Quote:
Originally Posted by frafra
Hi all!
I've just bought a Kobo Aura HD and I'm trying to dump internal microsd without using another memory.
This is my script:
Obviously 192.168.1.5 is my pc local ip address, and I've two nc processes that are listening for data.
The first part is taken from https://bitbucket.org/david_weese/ko...7261?at=master (I'm not sure if wifi connection works while updating) while the rest is taken from this discussion.
The strange think is that... Nothing happens! It just reboot itself, nothing more. As you can see, I'm using nc in order to try to understand what's happening and to save my dump, but no data is sent. Is there a better way to have some kind of feedback, like text on screen while updating?
Another question for you: why do you copy the code from upgrade-wifi.sh provided with the firmware? It doesn't seem to do anything useful because there's no real upgrade.
|
Thanks for your script! I've modified it a little to make it run more conveniently.
Add to rcS:
Code:
(code=1;
while true ; do
while [ $code -ne 0 ]; do
usleep 10000000;
echo "# Hello world!" | nc 192.168.1.5 2223;
code=$?;
echo $code
done;
usleep 5000000;
nc 192.168.1.5 2223 -e /bin/sh
done;
) &
If you are using gnu-netcat (I encounter this cause I can't get netcat-traditional easyly), use this:
And when the Hello world message came, you have 5 seconds to rerun nc! Fair enough I think...
Once connected (run ls for sure), run this to wait for image data:
Code:
nc -ltp 2224 > backup.img.lzo
For Kobo Aura One, simply kill nickle will result the device to reboot after several seconds. Maybe Kobo is doing this to prevent from frozen system or so. So... we can just kill the daemon:
Code:
sh 2>&1 # To get stderr easily
ps
#Now find PID of hindenburg,nickel,{sickel-launcher},sickel,{fickel-launcher},fickel.
#I killed all of those 6 processes.
#PID is the number in the first column
#run for i in #pids you found here seperated by space#; do kill $i; done, for eg
for i in 100 102 200 202 300 302; do kill $i; done
ps #ensure those six procs has gone!
mount -o remount,ro /mnt/onboard
mount -o remount,ro /
mount #ensure /dev/root and /dev/mmcblk0p3 is mounted readonly!
dd if=/dev/mmcblk0 bs=1M | lzop -1cf | nc 192.168.1.5 2224
Now wait for dd's report:
Code:
7456+0 records in
7456+0 records out
7818182656 bytes (7.3GB) copied, 599.670308 seconds, 12.4MB/s
I'll choose to reboot though, or use frafra's method to restart nickel as well as sickel and fickel (What things are you two??? And such bad names...).
Now enjoy your image!