All Activity

# Orange Pi Zero 3 (1.5GB) Random Crashes — Root Cause Found & Fixed (No Recompilation Required) **Board:** Orange Pi Zero 3 — 1.5GB LPDDR4 variant **SoC:** Allwinner H618 **OS:** Armbian Trixie **Kernels tested:** 6.6.75-legacy, 6.18.33-current, 6.16.8-edge **Status:** ✅ SOLVED — 72+ hours stable uptime confirmed --- ## The Problem My Orange Pi Zero 3 (1.5GB model) was randomly crashing every few hours under all three Armbian kernel branches. The symptoms were: - System freezes with no response (SSH drops, ping still works briefly) - Kernel oops logged: `kswapd0 → ext4_es_scan → paging request fault` - Random segfaults on normal binaries (`sudo`, `curl`, `apt`) - Corrupted library names in memory (e.g. `libGap-Jg.sK.0` instead of `libcrypto.so.3`) - `memtester` passed clean (2 loops, 1GB) - `badblocks` found zero bad blocks - `fsck` found no filesystem errors - Crashes happened even at full idle, with only ~200MB RAM in use The crashes occurred on **all three kernel branches** — legacy, current, and edge — ruling out a kernel-specific bug as the root cause. --- ## Investigation ### What we ruled out (in order) | Suspect | Test | Result | |---|---|---| | SD card physical failure | `badblocks -sv` | 0 bad blocks | | Filesystem corruption | `fsck -f` | Clean | | RAM hardware failure | `memtester 1G x2` | No errors | | eth-optimize.sh (ring buffer 1024, txqueuelen 10000) | Removed from crontab | Extended stability but didn't fix | | Kernel version | Tested 6.6.75, 6.18.33, 6.16.8 | All crashed | | DRAM size detection bug | `dmesg \| grep Memory:` | Consistently showed 1572864K (correct 1.5GB) | ### The kernel oops — the real clue ``` kernel: Unable to handle kernel paging request at virtual address 14ae620800000018 kernel: CPU: 0 PID: 56 Comm: kswapd0 kernel: Call trace: kernel: __es_tree_search.isra.0+0x20/0xa4 kernel: es_reclaim_extents+0x58/0xf0 kernel: ext4_es_scan+0xf0/0x29c kernel: do_shrink_slab+0x174/0x298 kernel: shrink_slab+0xb4/0x2f0 kernel: kswapd+0x18c/0x3b8 ``` The address `14ae620800000018` is in "address between user and kernel address ranges" — a classic corrupted/invalid pointer. This happens when the kernel accesses memory that physically doesn't exist or wraps around. ### DRAM address wraparound — the root cause The H618 1.5GB LPDDR4 variant has a known DRAM address wraparound issue, documented in the linux-sunxi community (notably by **ag123** in Armbian forums and **Jernej Skrabec's** u-boot size scan patches): > *"It is quite possible the addresses wrap around in the 1.5GB LPDDR4 DRAM chips and the 'test' for memory there returns a false positive."* In this specific case, the **size** is correctly detected as 1.5GB — but the **upper address region** (above 1GB physical) exhibits wraparound behavior. When the kernel accesses the upper ~512MB region (via page cache, kswapd reclaim, or slab allocation), writes to those addresses silently corrupt data in the lower region — because the addresses "wrap" back. This explains why: - `memtester` passes (it tests sequentially in a contiguous region, doesn't trigger wraparound patterns) - Size appears correct (1572864K = 1.5GB every boot) - Crashes are random in timing (depends on when page cache/kswapd reaches the upper region) - ALL kernels are affected (this is a u-boot/hardware level issue, not kernel-specific) --- ## The Fix ### Standard approach (requires u-boot recompilation) The proper fix is `CONFIG_DRAM_SUN50I_H616_TRIM_SIZE=y` in u-boot, or Jernej Skrabec's adjusted size scan procedure (columns-first scanning). This correctly detects and excludes the wraparound region, allowing