > > tmpfs /var/log tmpfs defaults,noatime,nosuid,mode=0777,size=128M 0 > 0 > > tmpfs /var/tmp tmpfs defaults,noatime,nosuid,size=16M 0 0 > > tmpfs /tmp tmpfs defaults,noatime,nosuid,size=128M 0 0 > > > > Service generic-board-startup (a.k.a. > /opt/scripts/boot/generic-startup.sh) > > must run once with a writable root to set up a bunch of stuff (on later > > boots it complains, but it doesn't seem to matter). After that you're > stuck > > hacking all the software which expects to be able to write to disk. For > a > > working network, "/etc/resolv.conf" needs symlinking to a writable > location. > > Many services expect to persistently store their stuff under "/var/lib". > > > > I've used tmpfiles.d to create required symlinks, e.g. > > > > $ cat /etc/tmpfiles.d/10-ro-symlinks.conf > > L+ /etc/resolv.conf - - - - /var/run/symlinks/etc/resolv.conf > > L+ /etc/ppp/resolv.conf - - - - /var/run/symlinks/etc/ppp/resolv.conf > > Hmm having tried this I don't get it. This fragment seem to be assuming > that: > > * the dirs /var/run/symlinks/etc etc. are created somehow, maybe > with a d line in the same file? > > * the link target files /var/run/symlinks/etc/resolv.conf exist, > I guess they can just be empty as connection manager is going to fix > them up for me anyway? > > Thanks, > Britton > Right. I wanted to emphasize only the important lines, but prerequisites are important too :) Here's the full file #Type Path Mode UID GID Age Argument d /var/run/symlinks 0755 - - - d /var/run/symlinks/etc 0755 - - - d /var/run/symlinks/etc/ppp 0755 - - - f /var/run/symlinks/etc/resolv.conf 0755 - - - f /var/run/symlinks/etc/ppp/resolv.conf 0755 - - - # NB! Connman might override this by linking "/usr/lib/tmpfiles.d/connman_resolvconf.conf" to "/run/connman/resolv.conf" L+ /etc/resolv.conf - - - - /var/run/symlinks/etc/resolv.conf L+ /etc/ppp/resolv.conf - - - - /var/run/symlinks/etc/ppp/resolv.conf -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/a62e55b1-bc45-4783-b74f-f98715f6948a%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: RTC reset after Beaglebone boot
On Thursday, 14 September 2017 17:25:09 UTC+3, Ritisha Shettigar wrote: > > I have intefaced RTC DS1307 with Beaglbone. However, after the beaglebone > boots up, its time is reset and RTC takes this reset time. Use of NTP would > not be feasible since Beaglebone would be used in a remote location with no > internet connection. What would be the solution to this issue ? > There's a reasonably easy hack: https://learn.adafruit.com/adding-a-real-time-clock-to-beaglebone-black/set-rtc-time A proper solution would probably involve a kernel change to replace the default RTC device (which is currently the chip's nearly useless internal one at /dev/rtc0). I'd love to figure out how that works, as I need to do the same thing. -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/b98c9c08-c8e1-4383-be4c-da4ac80366ae%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: how to get read-only files system these days?
On Tuesday, September 5, 2017 at 1:45:05 AM UTC+3, Britton Kerin wrote: > > Any help modifying the above recipe or pointer to how to do it these > days would be appreciated. > I just declare the root file system as read-only in fstab and create some ramdisks for stuff that needs to be writable. LABEL=rootfs / ext4 ro,noatime,errors=remount-ro 0 1 debugfs /sys/kernel/debug debugfs defaults 0 0 tmpfs /var/log tmpfs defaults,noatime,nosuid,mode=0777,size=128M 0 0 tmpfs /var/tmp tmpfs defaults,noatime,nosuid,size=16M 0 0 tmpfs /tmp tmpfs defaults,noatime,nosuid,size=128M 0 0 Service generic-board-startup (a.k.a. /opt/scripts/boot/generic-startup.sh) must run once with a writable root to set up a bunch of stuff (on later boots it complains, but it doesn't seem to matter). After that you're stuck hacking all the software which expects to be able to write to disk. For a working network, "/etc/resolv.conf" needs symlinking to a writable location. Many services expect to persistently store their stuff under "/var/lib". I've used tmpfiles.d to create required symlinks, e.g. $ cat /etc/tmpfiles.d/10-ro-symlinks.conf L+ /etc/resolv.conf - - - - /var/run/symlinks/etc/resolv.conf L+ /etc/ppp/resolv.conf - - - - /var/run/symlinks/etc/ppp/resolv.conf -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/66d83657-ac4e-4be3-a2a9-6113408660a0%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
[beagleboard] Re: Beaglebone Black - Power supply and shutdown process.
