@bodhi I was tempted to install your newest uBoot but not seeing ix2-200 specifics. That read error has been ongoing but with the previous uInitrd it passes checksum. Can I get us more detail somehow? Here's what the previous does NAND read: device 0 offset 0x100000, size 0x500000 5242880 bytes read: OK NAND read: device 0 offset 0x1000000, size 0xe00000 14680064 bytes read: ERRby damnit_ - Debian
Now that my boot up issues are resolved per https://forum.doozan.com/read.php?2,95424, thank you very much @bodhi I have new issue trying to migrate to RAID1 as shown here https://forum.doozan.com/read.php?2,27308 Made new uInitrd and flashed to NAND after initramfs updated from dpkg-reconfigure mdadm But uBoot hates it NAND read: device 0 offset 0x100000, size 0x500000 5242880 bytesby damnit_ - Debian
Found a typo in bootcmd setenv bootcmd 'setenv bootargs $(bootargs_console) root=LABEL=rootfs rootfdelay=10 earlyprintk=serial; run bootcmd_usb; bootcmd_stock' Should be setenv bootcmd 'setenv bootargs $(bootargs_console) root=LABEL=rootfs rootfdelay=10 earlyprintk=serial; run bootcmd_usb; run bootcmd_stock' Fixed all 100% working!! @bodhi thank you so much for aby damnit_ - Debian
@bodhi Now I attempt boot without USB stick attached, NAND boot looks unhappy.. ends in kernel panic trying to find HDD rootfs Hit any key to stop autoboot: 0 NAND read: device 0 offset 0x100000, size 0x500000 5242880 bytes read: OK NAND read: device 0 offset 0x1000000, size 0xe00000 14680064 bytes read: ERROR ## Booting image at 00800000 ... Image Name: Linux-5.2.9-kirkby damnit_ - Debian
Ah, you taught me earlier, just had to do some hexadecimal mathematics... setenv mtdparts 'mtdparts=orion_nand:0x100000@0x000000(uboot)ro,0x20000@0xA0000(uboot_env),0x500000@0x100000(uImage),0xE00000@0x1000000(uInitrd)' Seems to fit now! root@debian:~# dmesg | grep 0x0 [ 0.000000] Booting Linux on physical CPU 0x0 [ 0.000000] Kernel command line: console=ttyS0,115200 mby damnit_ - Debian
Ultimately I still have the issue of your uImage not fitting in nand.. Can I make /dev/mtd2 larger somehow? root@debian:~# cat /proc/mtd dev: size erasesize name mtd0: 00100000 00004000 "uboot" mtd1: 00020000 00004000 "uboot_env" mtd2: 00300000 00004000 "uImage" mtd3: 01000000 00004000 "uInitrd" root@debian:~# ls -al /boot total 65492 drby damnit_ - Debian
@bodhi, Apparently there's still something it does not agree with... __ __ _ _ | \/ | __ _ _ ____ _____| | | | |\/| |/ _` | '__\ \ / / _ \ | | | | | | (_| | | \ V / __/ | | |_| |_|\__,_|_| \_/ \___|_|_| _ _ ____ _ | | | | | __ ) ___ ___ | |_ | | | |___| _ \ / _ \ / _ \| __by damnit_ - Debian
@bodhi I believe NAND mtd2 may still be intact as I had flashed back my mainline uImage when I saw that yours did not fit. I had already performed steps 1-4 to HDD from my debianlive on laptop, so for your test I simply copied that rootfs to USB stick so I'm ready to boot from either HDD or USB. Let me know how the envs should be set. Thanks! Marvell>> printenv baudrateby damnit_ - Debian
I reloaded my original HDDs with jessie rootfs and booted from the USB stick to try nandwrite with your uImage & uInitrd but it fails root@debian:~# nandwrite -p /dev/mtd2 /mnt/uImage-bodhi Image 4975416 bytes, NAND page 512 bytes, OOB area 16 bytes, device size 3145728 bytes nandwrite: error!: Input file does not fit into device error 0 (Success) nandwrite: error!: Data wasby damnit_ - Debian
I created it per your method https://forum.doozan.com/read.php?2,12096by damnit_ - Debian
@bodhi Added the line to fstab, it was indeed missing. I still get kicked to initramfs prompt?!by damnit_ - Debian
@bodhi Indeed I neglected to label the rootfs correctly, however now that I've done that I still cannot boot the HDD?! user@debian:~$ sudo e2label /dev/sdb1 user@debian:~$ sudo e2label /dev/sdb1 rootfs user@debian:~$ sync user@debian:~$ sudo e2label /dev/sdb1 rootfs ALERT! LABEL=rootfs does not exist. Dropping to a shell!by damnit_ - Debian
