Quotetmcatalin Is it any technical reason to create from zero new partitions instead of using existing ones? Nope. Except that you can create a somewhat bigger array, by reclaiming the 2 487MB partitions. But Quote Did anyone using btrfs on these boxes see improved performances over mdadm? Does make sense to change from mdadm? I can't say anything about performance, but it does make senseby Mijzelf - Debian
I don't have real answers to your questions, but Quoteinstalling a new kernel with dpkg -i kirkwood-debian-xxx.deb automatically starts dkms, recompiling r8168 code from kernel source tree AFAIK that isn't true. dkms is designed to automatically compile modules which are not in the source tree. (Or which have an incompatible/outdated stock module). Yet it uses the header files fromby Mijzelf - Debian
r8168-8.051.02 is 3 years old. Meanwhile 8.055.00 is available, which might better match your kernel. (and compiler).by Mijzelf - Debian
Check the handshake on your terminal emulator. If hardware handshake is enabled no data is send.by Mijzelf - uBoot
> I guess you have recently updated the kernel and rootfs? Yes. There is a clean install of Armbian on emmc, which is up-to-date. I want to try to install Debian 13 on an SSD, using the netinstaller. But I probably will have to provide the files on μSD.by Mijzelf - uBoot
=> mii info NULL device name! No such device: <NULL> NULL device name! No such device: <NULL> NULL device name! No such device: <NULL> (repeated 32 times in total) => mdio list No MDIO bus found => setenv ipaddr 192.168.1.50 => setenv serverip 192.168.1.254 => ping 192.168.1.254 dw_dm_mdio_init: mdio node is missing, registering legacy mdio busdw_dmby Mijzelf - uBoot
I'm trying to load files over tftp to my Kobol Helios64. This box has 2 ethernet ports, one 2.5Gbit, and one 1Gbit. When I give any network related command, it gives a repeated 'dw_dm_mdio_init: mdio node is missing, registering legacy mdio busdw_dm_mdio_init'. But the command 'net list' (according to 'help net' the only subcommand of net) gives => netby Mijzelf - uBoot
Maybe a stupid question, have you checked the USB-serial device actually works?by Mijzelf - uBoot
What do you mean by officially? It is orderable at AliExpress and Amazon. https://sfconservancy.org/activities/openwrt-one.htmlby Mijzelf - Off-Topic
kokosz887 Wrote: ------------------------------------------------------- > These comments do not work for me on ubuntu. Ubuntu on a PC? You are supposed to execute this in a shell on the NAS itself.by Mijzelf - uBoot
bodhi Wrote: ------------------------------------------------------- > disconnect power for a while (at least 30 minutes) > and power up. Why is that?by Mijzelf - uBoot
Wouldn't that mean that you can simply use rtc1 instead of rtc0?by Mijzelf - Debian
Was the original firmware able to wakeup the NAS on RTC?by Mijzelf - Debian
It's common practice to use the names uImage and uInitrd (which could be symlinks to uImage-<version> etc) for u-boot images. vmlinux normally is a 'raw' kernel.by Mijzelf - uBoot
I suggest you to consider to create 3 6TB volumes instead. Apart from having one big volume raid0 has no advantages here. It won't be faster, as the NIC is the bottleneck, and not the raw throughput of a single disk. But it has disadvantages. A single disk failing will kill all your data, and all disks will have to spin up to access the volume.by Mijzelf - Debian
Linux can handle GPT. That is unrelated to how the bootloader managed to start it. So yes, if your u-boot doesn't support GPT and you use a single MBR disk to load Linux, Linux will be able to use GPT on the other disks. Probably it can even use GPT on the MBR disk, if you create a hybrid partition table. QuoteIf the box cannot handle GPT, can I use 4+ TB MBR drives, but with the sector siby Mijzelf - Debian
