Database Users

[Q] No result from parted /dev/block/mmcblk0

After restore to original stock 6.2.2 from TWRP, my KF is stucked in Kindle Fire logo. ADB on but always face exec system/bin/sh failed so I already made factory cable and put KF into fastboot successfully.
I tried to install TWRP again successfully but when I tried to upgrade FFF to latest version by running
Code:
fastboot -i 0x1949 flash bootloader u-boot.bin
I got the message FAILED, partition does not exist.
To make clear my partition is ok or not, I repeated TWRP installation and let it stay in Install Complete phase to go to recovery with ADB.
I run the command
Code:
adb shell
~# parted /dev/block/mmcblk0
And the result is
Code:
GNU Parted 1.8.8.1.179-aef3
Using /dev/block/mmcblk0
Welcome to GNU Parted! Type 'help' to view a list of commands.
(parted)
next I enter
Code:
print
but nothing happen and it seem that print command stuck.
Anyone please kindly advise what I should do next in order to recover my bricked KF ? Thank you very much

maybe this helps:
http://forum.xda-developers.com/showpost.php?p=22401099&postcount=2
follow the links also ...
good luck !

Thank you for your advice, I tried to follow the link and tried to run
Code:
fastboot -i 0x1949 boot bootloader u-boot.bin
I get failed, cannot load bootloader
but if I run
Code:
fastboot -i 0x1949 boot u-boot.bin
The result is
Code:
C:\kfu\tools>fastboot -i 0x1949 boot u-boot.bin
creating boot image...
creating boot image - 219136 bytes
downloading 'boot.img'... OKAY [ 0.128s]
booting... OKAY [ 0.000s]
finished. total time: 0.128s
And when I tried to run oem format, it stuck.
Code:
C:\kfu\tools>fastboot -i 0x1949 oem format
...
I have no experience how long the re-partitioning takes, still wonder whether cancel this command or let it run T_T

would try the format command without -i 0x1949
the initialization is'nt needed with u-boot.bin ver.1.2 and the command is'nt designed to work with initialization

I tried to run command but failed. <waiting device>

you have allready the hot booted fff loaded ?
give -i 0x18d1 a try at the format command
bradys had a simmilar problem:
http://forum.xda-developers.com/showthread.php?t=1369405&page=10
follow the next 2 sites ...
hope this helps

you have allready the hot booted fff loaded ?
Click to expand...
Click to collapse
I guess not because fastboot -i 0x1949 boot bootloader u-boot.bin did not work and fastboot -i 0x1949 boot u-boot.bin either because after that command finishs, the fastboot oem format return <waiting device>, it means that fff 1.2 is not loaded successfully.
give -i 0x18d1 a try at the format command
Click to expand...
Click to collapse
tried but <waiting device>
bradys had a simmilar problem:
http://forum.xda-developers.com/show...369405&page=10
follow the next 2 sites ...
hope this helps
Click to expand...
Click to collapse
Thank you for your prompt support.
Not really, the symtoms are similar but in his case, the partition table still able to read but mine couldn't be. And he also has chance to rebuild partition mannually but I do not; I cannot run parted /dev/block/mmcblk0, this command return nothing as i mentioned above. T_T

yeutinh said:
I guess not because fastboot -i 0x1949 boot bootloader u-boot.bin did not work and fastboot -i 0x1949 boot u-boot.bin either because after that command finishs, the fastboot oem format return <waiting device>, it means that fff 1.2 is not loaded successfully.
Click to expand...
Click to collapse
your right about that - have also read it in an other thread - loads but don't reload itself
are you able to issue mklabel, mkpart and the other commands like here:
http://forum.xda-developers.com/showpost.php?p=20880465&postcount=58
maybe give eldarerathis a pm - he's the guru on partitioning:
http://forum.xda-developers.com/member.php?u=2676559

are you able to issue mklabel, mkpart and the other commands like here:
http://forum.xda-developers.com/show...5&postcount=58
Click to expand...
Click to collapse
Tried but failed, command stuck
maybe give eldarerathis a pm - he's the guru on partitioning:
http://forum.xda-developers.com/member.php?u=2676559
Click to expand...
Click to collapse
Thanks for advice, already PM him.
Code:
cat /proc/partitions
major minor #blocks name
179 0 7553024 mmcblk0
All partitions gone T_T

