amouiche
commented
2 years ago Hello David.
Thank you for reporting this issue. Looks like fw_setenv had some issues to properly configure the uboot env.
- Did you, by chance, still have the backup of the /dev/mtd4 before you performed the partition resize (? (ie. the uboot env partition)
- I'm also interested by a dump of your current /dev/mtd4 to fully understand what was done. Can you do it and share with us ?
as root: cat /dev/mtd4 > /tmp/mtd4.dump
If you zip those files they will be very small.
Regards, Arnaud
Le 21/12/2021 à 19:38, David J. Fiddes a écrit :
Discovered a problem when using qnap_mtd_resize.py at c9ccf4d https://github.com/amouiche/qnap_mtd_resize_for_bullseye/commit/c9ccf4d4b6fe9802d6558b9a88609c68709b2ddf to upgrade my TS-219P+. This unit uses Debian Bullseye with the Buster kernel pinned as per the official release notes. The upgrade appeared to work correctly but when rebooting it failed to boot reporting "Bad Magic" and returning to the u-boot prompt.
The log of the script runing follows:
|[Check of the QNAP model and see if supported] kirkwood-qnap: machine: QNAP TS219 family DTB file: kirkwood-ts219-6282.dtb Checking: flashcp -V Checking: flash_erase --version ici Using 'u-boot-tools' package [find on which MTD device partitions are currently mounted] spi0.0 [Dump current U-boot config'] Current U-boot bootcmd: cp.l 0xf8200000 0x800000 0x080000;cp.l 0xf8400000 0xa00000 0x240000;bootm 0x800000 Current U-boot bootargs: console=ttyS0,115200 root=/dev/ram initrd=0xa00000,0x900000 ramdisk=34816 [Prepare new 'bootcmd'] Old: cp.l 0xf8200000 0x800000 0x080000;cp.l 0xf8400000 0xa00000 0x240000;bootm 0x800000 New: cp.l 0xf8100000 0x800000 0xc0000;cp.l 0xf8400000 0xb00000 0x300000;bootm 0x800000;echo Kernel_legacy layout fallback;bootm 0x900000 [Prepare new 'bootargs'] Old: console=ttyS0,115200 root=/dev/ram initrd=0xa00000,0x900000 ramdisk=34816 New: console=ttyS0,115200 root=/dev/ram initrd=0xb00000,0xc00000 ramdisk=34816 @.**@*.**@*.**@*.**@*.(U-Boot @.(NAS Config)" @.**@*.**@*.**@*.**@*.(U-Boot @.(NAS Config)" [Prepare fw_setenv script (/tmp/fw_setenv.script)] [Dump current 'NAS config' and 'Kernel' images] + cat /dev/mtd5 [Resize 'NAS config' dump from 1280KB to 256KB.] + modprobe loop + losetup --show -f /tmp/mtd_nas_config.dump + loopdev=/dev/loop0 + e2fsck -f -p -v /dev/loop0 17 inodes used (13.28%, out of 128) 2 non-contiguous files (11.8%) 0 non-contiguous directories (0.0%) # of inodes with ind/dind/tind blocks: 0/0/0 51 blocks used (4.98%, out of 1024) 0 bad blocks 0 large files 6 regular files 2 directories 0 character device files 0 block device files 0 fifos 0 links 0 symbolic links (0 fast symbolic links) 0 sockets ------------ 8 files + e2fsck -f -p -v /dev/loop0 17 inodes used (13.28%, out of 128) 2 non-contiguous files (11.8%) 0 non-contiguous directories (0.0%) # of inodes with ind/dind/tind blocks: 0/0/0 51 blocks used (4.98%, out of 1024) 0 bad blocks 0 large files 6 regular files 2 directories 0 character device files 0 block device files 0 fifos 0 links 0 symbolic links (0 fast symbolic links) 0 sockets ------------ 8 files + resize2fs /dev/loop0 128 resize2fs 1.46.2 (28-Feb-2021) Resizing the filesystem on /dev/loop0 to 128 (1k) blocks. The filesystem on /dev/loop0 is now 128 (1k) blocks long. + losetup -d /dev/loop0 [Concatenate first 256K of 'NAS config' with first 1MB of Kernel > /tmp/mtd_nas_config.new] [Prepare second 1MB of kernel tail > /tmp/mtd_kernel.tail] ------------------------------------------------------------ !!!! Warning !!!! Everything is fine up to now. Next steps will write the flash and may be subject to failures. It is highly recommended to perform a MTD backup and save the files somewhere (USB device, PC) cat /dev/mtd0 > /tmp/mtd0.uboot.backup cat /dev/mtd1 > /tmp/mtd1.kernel.backup cat /dev/mtd2 > /tmp/mtd2.rootfs1.backup cat /dev/mtd3 > /tmp/mtd3.rootfs2.backup cat /dev/mtd4 > /tmp/mtd4.uboot-config.backup cat /dev/mtd5 > /tmp/mtd5.nas-config.backup fw_printenv -c /tmp/fw_env.config > /tmp/uboot_config.backup.txt cd /tmp tar cvzf mtd_backup.tgz mtd?.