@bodhi I've given up on my install with RAID for the time being. I formatted both drives and followed your technique to install buster on ext3 partition on one drive. Can you advise how I should now configure the boot parameters to make this work? Presently it fails to find rootfs and kicks me to initramfs prompt. Also should the rootfs be in first or second drive slot? Marvell>>by damnit_ - Debian
The USB “uInitrd-new” and “uImage-new” that I can boot OK with and utilize Debian rootfs on HDD were created in the Debian instance. Netboot did all the dpkg for me. USB was originally created with an Arch Virtualbox solely for the netboot. Then uInitrd & uImage we created and cp’d from working Debian instance on HDD, to the USB. Hence I know not why those images will not boot direcby damnit_ - Debian
Ah, I neglected to indicate my intended rootfs is on HDD. The USB is just a jessie netboot I used for install to the RAID1. Yet if I issue u-boot params to use uImage & uInitrd of USB, I can already boot to Debian rootfs of HDD?! However those files we’re created in Debian and simply copied to USB, so I’m not fully understanding why I cannot already use them to boot directly to HDD rootfsby damnit_ - Debian
Almost! Still a NAND read error, but 5.2.9 initramfs checksum OK. Seems it can't find rootfs for some reason... __ __ _ _ | \/ | __ _ _ ____ _____| | | | |\/| |/ _` | '__\ \ / / _ \ | | | | | | (_| | | \ V / __/ | | |_| |_|\__,_|_| \_/ \___|_|_| _ _ ____ _ | | | | | __ ) ___by damnit_ - Debian
Same result as before unfortunately: root@debian:~# dmesg | grep 0x0 [ 0.000000] Booting Linux on physical CPU 0x0 [ 0.000000] Kernel command line: console=ttyS0,115200 mtdparts=orion_nand:0x100000@0x000000(uboot)ro,0x20000@0xA0000(uboot_env),0x300000@0x100000(uImage),0x1000000@0x1000000(uInitrd) root=/dev/mapper/debian--vg-root rootdelay=10 [ 14.697747] pci 0000:00:00.0: [11ab:62by damnit_ - Debian
Here's the result, not sure that worked... root@debian:~# dmesg | grep 0x0 [ 0.000000] Booting Linux on physical CPU 0x0 [ 0.000000] Kernel command line: console=ttyS0,115200 mtdparts=orion_nand:0x100000@0x000000(uboot)ro,0x20000@0xA0000(uboot_env),0x300000@0x100000(uImage),-0x1000000(uInitrd) root=/dev/mapper/debian--vg-root rootdelay=10 [ 16.766926] pci 0000:00:00.0: [11ab:by damnit_ - Debian
@bodhi Best process for strip? Repart? This starts to extend beyond my *nix skillset. Will repart potentially be more reliable re futureproofing? Also, so I may understand, the bad block is on flash not hdd so alternate hdd will have no bearing? Thanks again and again! How about your newest kirkwood setup and rootfs in this instance?by damnit_ - Debian
@bodhi thanks so much for your prompt help, I should have posted those details with my last. Could that bad block be messing up my checksum?? root@debian:~# cat /proc/mtd dev: size erasesize name mtd0: 00100000 00004000 "uboot" mtd1: 00020000 00004000 "uboot_env" mtd2: 00300000 00004000 "uImage" mtd3: 01000000 00004000 "uInitrd" root@debby damnit_ - Debian
OK thank you! With corrected mtdparts argument Debian now shows as it should. root@debian:~# cat /proc/mtd dev: size erasesize name mtd0: 00100000 00004000 "uboot" mtd1: 00020000 00004000 "uboot_env" mtd2: 00300000 00004000 "uImage" mtd3: 01000000 00004000 "uInitrd" I proceed with flash_erase and nandwrite to /dev/mtd2 and /dev/mtd3 respecby damnit_ - Debian
Well now I have a new wrinkle. I did not experience the nandwrite error for mtd0 the first time around which appears now. I did see the mtd1 bad block but assumed it was maybe from the factory. These uImage and uInitrd were from mkimage in fresh Debian and are the ones that boot fine as cp'd to USB. cd /boot mkimage -A arm -O linux -T ramdisk -C gzip -a 0x00000000 -e 0x00000000 -n iniby damnit_ - Debian
My ix2-200 running stock is setup per Nobiscuit but I'm not able to boot from internal flash and disk. Works fine with USB boot to disk. When I flashed my uImage and uInitrd files to nand there was a bad block, not sure if that's the cause or how to resolve but here's the pertinent info: __ __ _ _ | \/ | __ _ _ ____ _____| | |by damnit_ - Debian