Got your PM. I haven't seen anything quite like this before, so all I can really do is take something of an educated guess here. What it looks like may have happened is that the partition table got corrupted or erased, so parted can't read the partition list (again, just a bit of a guess).
Try doing this and post the output that you see:
Boot into recovery
Use adb shell to connect to your Fire
Run "fdisk -l /dev/block/mmcblk0" (that's a lowercase "L")
If (3) doesn't seem to work, try "fdisk /dev/block/mmcblk0". It should give you a little prompt kind of like parted does. Type "p" and hit enter at the fdisk prompt (then use "q" to quit).
If fdisk gives you something sane then your partition table is okay. If it blows up or throws back an error then it's toast, but you might be able to use fdisk to create a new (blank) partition table and then re-create the correct partition layout. Either way, it should get you some more information about what's actually going on with your disk.

Thank you for your support, I appreciate that
Code:
~ # fdisk -l /dev/block/mmcblk0
fdisk -l /dev/block/mmcblk0
fdisk: can't open '/dev/block/mmcblk0': I/O error
~ # fdisk /dev/block/mmcblk0
fdisk /dev/block/mmcblk0
fdisk: can't read from /dev/block/mmcblk0
it means the filesystem on eMMC was accidentally broken down, correct ? But I wonder what make FS down because I did not do anything that impact to partitions, I think
and when I run mount, the result is
mount
rootfs on / type rootfs (rw)
tmpfs on /dev type tmpfs (rw,relatime,mode=755)
devpts on /dev/pts type devpts (rw,relatime,mode=600)
proc on /proc type proc (rw,relatime)
sysfs on /sys type sysfs (rw,relatime)
Click to expand...
Click to collapse
Already checked whether eMMC dead or alive http://forum.xda-developers.com/showpost.php?p=13895268&postcount=48
Code:
~ # cat /sys/devices/platform/mmci-omap-hs.1/mmc_host/mmc0/mmc0:0001/name
cat /sys/devices/platform/mmci-omap-hs.1/mmc_host/mmc0/mmc0:0001/name
M8G2FA
Returned name of eMMC
Code:
~ # cat /proc/kmsg | grep mmc0
cat /proc/kmsg | grep mmc0
<6>[ 2.438171] mmc0: new high speed DDR MMC card at address 0001
<6>[ 2.456115] mmcblk0: mmc0:0001 M8G2FA 7.20 GiB
it means that my chip still alive Need your help to clear the messed FS

The I/O error from fdisk is concerning, and kind of makes it sound like the hardware up and died somehow. I was thinking you might be able to use fdisk to create a blank MSDOS partition table, then use parted to switch it to a GPT one (the Fire uses GPT and fdisk doesn't support that), but if it can't even read the disk then that won't work. The fact that FFF's 'fastboot oem format' isn't working also kind of lends to the hardware possibility, but I don't know exactly how it performs the format (you could try PM'ing pokey9000 and see if he'll explain it to you, maybe).
To be honest, I don't think I've seen a lot of recovery options for this sort of thing. The only stuff I can think of are all kind of Linux-geared, and Android is unfortunately lacking in some of the tools that it has available. For instance, one option on a Linux box would be to attempt to zero out the whole disk and then try to re-create everything, but I don't know how you'd handle things like the MBR on Android (or if you even need to). It still may be a possibility, but I have no idea if it would work and your data would all be lost. Worst case you'd end up with a permanently unbootable drive if the MBR/partition table are zeroed out and you can't re-create them, but I guess that's kind of what you have now, in't it?

