https://tinker-board.asus.com/forum/index.php?/discover/Tinker Board Forum - All ActivityenUART0 is a little different because its CTS/RTS pins are also exposed on the 40-pin header. However, with rtscts=False and the normal 8N1 configuration, hardware flow control should not be required. I would first check the pin mux/configuration. On Tinker Board 3S, UART0 must be enabled in /boot/config.txt: intf:uart0=on Then reboot and verify that /dev/ttyS0 is actually mapped to the expected TX/RX pins. UART0 TX/RX are on pins 8 and 10, while CTS/RTS are on pins 16 and 18. For a basic MCU connection, try only TX, RX and GND (cross TX↔RX) and test with: serial.Serial("/dev/ttyS0", 115200, timeout=1, rtscts=False, dsrdtr=False) If UART1/4/9 work but UART0 does not, I would suspect the pin mux/device-tree configuration or wiring before the Python code.Wed, 26 Aug 2026 17:52:04 +0000Hi everyone, I'm planning to build an AI agent that can automate tasks, interact with APIs, and make decisions based on user input. My goal is to create something that can handle real-world workflows rather than just answer questions. When I searched on the internet about the tools that can help me to build the AI agent i found this post and https://www.reddit.com/r/AI_Agents/comments/1vg730b/ai_agent_builders_in_2026_still_dont_meet_me/ this reddit article. author listed around 10, but when I searched on the intenent about all of them very confused. basically i'm trying to understand: Which AI agent framework would you recommend (LangGraph, AutoGen, CrewAI, OpenAI Agents SDK, etc.)? What tech stack are you using for production AI agents? How do you handle memory, tool calling, and long-running workflows? What are the biggest challenges you've faced while building and deploying AI agents? Any open-source templates, GitHub repositories, or tutorials you'd recommend for someone building a scalable AI agent? I'd really appreciate hearing about your experiences, best practices, and any lessons learned. Thanks in advance!Thu, 06 Aug 2026 15:41:31 +0000now it is 2026 and the exactly same problem here - same camera, same OS image, same error in log - Failed to detect device. This camera worked well with RPi. (gst-launch-1.0:1505): GStreamer-CRITICAL **: 11:56:58.810: gst_value_set_int_range_step: assertion 'end % step == 0' failed ERROR: from element /GstPipeline:pipeline0/GstV4l2Src:v4l2src0: Internal data stream error. Additional debug info: gstbasesrc.c(3055): gst_base_src_loop (): /GstPipeline:pipeline0/GstV4l2Src:v4l2src0: streaming stopped, reason not-linked (-1)Thu, 25 Jun 2026 11:58:05 +0000@Frank_Chiang Thank you so much for your kind response! Best.Thu, 09 Apr 2026 23:58:58 +0000Hi Sonyan, In general, we don’t recommend disabling the UART log, since it removes an important recovery and debugging channel. If you still need to disable it, please try to disable the status of /fiq-debugger in device tree. This will stop console output during the kernel and system stages. For U-Boot, we recommend using CONFIG_SILENT_CONSOLE to silence its output. Thank you.Wed, 08 Apr 2026 01:49:57 +0000Hi all, I'm deploying Tinker Board 3S units in an embedded product where physical debug interfaces need to be locked down for security. I want to completely disable the debug UART console (3-pin header, ttyFIQ0 @ 1500000 baud) so that no shell access or boot log output is possible through the serial port. What I've considered so far: - Removing "console=ttyFIQ0,1500000" from the kernel command line - Masking getty@ttyFIQ0.service - Disabling the UART node in the device tree (status = "disabled") My questions: 1. Is removing the kernel console parameter + masking getty sufficient, or does U-Boot itself also output to this UART before the kernel takes over? If so, is there a way to silence U-Boot output as well? 2. Is there a recommended device tree overlay approach to fully disable this UART on the RK3566? 3. Are there any side effects to disabling this UART (e.g., boot failures, watchdog issues)? Environment: - TinkerOS Debian on eMMC - Headless deployment, SSH-only access over Ethernet Any guidance or experience with production hardening on TB3S would be appreciated. Thanks!Sat, 04 Apr 2026 15:33:45 +0000Update. Hi Owen Chen, The original dts file is the file for mipi2lvds2_G156HAN02.dts , I've tweaked the timings a bit. I found the problem: the LVDS power supply from the ThinkBoard (3.3V) drops below 3.1V under panel load. The panel requires 3.15V to 3.6V, which is causing the image distortion. The panel works perfectly with an external power supply. Do you know the maximum current the ThinkBoard can deliver at its LVDS output? Or is there a solution for this?Wed, 04 Mar 2026 16:13:46 +0000Hi Chris, Thank you for the information. You mentioned using an overlay from the /boot/overlays/ folder. Could you let us know where this specific .dts file originated from? We are asking because we actually do not include this file in our official release, and we do not support the G156HCE-L01 panel on our Debian system.Wed, 04 Mar 2026 09:15:56 +0000Thanks for the reply, I got Debian from https://tinker-board.asus.com/download.html#os Here is the screenshot of the command output:Wed, 04 Mar 2026 08:17:29 +0000Hi Chris, Thank you for choosing our product. To help us better analyze the situation, could you please let us know which operating system and version you are currently using? For example, you can usually retrieve this information by running the following commands in your terminal: $cat /etc/version $cat /etc/os-release We look forward to your reply so we can assist you further.Wed, 04 Mar 2026 06:54:38 +0000Hallo, I have a