full 1.5GB usage. ### Our approach — kernel `mem=` parameter (no recompilation needed) Since the wraparound occurs in the **upper address region**, we can simply tell the kernel to only use the safe lower 1GB: Edit `/boot/armbianEnv.txt` and add `mem=1024M` to `extraargs`: ```bash sudo nano /boot/armbianEnv.txt ``` Change: ``` extraargs=... existing params ... ``` To: ``` extraargs=mem=1024M ... existing params ... ``` Reboot and verify: ```bash cat /proc/cmdline | grep -o "mem=1024M" cat /proc/iomem | grep "System RAM" free -h ``` Expected output: ``` mem=1024M 40080000-7fffffff : System RAM total used free Mem: 974Mi ... ``` The `iomem` map confirms the kernel's physical RAM range is **exactly** `0x40080000–0x7fffffff` — stopping at the 1GB boundary. The upper wraparound region (`0x80000000` and above) is completely excluded from the kernel's memory map. ### Why this works The wraparound region begins at or above `0x80000000` (1GB physical offset). With `mem=1024M`: - Kernel allocator, page cache, slab, and all processes stay within `0x40000000–0x80000000` - `kswapd` never reclaims pages from the problematic upper region - The corrupted-pointer crash path is never triggered ### Tradeoffs vs. u-boot fix | | mem=1024M | u-boot TRIM fix | |---|---|---| | Requires recompilation | ❌ No | ✅ Yes | | RAM available | ~974MB | ~1.4GB | | Works on all kernels | ✅ Yes | ✅ Yes | | Fixes root cause | Workaround | Proper fix | For SDR servers, home automation, and similar light workloads where ~974MB is more than sufficient (our system uses ~130–200MB), this is a perfectly viable permanent solution. --- ## Results | Metric | Before fix | After fix | |---|---|---| | Uptime | 1–8 hours max | **72+ hours** (and counting) | | Kernel oops | Multiple per session | **Zero** | | Segfaults | Random, on any binary | **Zero** | | `sudo apt update` | Crashed frequently | Runs perfectly | | Memory corruption | Frequent bit-flips | **None observed** | System configuration at time of testing: - **Kernel:** 6.16.8-edge-sunxi64 - **Kernel param:** `mem=1024M` - **CPU governor:** ondemand - **Workload:** OpenWebRX+, SpyServer, RTL-TCP (x2), radiosonde_auto_rx, Xray proxy --- ## How to apply ```bash # Backup first sudo cp /boot/armbianEnv.txt /boot/armbianEnv.txt.bak # Add mem=1024M to extraargs sudo sed -i 's/^extraargs=/extraargs=mem=1024M /' /boot/armbianEnv.txt # Verify grep extraargs /boot/armbianEnv.txt # Reboot sudo reboot ``` After reboot, confirm: ```bash # Should show mem=1024M cat /proc/cmdline | grep -o "mem=1024M" # Should show ~974Mi total free -h # Should show System RAM ending at 0x7fffffff cat /proc/iomem | grep "System RAM" ``` --- ## References - ag123's analysis of 1.5GB wraparound: Armbian Community Forums - Jernej Skrabec's u-boot DRAM size scan fix: [linux-sunxi mailing list](https://www.mail-archive.com/u-boot@lists.denx.de/msg516769.html) - H618 LPDDR4 support patch: [u-boot mailing list](https://lore.kernel.org/all/20231111091000.26744-2-iuncuim@gmail.com/T/) - ag88's 1.5GB u-boot fix (alternative approach): GitHub --- ## Notes This fix was developed and tested on a **1.5GB LPDDR4** Orange Pi Zero 3. The 1GB, 2GB, and 4GB variants use different DRAM configurations and may not be affected by this specific issue. If you have a 1.5GB board and experience similar random crashes, try `mem=1024M` before spending time on kernel debugging or hardware replacement. It may save you days of investigation. **Tested and confirmed working by:** Özgür Çetinoğlu **Location:** Athens, Greece **Setup:** 24/7 SDR server (OpenWebRX+, radiosonde tracking) **Date:** June 2026