I think you are right T_T. Will look for workaround solution if any and after that will return to Amazon T_T

You might as well try it.
dd if=/dev/zero of=/dev/block/mmcblk0
Immediately flash a complete system image after this (complete with everything from xloader to media).
If this command says can't open '/dev/block/mmcblk0': I/O error, the chip is probably toast. Unless it's some silly partition abstraction layer caused by twrp or something, in which case you can try to flash a cwm image via fastboot...
If even that fails, you can try to pin-mod that flash -- resetting it. But opening your kindle is probably the last thing you'd want to do if you have the chance to return it, heheh...

I tried dd command already, it returned I/O error also.
Will consider pin-mod method, but could you pls advise more details hehe. It's better than return it to Amazon

This afternoon I tried to play with KF again. I put it to recovery and
Code:
~ # echo /dev/block/mmcblk0 > /sys/dev/devices/platform/usb_mass_storage/lun0/file
echo /dev/block/mmcblk0 > /sys/dev/devices/platform/usb_mass_storage/lun0/fi
le
And come back to Windows to check whether OS can detect the usb_mass_storage of KF.
And when I open HP USB Format Tool, it can detect the usb_mass_storage of KF as Amazon Kindle Fire 0001 (7376MB). I tried to format it but the error return that write protected device. Anyone can explain me what is the problem of my eMMC ?
And when using with GParted
Device information : Amazon Kindle
Size: 7.2 GiB
Path: /dev/sdb
Partition Table: msdos
Head 255
Sectors/track 63
Cylinders 940
Total sector 15106048
Sector size 512
Click to expand...
Click to collapse

what about mk2fs command via adb shell?
my mke2fs /dev/block/mmcblk0p11 brings to mk2fs: short write - what it could be?

Did you fix it yet? If fix, please intro for me. I have problem the same.

a similar problem, someone has already solved it?

[Q] How to recover files from formatted partition, find Userdata block

I just downgraded my Verizon G2 from 12b to 11a with this guide: (because of the push email issue)
http://forum.xda-developers.com/showthread.php?t=2448960
What I didn't realize was it would erase the data partition and now all my pictures are gone (idiot, I know)
Recovery software wont work because the G2 is only listed as a MTP device and they software doesn't see it.
I tried using this guide here, http://forum.xda-developers.com/showthread.php?t=1994705, which has instructions for the Galaxy Nexus which has the same MTP only issue. It basically dumps the partition as a .raw file to the computer through adb and then you can recover from that. I'm having a problem with this command "/system/bin/busybox nc -l -p 5555 -e /system/bin/busybox dd if=/dev/block/mmcblk0p12" mmcblk0p12 is the block on the galaxy nexus where the userdata was stored. I thought I had found the G2's block at "/dev/block/platform/msmsdcc.1/by-name/userdata" but the command doesn't work
Does anyone have any idea what the block path really is or an alternative to recover my lost stuff? Thanks.

Also tried to flash TWRP and then mount from there so software will see it but mounting does nothing

indigoataxia said:
I just downgraded my Verizon G2 from 12b to 11a with this guide: (because of the push email issue)
http://forum.xda-developers.com/showthread.php?t=2448960
What I didn't realize was it would erase the data partition and now all my pictures are gone (idiot, I know)
Recovery software wont work because the G2 is only listed as a MTP device and they software doesn't see it.
I tried using this guide here, http://forum.xda-developers.com/showthread.php?t=1994705, which has instructions for the Galaxy Nexus which has the same MTP only issue. It basically dumps the partition as a .raw file to the computer through adb and then you can recover from that. I'm having a problem with this command "/system/bin/busybox nc -l -p 5555 -e /system/bin/busybox dd if=/dev/block/mmcblk0p12" mmcblk0p12 is the block on the galaxy nexus where the userdata was stored. I thought I had found the G2's block at "/dev/block/platform/msmsdcc.1/by-name/userdata" but the command doesn't work
Does anyone have any idea what the block path really is or an alternative to recover my lost stuff? Thanks.
Click to expand...
Click to collapse
Hi there,
You should identify which partition is your data partition (link in step 1 of the guide). I did it, it shows me mmcblk0p35 (i have D802). But I had another problem.
I have rooted stock rom (D80210f-EUR-XX) with stock recovery. After modyfying /system/build.prop i got bootloop. the only thing I could do was to enter stock recovery and "Reset to defaults", so I did, but I lost all my data, messages, settings etc.
I tried to recover data partition using the same guide, but i'm stuck at step 6. It shows me speed of 0 B/s and creates file of 0B nomatter how long I wait.
Can anybody help please?