Quote ## Loading Ramdisk Image at 02000000 ... Bad Magic Number� OpenWrt doesn't use a ramdisk. It's flash space will/is being used by the overlay. So you'll have to change the u-boot environment. https://github.com/openwrt/openwrt/commit/4e46ae1f69b5ff1cf09b2e2eee69a7c38a00202dby Mijzelf - uBoot
There are at least 2 different versions of the Networkspace 2. https://web.archive.org/web/20180720230834/http://lacie.nas-central.org/wiki/Category:Network_Space_2 I suppose that means that there should be 2 different u-boot's either.by Mijzelf - uBoot
The support for the original i386 was dropped in Linux 3.8, more than 10 years ago. AFAIK the main problem with the i386 was a missing CMPXCHG opcode, which made it hard to maintain.by Mijzelf - Debian
I'm afraid my 2Big2 died a few years ago. Bad capacitor, I thought, but exchanging it didn't solve the problem. So either my diagnostic or my soldering capacities are bad. However, according to strings the u-boot of the D2 Network2 supports the usb subsystem. And while I was searching for a u-boot dump for the 2Big2 or 5Big2 I found this: https://plugout.net/viewtopic.php?f=26&t=by Mijzelf - Debian
I think it's a bit more complicated than that. My old 2Big2 didn't have a writable u-boot config, and the stock environment reads a uImage from the raw partition sda6, and then boots it with the kernel commandline root=/dev/sda7. A default Debian boot with initramfs is not possible without exchanging u-boot, AFAIK. However, you can simply write your rootfs to /dev/sda7 and write the uby Mijzelf - Debian
The Mindspeed Comcerto alas doen't have upstream support in the Linux kernel. Only the kernel provided in the vendor provided SDK is available (3.2.54). In your case using the stock kernel is also problematic, as the SoC supports both 4k and 64k memory pages, and the stock kernel uses 64k, which is not compatible with the 'normal' Debian arm7hf rootfs. But you can have a look herby Mijzelf - uBoot
It should be something like eth0.ipaddr=192.168.1.191 eth0.serverip=192.168.1.139 https://community.zyxel.com/en/discussion/comment/35771/#Comment_35771by Mijzelf - uBoot
Quote HD44780: open(/dev/lcd) failed: Operation not permitted What does 'ls -l /dev/lcd*' say? Quote What indicates that the LCD display on the RN104 is a USB device and not a parallel port device? It's not a patallel port device. What was the last time you saw a parallel port? It could be usb, or i2c, or gpio connected, or... . The driver should take care for that.by Mijzelf - Displays
I can't imagine you'd need /dev/parport0. I suppose that is used when you have connected an LCD display to the parallel port, and lcd4linux does the bitbanging in user mode. But you have a driver which should do that work. According to the documentation there should be a /dev/lcd. This wiki suggests that lcd4linux can also handle that.by Mijzelf - Displays
When the position and sizes of the u-boot and u-boot env partitions didn't change, I *think* you only need the stock u-boot environment. The firmware itself is capable of filling the other partitions. Put back the original u-boot environment, download the right RescueStick here, unzip it to a FAT formatted USB thumb drive. Extract the uImage from the file ras.bin (binwalk can tell you whereby Mijzelf - uBoot
As long as the bootloader isn't damaged, you can recover it. But you'll need serial connection.by Mijzelf - uBoot
@martini: That is a neat trick. Do you have any idea why it works? I'd say that for ddr training it doesn't matter if the binary is read from uart or from nand. So which subtle effect is causing this?by Mijzelf - uBoot
In that case mkimage will be dependent on other files which cannot be found due to the different path. You can try to chroot it: chroot /volumeUSB1/usbshare/ /bin/sh That creates a new root in /volumeUSB1/usbshare/, and executes /bin/sh from there. So /volumeUSB1/usbshare/ becomes / in that shell. Then you can cd /boot /usr/bin/mkimage ... By executing 'exit' you leave theby Mijzelf - Debian