@Taz the sources for that repository is a clone of my build. It’s a very basic attempt to build a 6.19 kernel. You are better off using my build. If you got 6.6 booting then 6.18 should work as well.

@sven-ola About the issue you saw (re. UUID=63ee7593-e111-4547-ac2f-6bdb8519ce11..), what I have done is flash either of the two images at https://sven-ola.commando.de/privat-in/ to a 128GB MicroSD card using Win32DiskImager or Rufus on Windows, and then booted the RV2 with it, that's all. The RV2 has had two PCI devices connected at boot. This should not affect the boot sequence. I sent you all kinds of weird boot sequences I had especially with the Linux 7.10 Armbian image. I gave a few hours to try these Armbian images on the RV2, and ultimately it's not stable. The biggest issue is to boot from the M.2 SSD. Then, the 6.18 boots well from MicroSD but the 7.10 not. Finally my SFP+ NIC doesn't work well on the 6.18. I had more success with the testing in the beginning, e.g. I got the 7.10 image to work booted from the SSD. Could there be an issue with power supply, for example with the 5V to 3.3V converter on the PCB. For power supply I use a good 5V @ 5A USBC power supply. The most successful test I did was, flash a MicroSD card with the 6.18 image, and boot it on the RV2. Then from inside that environment, do dd if=Armbian-unofficial_26.05.0-trunk_Orangepirv2_trixie_edge_7.1.0-rc3_minimal.img of=/dev/nvme0n1 bs=100M status=progress; sync . Booting from that NVMe did work a few times. I installed the u-boot bundled with Armbian to the SPI using armbian-config. I'm not sure this was a good idea, at least it did not make NVMe booting work better.

This article describes how I successfully configured a PWM fan on an Orange Pi 5 Max running Armbian Debian Trixie Minimal vendor 6.1.115, using the opifancontrol utility. Special thanks to Jamie Sinclair for creating and sharing the opifancontrol project with the community. I have an Orange Pi 5 Max installed in a metal case and running several Docker containers, including Frigate, Immich, Samba, and others. The case originally came with a internal 30x30x10 mm fan, which was not able to provide sufficient cooling. Under load, CPU temperatures were reaching nearly 70°C. To improve cooling, I installed an external 40x40x10 mm PWM fan that was originally used on an Odroid HC4. The fan was mounted using screws, which required drilling holes in the case cover. I also added several ventilation holes, in the case cover above the CPU area to improve airflow. After these modifications, the CPU temperature now stays around 48°C. 1. Install wiringOP Follow the instructions from the wiringOP project: Download wiringOP $ apt-get update $ apt-get install -y git $ git clone https://github.com/orangepi-xunlong/wiringOP.git Build wiringOP $ cd wiringOP $ ./build clean $ ./build Verify the installation: $ gpio readall This command should display the Orange Pi 5 Max GPIO pin map. 2. Configure GPIO Pin 7 as a PWM Output In Armbian, configure GPIO Pin 7 (PWM3_IR_M3) as a PWM output. $ sudo armbian-config Navigate to: System └─ Kernel └─ Overlays Select: rk3588-pwm3-m3 Press the space bar to enable it, save the configuration, and reboot the system. 3. Install opifancontrol: https://github.com/jamsinclair/opifancontrol $ su $ cd /usr/local/bin/ $ wget https://github.com/jamsinclair/opifancontrol/blob/main/opifancontrol.sh $ chmod +x /usr/local/bin/opifancontrol.sh $ curl -sSL https://raw.githubusercontent.com/jamsinclair/opifancontrol/main/install.sh | bash Enable the service to start automatically at boot: $ systemctl enable opifancontrol.service Start the service: $ systemctl start opifancontrol.service Check the service status: $ systemctl status opifancontrol.service 4. PWM Fan Wiring I used a 5V, 4-wire, 4,000 RPM fan from an Odroid HC4. Connection details: Fan Wire Function GPIO Pin Red +5V Power Pin 4 Black GND Pin 6 Blue PWM Speed Control Pin 7 Yellow TACH (RPM Sensor) Not Connected 5. Adjust the Fan Control Configuration Edit the configuration file: $ sudo nano /etc/opifancontrol.conf My settings are: TEMP_LOW=45 FAN_LOW=60 TEMP_MED=60 FAN_MED=75 TEMP_HIGH=70 FAN_HIGH=100 RAMP_UP_DELAY_SECONDS=30 RAMP_DOWN_DELAY_SECONDS=60 After saving the configuration file, restart the service: $ sudo systemctl stop opifancontrol.service $ sudo systemctl start opifancontrol.service With these settings and the hardware modifications described above, my Orange Pi 5 Max now operates significantly cooler, maintaining temperatures around 48°C while running multiple Docker services