indigoataxia said:
I just downgraded my Verizon G2 from 12b to 11a with this guide: (because of the push email issue)
http://forum.xda-developers.com/showthread.php?t=2448960
What I didn't realize was it would erase the data partition and now all my pictures are gone (idiot, I know)
Recovery software wont work because the G2 is only listed as a MTP device and they software doesn't see it.
I tried using this guide here, http://forum.xda-developers.com/showthread.php?t=1994705, which has instructions for the Galaxy Nexus which has the same MTP only issue. It basically dumps the partition as a .raw file to the computer through adb and then you can recover from that. I'm having a problem with this command "/system/bin/busybox nc -l -p 5555 -e /system/bin/busybox dd if=/dev/block/mmcblk0p12" mmcblk0p12 is the block on the galaxy nexus where the userdata was stored. I thought I had found the G2's block at "/dev/block/platform/msmsdcc.1/by-name/userdata" but the command doesn't work
Does anyone have any idea what the block path really is or an alternative to recover my lost stuff? Thanks.
Click to expand...
Click to collapse
I'have your same issue (LG G2 D802)
I tried with:
/system/xbin/busybox ncat -l -p 5555 -e /system/xbin/busybox dd if=/dev/block/platform/msm_sdcc.1/mmcblk0p35
/system/xbin/busybox nc -l -p 5555 -e /system/xbin/busybox dd if=/dev/block/platform/msm_sdcc.1/mmcblk0p35
/system/xbin/busybox ncat -l -p 5555 -e /system/xbin/busybox dd if=/dev/block/mmcblk0p35
/system/xbin/busybox nc -l -p 5555 -e /system/xbin/busybox dd if=/dev/block/mmcblk0p35
but nothing happen!
did you solve? if yes, how???

tetractius said:
did you solve? if yes, how???
Click to expand...
Click to collapse
No, never was able to recover.

Sorry for reviving a dead thread.
Has anyone been able to successfully recover their deep formatted drive? Like the guide mentioned does? I'm currently trying, but no success..

SuperSU not autostarting

Hello, I hope someone has seen this before. I am not sure how I managed to break S7, as it was working, initially. I am running PE1 firmware, and factory reset before following the root guide, in the verizon guide forums. But now I have the problem where supersu will not auto start on bootup. I have to connect the usb cable to my pc, and use adb and manually start it. Root and Supersu are loaded. The only thing I have to run is:
C:\adb\adb.exe shell /system/etc/launch_daemonsu.sh
Then disconnect the usb cable, and all is well. Everything seems to be correct. What am I missing? I really do not want to start completely over. I have run the verizon debloat v2 and v8. I had to manually start supersu to flashfire the v8 script. Also in the same guide. All the root checker apps will see that you have root, but they say you are missing a super user app.
C:\adb\adb.exe shell /system/etc/launch_daemonsu.sh
mkdir: '/su': File exists
cp: /cache/stock_boot_*: No such file or directory
cp: bad '/cache/stock_boot_*': No such file or directory
/data/su.img: recovering journal
ext2fs_close2 : fs->write_bitmaps is null
ext2fs_close2 : normal operation, return 0
skipping journal recoverybecause INCOMPAT_RECOVER was clear.
check whether gdt & bitmap free count is vaild
/data/su.img: clean, 25/2048 files, 1399/8192 blocks
ext2fs_close2 : fs->write_bitmaps is null
ext2fs_close2 : normal operation, return 0
mount: No such file or directory
mount: No such file or directory
supolicy v2.74 (ndk:arm64-v8a) - Copyright (C) 2014-2016 - Chainfire
Patching policy ...
(Android M policy compatibility mode)
- Success
<Then disconnect the usb cable, supersu will work fine now.>
C:\adb>adb shell cat /system/etc/init.sec.boot.sh
#!/system/bin/sh
echo "init.sec.boot.sh: start" > /dev/kmsg
# start deferred initcalls
cat /proc/deferred_initcalls
## strace for system_server
#str=""
#while [ "$str" = "" ]; do
# str=`ps | grep system_server`
# sleep 0.1
#done
#
#pid=${str:10:4}
#echo "init.sec.boot.sh: strace -tt -T -o /data/log/strace.txt -p ${pid}" > /dev/kmsg
#strace -tt -T -o /data/log/strace.txt -p ${pid}
/system/etc/launch_daemonsu.sh
/data/s7startup/startupscript.sh
C:\adb>adb shell ls -la /system/etc/launch_daemonsu.sh
-rwx------ root root 4686 2016-07-03 13:32 launch_daemonsu.sh
Thanks again.