ThinkBoard 3N Lite and want to control a G156HCE-L01 panel via LVDS. I used one of the existing overlays from the /boot/overlays/ folder and tried to adapt it to my panel. After several attempts, I'm still having the problem that simple images (login screen and normal desktop) work. However, when I have a window or the black terminal on the screen, the panel starts flickering, or black/colored stripes appear. I think there might be something wrong with the timing values or LVDS settings. What is worng with the LVDS settings? Can anyone help me? mipi2lvds2_G156HCE_L01.dtsTue, 03 Mar 2026 14:26:17 +0000We are currently working on Tinker 2S board for one of our project. We have the requirement of using both SPI buses of the Tinker 2S board, but currently we are facing an issue with the SPI5 bus. Scenario 1 : We connected MOSI and MISO in a loop and enabled loop back, it worked with a normal wire length but after increasing the wire length it stopped working. We tried the same scenario with the SPI1 bus and it was working without any issue. But we used twisted pair wire with the same increased length on the SPI5 bus and it was working fine that time. Scenario 2 : We connected a SPI slave to SPI1 bus and it worked without any issue, but connecting the same slave to SPI5 and it was not working. We enabled SPI5 from OS and all connections was as per the pin out diagram itself. What may be the issue that SPI5 bus is not working properly?Thu, 19 Feb 2026 09:21:30 +0000I am using the Tinker 2S board and am unable to use SPI5, as it always returns junk values. However, the same SPI slave works correctly with SPI1. What could be the issue?Thu, 19 Feb 2026 04:45:02 +0000Hi TzuWen, Thank you for the information. I disabled usbdevice.service as you suggested, and I have confirmed that the service was successfully stopped. This resolved my issue. Thanks again for your support. Best regards,Thu, 15 Jan 2026 00:58:22 +0000Hi Toru, You can stop the usbdevice.service directly. Please use the following command: sudo systemctl disable usbdevice.service BRs, TzuWenFri, 26 Dec 2025 09:36:58 +0000To people who is having same issue, I found possible solution for this version of TinkerOS. The problem was that dhcpcd5 software is accessing network interfaces before udev tries to set Predictable Names for network interfaces which actually blocks udev executing the change. So the possible solution is to delay starting dhcpcd service until udev completes its task. This is why network interfaces get their predictable names correctly when dhcpcd is disabled to start at boot. Possible solution includes creating a file at location: /etc/systemd/system/dhcpcd.service.d/override.conf with further contents: [Unit] After=network-online.target Wants=network-online.target After this file is created, dhcpcd service can be enabled to start at boot.Mon, 22 Dec 2025 11:31:12 +0000Hello, I would like to ask how to properly disable ADB on TinkerOS. I am using a Tinker Board 3S with TinkerOS (Debian 11 V1.0.7). For security reasons, I want to prevent ADB from automatically starting when a USB device is connected, since this device will be used to run a custom application in a production environment. At the moment, I am disabling ADB by manually removing the ADB function from the USB gadget configuration: sudo rm /sys/kernel/config/usb_gadget/asus/configs/b.1/f-ffs.adb sudo rmdir /sys/kernel/config/usb_gadget/asus/functions/ffs.adb I also commented out the ADB-related sections inside /usr/bin/usbdevice (lines 260–272): # /usr/bin/usbdevice adb_prepare() { # usb_mount adb /dev/usb-ffs/adb -o uid=2000,gid=2000 -t functionfs # usb_start_daemon /usr/bin/adbd # usb_wait_files -m /dev/usb-ffs/adb } adb_stop() { # usb_stop_daemon /usr/bin/adbd } Are there any concerns or drawbacks with disabling ADB in this way? Is there an official or recommended method to disable ADB in TinkerOS? Any advice would be greatly appreciated.Thu, 11 Dec 2025 05:41:52 +0000Hi Anyone can advise how to install python3-ASUS.gpio. Now error is unable to locate packageWed, 10 Dec 2025 04:42:30 +0000I have mistaken when downloading old source code. After trying with latest source code, Audio works normally now. Please ignore. Thank you !Sun, 16 Nov 2025 04:42:55 +0000Sat, 15 Nov 2025 14:46:56 +0000That’s an impressive milestone for the Armbian team — they’ve managed to balance performance improvements with community-driven innovation, which isn’t easy in open-source projects of this scale.Mon, 03 Nov 2025 20:34:36 +0000Tue, 28 Oct 2025 07:44:36 +0000hello @LucD, sorry to hear that. could you provide a screencap of the failure so we can further investigate the issue?Thu, 23 Oct 2025 14:48:34 +0000Replying to this old thread because I had the same problem this week on the Tinker Board (not S) R2.0: I tried both of the Android 11 images with the same issue as @AJofre - after running adb reboot-bootloader the logo screen appears but the device gets stuck on that screen and the device is not visible when I run fastboot devices Tinker Board R2.0 /Tinker Board S R2.0 Android 11 V1.0.6 Tinker Board S R2.0 Android 11 V1.0.0 However, I was able to work around the problem as follows: Reboot the device to boot normally (I had to power off the Tinker Board, remove the SD card, power it on again, power off, reinsert SD card, power on again) adb shell su dd if=/storage/emulated/0/Download/magisk_patched-XXX.img of=/dev/block/by-name/boot adb reboot After the reboot, I'm able to grant root permissions to apps that request them.Tue, 21 Oct 2025 20:16:34 +0000@TzuWen_Chang Hello. Thank you so much for your kind advice. I will try. Best regards,Tue, 21 Oct 2025 05:52:04 +0000