Then wipe spi and retry. You can also pull a backup, simply via dd

I'm having trouble booting my Orange Pi5 off an SD Armbian_26.5.1_Orangepi5_trixie_vendor_6.1.115_minimal.img. I just get a blank screen. Older versions of Armbian do boot off an SD card, it also boots off NVME. I do have an old version of U-Boot flashed on SPI (Thanks to Joshua Riek), which I use because it is compatible with my KingSpec NVME drives, other versions of U-Boot did not recognize KingSpec NVME, including Armbian versions of U-Boot. I last tested Armbian U-Boot about 6 months ago, it failed. The Console log looks normal-ish right up to the point it just stops. Console Log --- DDR 9fffbe1e78 cym 24/02/04-10:09:20,fwver: v1.16 LPDDR4X, 2112MHz channel[0] BW=16 Col=10 Bk=8 CS0 Row=16 CS1 Row=16 CS=2 Die BW=16 Size=2048MB channel[1] BW=16 Col=10 Bk=8 CS0 Row=16 CS1 Row=16 CS=2 Die BW=16 Size=2048MB channel[2] BW=16 Col=10 Bk=8 CS0 Row=16 CS1 Row=16 CS=2 Die BW=16 Size=2048MB channel[3] BW=16 Col=10 Bk=8 CS0 Row=16 CS1 Row=16 CS=2 Die BW=16 Size=2048MB Manufacturer ID:0x1 CH0 RX Vref:28.5%, TX Vref:19.8%,20.8% CH1 RX Vref:29.3%, TX Vref:20.8%,20.8% CH2 RX Vref:28.5%, TX Vref:21.8%,21.8% CH3 RX Vref:29.7%, TX Vref:22.8%,21.8% change to F1: 528MHz change to F2: 1068MHz change to F3: 1560MHz change to F0: 2112MHz out U-Boot SPL board init U-Boot SPL 2017.09 (Aug 31 2024 - 15:22:22) unknown raw ID 41 18 20 Trying to boot from MMC2 spl: partition error Trying fit image at 0x4000 sector ## Verified-boot: 0 ## Checking atf-1 0x00040000 ... sha256(7612223b82...) + OK ## Checking u-boot 0x00800000 ... sha256(642bfeda4e...) + OK ## Checking fdt-1 0x008d6c48 ... sha256(7b2c4c6dbe...) + OK ## Checking atf-2 0x000f0000 ... sha256(b2af21b504...) + OK ## Checking atf-3 0xff100000 ... sha256(70505bb764...) + OK Jumping to U-Boot(0x00800000) via ARM Trusted Firmware(0x00040000) Total: 266.68/489.129 ms --- +

I have posted a new release on GitHub. Improved version. https://github.com/joilg/x88pro New: Support for Seekwave EA6521 Wifi Chip on Hardware Version 1.4 Frontdisplay shows now Systen Clock. USB3 support now SuperSpeed (5Gbps) IR Remote control implenented Target OS Ubuntu resolute 26.01 LAN is limited to 100mbps. (major advantage is the low quiescent current in standby mode. An external PHY with 1Gbps consumes a significant amount of power in Wake-on-LAN mod) Feedback welcome.

Hello @4A studio. The armbian-config tool may change large parts of the system configuration without a real "undo". I am not sure if I understand your situation correct: Board does not boot. You see UART output but no UART login prompt? Board is booting. But you don't see a live picture on your HDMI monitor?

Hey @armfan. UUID=63ee7593-e111-4547-ac2f-6bdb8519ce11 was read from /boot/extlinux/extlinux.conf which is correct for the edge/7.1.0rc3 image. However, something must be wrong with your SD card if u-boot can read UUID from extlinux.conf, but kernel does not find corresponding file system on the same boot media. This is the expected kernel startup : [ 0.000000] Booting Linux on hartid 0 [ 0.000000] Linux version 7.1.0-rc3-edge-spacemit (build@armbian) (riscv64-linux-gnu-gcc (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0, GNU l [ 0.000000] Machine model: OrangePi RV2 [ 0.000000] SBI specification v1.0 detected [ 0.000000] SBI implementation ID=0x1 Version=0x10003 [ 0.000000] SBI IPI extension detected [ 0.000000] SBI RFENCE extension detected [ 0.000000] earlycon: sbi0 at I/O port 0x0 (options '') [ 0.000000] printk: legacy bootconsole [sbi0] enabled Loading, please wait... Starting systemd-udevd version 257.13-1~deb13u1 Begin: Loading essential drivers ... done. Begin: Running /scripts/init-premount ... done. Begin: Mounting root file system ... Begin: Running /scripts/local-top ... done. HTH // Sven-Ola

@hmoob What part do you need clarify? I can help you but I need more information.