Did you push the supersu app to you device?

Is that to say uve got supersu installed but it won't start? If that's so I would check the box for start supersu during bootup. Unless im just completely missing your issue and im sorry if I am.

Nook Glowlight 3 - Stuck at infinite startup loading screen

So here's my issue. I transfered an epub file to my Nook Glowlight 3 using calibre, but when i ejected the device and started using the Nook. Not only did it not recognize my Epub file, the device slowed to a crawl! I restarted my Nook to see if this would fix the issue and now it's stuck at an infinite loading screen. I have tried hard resetting by holding the power button, but it keeps rebooting to an infinite loading screen. Could my device be possibly bricked? I think the issue might have been that I transfered a wrong type of epub file into the nook (the file was imported using calibre), which is incompatible with the nook software.
Any help would be appreciated, how do i get out of this infinite loading screen! I just want to be able to use my device again :/
Thanks.

No, it sounds like you did something else.
Do you have ADB working? Do:
Code:
# ps|grep u0 [color="red"]<- that's a zero[/color]
Do you see a bunch of processes?
That means that the Android subsystem is mostly running ok.
Is your launcher listed?
Code:
# getprop|grep bootanim
What does bn.bootanim.exit and service.bootanim.exit say?
You can disable the bootanimation entirely:
Code:
# mount -o rw,remount /system
# cd /system/bin
# mv bootanimation bootanimation.bak
# reboot
Can you start stuff?
Code:
# am start -n com.android.settings/.Settings

I'm actually having a *very* similar problem *only* when i have a 'real' launcher installed. In this case i'm using ReLaunchX, a launcher/file manager built for epaper devices. During boot the screen seems to be stuck on the boot animation. Here are some of the values from the commands you've reccomended to the OP:
Code:
[email protected]_6sl:/ $ ps|grep u0
u0_a2 2501 2080 321148 24152 ffffffff 00000000 S com.android.systemui
u0_a3 2517 2080 449348 87660 ffffffff 00000000 S com.nook.partner
u0_a0 2588 2080 319964 24852 ffffffff 00000000 S android.process.media
[email protected]_6sl:/ $ getprop|grep bootanim
[init.svc.bootanim]: [running]
[service.bootanim.exit]: [1]
[email protected]_6sl:/ $ am start -n com.android.settings/.Settings
Starting: Intent { cmp=com.android.settings/.Settings }
I have gone through the process of installing the launcher, rebooting, waiting for boot anim to finish (for up to 15-20 minutes), removing my launcher through adb, rebooting then re-installing the launcher. The launcher is the only thing that i've been chaning and it gets stuck on reboot *everytime* with the launcher installed.
P.S. I don't have root and i'm not interested in root, just the launcher and my safari queue app

If you have something taking over the Home intent, then the B&N stuff doesn't run.
The B&N stuff has the bootanimation termination hijacked.
Therefore the bootanimation will run for ever.
This will stop the bootanimation (just as a test).
Code:
$ setprop bn.bootanim.exit 1
You could write an app to do this.
You could modify the Glow3 bootanimation.
You could replace the Glow3 bootanimation with the Glow2 bootanimation.
You could delete the bootanimation.

