Empty tcrp, can't boot #33
-
|
Thanks for the awesome work! One week in havenI managed to install it last week, and was up and running with no Problem. After a power loss when I rebooted it wouldn't boot anymore. I tried to re-build, re-install the image, different flash drives, etc, but no luck. Setup
OutputOutput on boot`Loading Linux... Please press Enter to activate this console. [2024/03/20 20:55:03:0670] N: ttyd 1.7.3 (libwebsockets 4.3.2-2023.08.4) After boot pressing key / running boot.sh`root@tcrpfriend:~# ./boot.sh !!NOTE: This file is empty!! Populating user_config.jsonAt this point I can see the it on the network at \192.168.1.198\tcrpfriend\tcrp Output after config`root@tcrpfriend:# ./boot.sh device node not found Model : DS3622xs+, Serial : 20C0SQRTZSWZA, Mac :c1d297c8b765, Build : 7.2.1-69057, Update : 0, LKM : prod Experiments/observationsHonorable mentions in logWhen running Re-Install DSM:: Mounting devtmpfs ... [FAILED] [ 7.221787] redpill: module verification failed: signature and/or required key missing - tainting kernel ` Constant errorSome constant warningI get this every minute or so, but I got it even when the thing was working Maybe from the onboard USB storage[ 202.162026] usb 8-1: device descriptor read/64, error -110 Things I tried
Bonus questionNow that I have formatted one of the 4 drives from Raid 5 it should theoretically still be possible to recover the data if I get it working right? Is it complicated? I've only had a week's worth of data there, but I recovered some old pictures from a failed drive and I'm not sure if it survives an other deep scan Could someone help me with some guidance, please? |
Beta Was this translation helpful? Give feedback.
Replies: 4 comments 6 replies
-
|
It has been moved from the issues section to the discussion section. The boot device you soldered in must be either Sata or USB. fdisk -l | grep "Disk /dev/sd" udevadm info --query property --name /dev/sdv |
Beta Was this translation helpful? Give feedback.
-
|
Please excuse me, I didn't know. Thank you for getting back to me so quickly. ResultsNote: (I'm not sure if it is important) Outputsfdisk -l | grep "Disk /dev/sdroot@tcrpfriend:# " It is a 64Gb Kingstone Datatravel There is no SDV, but there is a,b,c udevadm info --query property --name /dev/sdvdevice node not found udevadm info --query property --name /dev/sdaDEVLINKS=/dev/disk/by-diskseq/9 /dev/disk/by-id/ata-ST4000VN008-2DR166_ZDH9KSDT /dev/disk/by-id/wwn-0x5000c500c92b036f udevadm info --query property --name /dev/sdbDEVLINKS=/dev/disk/by-diskseq/10 /dev/disk/by-id/usb-SMI_USB_DISK_AA04012700008612-0:0 /dev/disk/by-path/pci-0000:00:1d.7-usb-0:1:1.0-scsi-0:0:0:0 udevadm info --query property --name /dev/sdcDEVLINKS=/dev/disk/by-diskseq/11 /dev/disk/by-id/usb-Kingston_DataTraveler_3.0_E0D55E6C71161580E9118F56-0:0 /dev/disk/by-path/pci-0000:00:1d.7-usb-0:4:1.0-scsi-0:0:0:0 SumSo basically
NotesAlso, I see this on the network. A shared drive appears with three folders: This is the structure: Ghost sdvI don't know why is there no SDV suddenly. boot log of reinstall (When booting Re-install.log)========== BEGIN DUMP OF ALL PARTITIONS DETECTED =========== /dev/sda1 8192 16785407 16777216 fd /dev/sda2 16785408 20979711 4194304 fd /dev/sda3 21241856 7813832351 7792590496 fd /dev/sdb1 8192 16785407 16777216 fd /dev/sdb2 16785408 20979711 4194304 fd /dev/sdb3 21241856 7813832351 7792590496 fd /dev/sdc1 8192 16785407 16777216 fd /dev/sdc2 16785408 20979711 4194304 fd /dev/sdc3 21241856 5860328351 5839086496 fd /dev/sdd1 8192 16785407 16777216 fd /dev/sdd2 16785408 20979711 4194304 fd /dev/sdd3 21241856 5860328351 5839086496 fd /dev/sdv1 2048 147455 145408 83 /dev/sdv2 147456 301055 153600 83 /dev/sdv3 301056 3147775 2846720 83 |
Beta Was this translation helpful? Give feedback.
-
|
Do you suspect that /dev/sdb was a tcrp bootloader used in the past and was damaged due to a power outage? I think you need to check additionally with the two commands below to be sure. fdisk -l /dev/sdb TCRP requires only one bootloader. I think the problematic /dev/sdb is being used first. /dev/sdb, /dev/sdc |
Beta Was this translation helpful? Give feedback.
-
No.
But it was always like that right? USB prioiorityI've also filled out VID/PID in the user config.json Outputsfdisk -l /dev/sdbDisk /dev/sdb: 956 MiB, 1002438656 bytes, 1957888 sectors Device Boot Start End Sectors Size Id Type It is a 64Gb Kingstone Datatravel fdisk -l /dev/sdcDisk /dev/sdc: 57.73 GiB, 61991813632 bytes, 121077761 sectors Device Boot Start End Sectors Size Id Type It is a 64Gb Kingstone Datatravel I am a bit reluctant to format the dev/sdb flash drive, because that has the original software, and now that is the only thing that is working. UpdateI also used to give device node not found, and I see the effect of yesterday's changes, so instead now, I get Not Supported Loader BUS Type, program Exit!!! Since then I don't see the mounted folders if I access the drive through network sharing(I mean tcrp, tcrp-p1, p2 are not visible anymore) |
Beta Was this translation helpful? Give feedback.
-
|
In that case, you should leave /dev/sdb and skip it. Recent changes add the ability to use nvme and mmc as loader boot devices. I'm not sure if you can understand the script, but change it like below. |
Beta Was this translation helpful? Give feedback.
-
|
So cool! I am an embedded software engineer, so I know scripts, just not overly familiar with Unix systems It seems so that it worked with the diff you sent. Will keep you posted |
Beta Was this translation helpful? Give feedback.
-
|
Even if /dev/sdb exists, menu_m.sh, which was changed to the latest version, does not seem to be affected. The final changes to menu_m.sh are as follows: update_tinycore() has nothing to do with updating menu_m.sh. Can you take a screenshot of what the latest menu_m.sh looks like on your PC? Separately from this, friend v0.1.0v has now been released. ./down.sh |
Beta Was this translation helpful? Give feedback.
-
|
Thank you so much for the continuous help, and suggestions. After reinstalling while migrating from the three original drives I was able to start up, and begin a repair of my storage pool. I am extremely happy |
Beta Was this translation helpful? Give feedback.
-
|
Thank you for your donation. |
Beta Was this translation helpful? Give feedback.
In that case, you should leave /dev/sdb and skip it.
I am planning to modify and distribute the function to find the bootloader in friends.
I think it will be friend version v0.1.0v.
Recent changes add the ability to use nvme and mmc as loader boot devices.
In the process, boot devices that do not correspond to usb / sata / nvme / mmc, etc. may be recognized as errors.
Since this is the case for your 1G USB DISK device, we are adding a function to skip it.
I'm not sure if you can understand the script, but change it like below.
PeterSuh-Q3/tcrpfriend@ab0468a
I will distribute it after finishing testing.