On Tuesday, September 5, 2017 at 3:12:44 AM UTC+3, michael.ja...@gmail.com wrote: > > I am attempting to design a power backup for the beaglebone to ensure > corruption does not occur when power is lost. I understand this can happen. > Apparently it has happened with this product. > My experience shows that the ext4 file system with journaling enabled does occasionally become corrupt when pulling the plug, although I've tried to avoid writing to the eMMC (no read-only root, just redirected all disk-hoggers to tmpfs-s in RAM). Still, during first half of the year I've had 3-4 cases of file system corruption among a set of ~10 devices which get their plug pulled daily (they are powered by street lighting circuits which are off during daytime). The easiest software "fix" I've come up with so far is mandatory fsck (the file system checker) on boot. This requires adding "fsck.mode=force fsck.repair=yes" to the kernel command line in /boot/uEnv.txt. On my desk it revives the corrupt ext4 file system automatically. Definitely not ideal, but the probability of ending up with an non-booting system is reduced (can't give stats, been only running this for 2 months). > Q1. In the real world, how long does the beaglebone black take to shutdown? > For me, around 8 seconds with a recent Debian 8.6 or 9.1 from https://rcn-ee.net/ (some services, some networking). I've seen bad network configuration delay it by another 5-10 seconds. > Q2. Apparently the power button is held down for 8 seconds. Does this > shutdown period commence from the end of that 8 seconds? > Pressing and immediately releasing the power button will trigger a software-initiated shutdown. Holding the power button down will probably do a "hard" shutdown, i.e. power will be cut. > Q3. From the notes left to me, I'm assuming the continuous current > consumption to be 500mA at the 5V supply. Would anyone have an idea of what > that may be during the shutdown period. > You'd have to measure that. I doubt it'll be any different from normal operation, unless shutdown triggers some externally connected HW to consume more. -- Kind regards, Tarmo Kuuse -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/2a71fca0-cced-4ae4-a020-b0b926ff2ebb%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] Issues loading capes
On Wednesday, September 6, 2017 at 3:51:46 AM UTC+3, RobertCNelson wrote: > > > > On Sep 5, 2017 5:07 PM, "Peter Lawler" > > wrote: > > Thanks, I'll check them out. > > Just glancing at the doc now, I'm not immediately seeing to how to load > while.runing a kernel, unload then reload an newer version. I'm sure it'll > become more obvious once my caffeine levels increase. > > > > Yeah we've given up on, loading, unloading, loading in kernel land. Too > many race conditions.. Everyone has pretty much given up, and the focus is > doing it in UBoot. > > Regards, > Thank you for the explanation. As a suggestion - perhaps you could add that to http://elinux.org/Beagleboard:BeagleBoneBlack_Debian#U-Boot_Override_external_capes or somewhere? I just spent a rather frustrating day trying to get my capes to load via capemgr/slots on a fresh image and came here to ask the exact same question as Peter. Can't say I found it obvious that this is no longer possible :) -- Kind regards, Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/19712e49-d432-4b9f-a20b-542c80aedeac%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
Re: [beagleboard] Why two Ethernet over USB interfaces, usb0 and usb1?
On Thursday, August 31, 2017 at 5:20:06 PM UTC+3, RobertCNelson wrote: > > On Thu, Aug 31, 2017 at 9:15 AM, Tarmo Kuuse > wrote: > > On Thursday, August 31, 2017 at 5:11:49 PM UTC+3, RobertCNelson wrote: > >> > >> On Thu, Aug 31, 2017 at 9:07 AM, Tarmo Kuuse > wrote: > >> > I've been reading the am335x_evm.sh & friends for a while and don't > get > >> > it. > >> > Why are there two separate Ethernet over USB interfaces - eth0 and > eth1 > >> > - > >> > doing the same exact thing? They both show up when I connect a BB to > my > >> > Debian PC. > >> > > >> > Having no experience with gadget and libcomposite, I'd guess that > RNDIS > >> > and > >> > CDC ECM are both required simultaneously to somehow support > >> > Windows/Mac/Linux users out of the box? > >> > >> So this is Windows and Apple's fault.. > >> > >> Windows = CDC ECM = needs signed driver.. > >> Apple = RNDIS = needs signed driver... > >> > >> Windows will use the RNDIS with a built-in driver, thus user doesn't > >> have to install anything... > >> Apple will use the CDC ECM with a built-in driver, thus user doesn't > >> have to install anything... > >> > >> So in linux, we end with two interfaces, because Windows and Apple > >> don't have a unified built-in driver... > > > > > > Fair enough, thanks for the explanation. > > > > Since the comments in "autoconfigure_usb1.sh" say it's handling CDC/ECM, > I > > assume that the other one (usb0) does RNDIS. > > it's only because we enable rndis first thus it becomes usb0: > > > https://github.com/RobertCNelson/boot-scripts/blob/master/boot/am335x_evm.sh#L503-L531 > > > Regards, > > > -- > Robert Nelson > https://rcn-ee.com/ > Oky ... /long blank stare/ ... I think I mostly get it now. The sysfs hierarchy groups them both under "functions/*.usb0" which is an excellent red herring :) -- Tarmo -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/b1a20d7e-071c-4da3-8d8c-652e1d3a6138%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.