Providing logs with armbianmonitor -u helps with troubleshooting and significantly raises chances that issue gets addressed.

I have the same issue than @1111Windows had back in March. That's because I chose the wrong option in armbian-config. How do I revert to original (or, at least, working) u-boot? UART connection through putty is working.

Has anyone successfully set up a MIPI screen with their LCD ribbon connectors on Armbian? Any wisdom from someone who has done this would be appreciated.

@sven-olaTrying the latest RC2 and RC3 7.1 bundles at https://sven-ola.commando.de/privat-in/ by flashing it to a MicroSD card and booting from it, I get the following boot failure. Booting your 6.18 image works though. Any idea how make it work? By the way, is the U-Boot binary inside Armbian the same in your 7.1 and 6.18 versions? Perhaps I could boot the 6.18 from MicroSD card, use it to flash the SPI, and then try to boot the 7.1 from the SPI? Update: I booted the RV2 off MicroSD card via the old 6.18 version. From inside this 6.18, I did "dd if=the-7.1-version of=/dev/nvme0n1 bs=1M status=progress", powered off, took out the MicroSD, and booted, and it worked! Now a question back to you: @sven-ola Now that the boot process works, do I still need to install the latest Armbian-bundled u-boot on the SPI chip via "armbian-install"?

That sounds like a Klipper message, not from the armbian OS. If there is new functionality in Klipper it will show this error, but you don't always need to reflash the MCU depending on if you use those functions. You can post in a more detailed error and versions in a Klipper forum for advice.

I ran into this as well. Downgrade to kernel 6.12.58 fixed the physical interface. My board that initially had Trixie required the NetworkManager fix to get wifi working even after reverting. Something about the upgrade and/or downgrade broke netplan.

As a data point, I had a CB1 with Trixie that was working fine with netplan until I did an apt upgrade to 26.2.1 and kernel 6.18.xx. The upgrade made my physical interface disappear, which required a kernel downgrade (to 6.12.58) to get back. But even after the downgrade my wireless interfaces still wouldn't connect. The ultimate fix was to switch this board to NetworkManager like I did with my other CB1 that was running bookworm and required the NetworkManager fix from day one.

I'm getting this weird graphical artifact on a Radxa Rock 4SE connected to a TV via HDMI. It covers roughly a third of the right side of the display, note that the terminal window is covered by the artifact but the cursor is above it. I started from a Debian minimal image and added the desktop environment via armbian-config. The artifact is visible on both GNOME and KDE Plasma (with fully updated system on both rolling and stable) but not with XFCE. It also disappears with resolutions lower than 4K or switching to tty. Any idea what might be causing this?

Tested: - Odroid C4 / HC4 - Tinkerboard - Odroid XU4 - Rockpi E

Hi, I'm using it at the moment as a containers server, thanks for link. I not that changing kernel may run in trouble, so I still keep the 6.1.115 and wait for update. Thanks for your help Regards

Hi, It depends on what you want to do with your device. You should check this sheet if the features you depend on are actually available in mainline linux: https://gitlab.collabora.com/hardware-enablement/rockchip-3588/notes-for-rockchip-3588/-/blob/main/mainline-status.md Also take note that you may run into trouble booting when switching kernels. The reason is that vendor kernel images may (I don't know for this particular board) also come with (from guts feel from the stoneage) 2017 vendor uboot which may have issues booting an up to date mainline kernel. Fixing can be a bit of a pain if you don't have a second arm64 device which you could plug your microsd (via usb adapter for example) into, chroot into and fix this.