*.backup uboot_config.backup.txt # now use scp / sftp to push/pull mtd_backup.tgz on another PC. Be sure you will not cut the power until the end of operations. In case of failure, you way need to recover with a Serial Console to run U-boot commands https://www.cyrius.com/debian/kirkwood/qnap/ts-219/serial/ Continue and flash the new partitions ? (y/N) y [Flash 'NAS config' partition content (ie 'NAS config' + head of Kernel) (still a 'safe' op)] + flashcp -v /tmp/mtd_nas_config.new /dev/mtd5 Erasing blocks: 5/5 (100%) Writing data: 1280k/1280k (100%) Verifying data: 1280k/1280k (100%) [Change U-boot config with new values)] + fw_setenv -c /tmp/fw_env.config -s /tmp/fw_setenv.script [Flash tail of the kernel in old 'Kernel' Partition] + flashcp -v /tmp/mtd_kernel.tail /dev/mtd1 Erasing blocks: 4/4 (100%) Writing data: 1024k/1024k (100%) Verifying data: 1024k/1024k (100%) ------------------------------------------------------------ SUCCESS. Now, REBOOT ! Notes: - DO NOT PERFORM A KERNEL OR SYSTEM UPDATE before the next reboot !... so don't wait too long. - Consider compressing initrd with 'xz' to optimize the size with: echo "COMPRESS=xz" > /etc/initramfs-tools/conf.d/compress |
I have a serial console connected so was able to figure out what was happening and temporarily change the bootcmd and bootargs to what the script said they should be and the machine started up.
This was my original configuration: uboot_config.backup.txt https://github.com/amouiche/qnap_mtd_resize_for_bullseye/files/7757290/uboot_config.backup.txt
The failing configuration looked like this: qnap_mtd_resize-post-run-config-2021-12-21.txt https://github.com/amouiche/qnap_mtd_resize_for_bullseye/files/7757323/qnap_mtd_resize-post-run-config-2021-12-21.txt
I was able to permanently fix the config using fw_setenv after boot so it looks like: fixed-uboot-config-2021-12-21.txt https://github.com/amouiche/qnap_mtd_resize_for_bullseye/files/7757329/fixed-uboot-config-2021-12-21.txt. This now reboots cleanly and I was able to upgrade to the newer Bullseye 5.10 kernel and reboot successfully. Thanks!
— Reply to this email directly, view it on GitHub https://github.com/amouiche/qnap_mtd_resize_for_bullseye/issues/13, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADXHI5GX3EJV6QN5Y4ZSJIDUSDCSDANCNFSM5KQ2GWKA. Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.
You are receiving this because you are subscribed to this thread.Message ID: @.***>
davefiddes
cavokz
Discovered a problem when using qnap_mtd_resize.py at c9ccf4d4b6fe9802d6558b9a88609c68709b2ddf to upgrade my TS-219P+. This unit uses Debian Bullseye with the Buster kernel pinned as per the official release notes. The upgrade appeared to work correctly but when rebooting it failed to boot reporting "Bad Magic" and returning to the u-boot prompt.
The log of the script runing follows:
I have a serial console connected so was able to figure out what was happening and temporarily change the bootcmd and bootargs to what the script said they should be and the machine started up.
This was my original configuration: uboot_config.backup.txt
The failing configuration looked like this: qnap_mtd_resize-post-run-config-2021-12-21.txt
I was able to permanently fix the config using fw_setenv after boot so it looks like: fixed-uboot-config-2021-12-21.txt. This now reboots cleanly and I was able to upgrade to the newer Bullseye 5.10 kernel and reboot successfully. Thanks!