You could write an app to do this.
You could modify the Glow3 bootanimation.
You could replace the Glow3 bootanimation with the Glow2 bootanimation.
You could delete the bootanimation.
Click to expand...
Click to collapse
Wouldn't these options require root?

Well, writing an app wouldn't need any special privilege, but it would be a round-about way to do it.
There's root & root & root. There's different degrees:
Disable all security on your normal boot image
Install a "superuser" app on your normal boot image to allow some apps to run as root
Set ro.secure to 0 to allow root ADB access
Set ro.debuggable to 1 to allow ADB to reconnect as root
Use a rooted recovery image to allow modification of your normal boot image
Load a rooted image over fastboot to allow modification of your normal boot image
Use the hardware root console that's already inside the Nook case
I wouldn't do #1, #2.
The other options don't really undermine security.
#7 requires opening the case.

Renate NST said:
[*]Set ro.secure to 0 to allow root ADB access
[*]Set ro.debuggable to 1 to allow ADB to reconnect as root
Click to expand...
Click to collapse
Hi Renate! I've tried to set these settings on a Glowlight 3, but as they don't seem to have effect might they require a pre-existing root to run them?
Code:
DESKTOP:~$ adb shell
[email protected]_6sl:/ $ setprop ro.secure 0
[email protected]_6sl:/ $ setprop ro.debuggable 1
[email protected]_6sl:/ $ mount -o rw,remount /system
mount: Operation not permitted
255|[email protected]_6sl:/ $
And just for good measure:
Code:
DESKTOP:~$ adb shell
[email protected]_6sl:/ $ su
/system/bin/sh: su: not found
Thank you for pointing me in any direction.

dotancohen said:
I've tried to set these settings on a Glowlight 3, but as they don't seem to have effect...
Click to expand...
Click to collapse
All the "ro.whatever" properties are read-only. They can only be set once.
They are usually already set in init.rc, default.prop or /system/build.prop.
ro.secure is in default.prop which is packed in the ramdisk of the image.
To modify it manually (vs. whatever scripts some people use), you need to pull the image, extract default.prop, text edit it, replace it, push the image.

Renate NST said:
To modify it manually (vs. whatever scripts some people use), you need to pull the image, extract default.prop, text edit it, replace it, push the image.
Click to expand...
Click to collapse
Thank you, I will try to pull and push the files directly. Have a great week!

Clock always wrong after every reboot

Hi,
I'm really desperate atm! No matter what i've done or tried, the clock always switches back to oct.2019 06:XX after every reboot with every new rom (e.g. 20g or 20h).
I can correct that in system settings and it stays correct as long as I do not reboot.
One reboot later: Octobre'19 about 06 o'clock.
I'm on stock 20c now. Everytime I restore an old backup clock is correct and stays correct even after reboot. I then tried to dirty flash 20h => wrong clock after reboot => clock stays wrong. I always need to restore everything with TWRP. Only then, clock stays correct. If I just restore "system" from backup for example => wrong clock...
I think it has something to do with the rtc:
Code:
# adb shell su -c hwclock
Fri Oct 19 06:40:09 2018 0.000000 seconds
# adb shell su -c date
Wed Mar 6 10:38:59 CET 2019
Tried to fix that, but even root has no write permission to the rtc:
Code:
# adb shell su -c hwclock -w
hwclock: ioctl 4024700a: Invalid argument
Anyone could help, please?? This drives me crazy!
I am not able to update my rom. Do not want to stay with 20c forever... just cause of the clock!