I'm using Armbian Orange Pi 5 Plus based on Debian 13 with vendor kernel v26.2.1 for Orange Pi 5 Plus running Armbian Linux 6.1.115-vendor-rk35xx I'm thinking about switching to Armbian based on Debian 13 with kernel current 6.18.33 because 6.1.115 is an old one => current the kernel is more recent Is that a good idea to swith to vendor kernel to current ? How can I do this ?

It's fixed in 7.1.0-rc6-bleedingedge-rockchip64 Cheers !

Hi there, I recently upgraded my Orange Pi 5 Plus to "Armbian 26.5.1" and noticed in the MOTD (on logins) that it is still showing version "26.2.1". I found differences in this 2 files, and the latter is the one used as reference for MOTD: cat /etc/os-release: PRETTY_NAME="Armbian 26.5.1 trixie" NAME="Debian GNU/Linux" VERSION_ID="13" VERSION="13 (trixie)" VERSION_CODENAME=trixie DEBIAN_VERSION_FULL=13.5 ID=debian HOME_URL="https://www.armbian.com" SUPPORT_URL="https://forum.armbian.com" BUG_REPORT_URL="https://www.armbian.com/bugs" ARMBIAN_PRETTY_NAME="Armbian 26.5.1 trixie" cat /etc/armbian-release: # PLEASE DO NOT EDIT THIS FILE BOARD=orangepi5-plus BOARD_NAME="Orange Pi 5 Plus" BOARDFAMILY=rockchip-rk3588 BUILD_REPOSITORY_URL=https://github.com/armbian/build BUILD_REPOSITORY_COMMIT=77f919f6c LINUXFAMILY=rockchip64 ARCH=arm64 BOOT_SOC=rk3588 IMAGE_TYPE=stable BOARD_TYPE=conf INITRD_ARCH=arm64 KERNEL_IMAGE_TYPE=Image KERNEL_TARGET=current,edge,vendor KERNEL_TEST_TARGET=vendor,current FORCE_BOOTSCRIPT_UPDATE= FORCE_UBOOT_UPDATE= OVERLAY_DIR="/boot/dtb/rockchip/overlay" VENDOR="Armbian" VENDORCOLOR="247;16;0" VENDORDOCS="https://docs.armbian.com" VENDORURL="https://www.armbian.com" VENDORSUPPORT="https://forum.armbian.com" VENDORBUGS="https://www.armbian.com/bugs" BOOTSCRIPT_FORCE_UPDATE="no" BOOTSCRIPT_DST="boot.cmd" VERSION=26.2.1 REVISION=26.2.1 BRANCH=edge Can somebody explain why armbian-release doesn't get properly updated? I belive this is not the first time I see this, I recall seeing this behaviour from previuos upgrades too. Thanks!

These Chinese people are very creative. I never imagined seeing a TV box with a SATA port. @Astlin You will need to interact with the .dts file to build a .dtb file with this capability. Before taking any action, check if your board is not from this other JP-Box project. https://github.com/armbian/build/blob/main/patch/kernel/archive/rockchip64-7.1/dt/rk3566-jp-tvbox.dts The topic you should interact with is this one here: https://forum.armbian.com/topic/31887-jianpian-rk3566-tv-box-8g32g-develop-log/#comment-175700 I don't actually have access to the JP-box @Astlin. If you are using the H96 max RK3566 firmware, the SATA port may not be enabled, but you can enable kernel changes within the Armbian compilation menu and insert your changes into the board's DTS file. https://github.com/armbian/build/blob/main/patch/kernel/archive/rockchip64-7.1/dt/rk3566-h96-tvbox.dts I believe the person responsible for this sign is @tdleiyao @ning have tested our DTB file on JianPian device https://github.com/armbian/build/blob/main/config/boards/jp-tvbox-3566.tvb If you have time, we can update the JP-box project in the correct topic. https://forum.armbian.com/topic/31887-jianpian-rk3566-tv-box-8g32g-develop-log/#comment-175700 I have maintained a sample repository with the necessary data to enable RK3566 cards in Armbian. https://github.com/hqnicolas/ArmBoardBringUp The reason JP-box isn't working is that it was enabled in kernel 6.6, and we're already on 7.1, nobody updates this since then. You can use the H96 modifications as start point. this modifications apply to the JP-box, but some functions, such as the SATA port, require customization.