Proxmark3 community

Research, development and trades concerning the powerful Proxmark3 device.

Remember; sharing is caring. Bring something back to the community.


"Learn the tools of the trade the hard way." +Fravia

You are not logged in.

Announcement

Time changes and with it the technology
Proxmark3 @ discord

Users of this forum, please be aware that information stored on this site is not private.

#1 2017-05-11 17:22:02

WitchDoctor
Contributor
Registered: 2017-05-10
Posts: 7

Bootloop?

I have this interesting situation. Maybe someone has some input. Until the 22nd, my options for OS are kali or ubuntu live. My macbook is back home with my wife and I'm in the states for a funeral. I saw that someone had a similar issue (kind of) and they were able to fix it using Ubuntu so I have created the live usb for it. I've been working on this for a couple of days between projects and I have read every piece of literature on the device I can find. I'll just jump in here.

I followed the commands to update the bootrom. No problem. Then the firmware. If I plug the device in (USB3.0 and USB2.0) without pressing the button, nothing happens and I see only the "new high speed device..." Output below. When I hold the button, I can see it in dmesg. I can't get the bootrom (hid-flasher) flasher to recognize the device in any situation. The client flasher goes through the motions of updating the firmware, but nothing changes.



Here is a list of all of the commands and their outputs from the device:

When the device is connected to my computer via usb, there is no connection made as verified in the following code:

root@kali-system:~/proxmark3/client# ./proxmark3 /dev/ttyACM0
ERROR: invalid serial port
proxmark3> quit
Segmentation fault
root@kali-system:~/proxmark3/client#


Without any user interaction, my device does not establish a connection to the computer and then constantly resets:

root@kali-system:~/proxmark3/client# dmesg|grep -i usb
[ 2685.964270] usb 2-1: new full-speed USB device number 10 using xhci_hcd
[ 2702.332046] usb 2-1: new full-speed USB device number 11 using xhci_hcd
[ 2718.703709] usb 2-1: new full-speed USB device number 12 using xhci_hcd

However, when I hold the "button" on the device and plug it into the computer, a connection is made on /dev/ttyACM0. The connection persists only so long as I am holding the button. When the button is released, the device continues to "boot loop" and no connection is made:


root@kali-system:~/proxmark3/client# dmesg|grep -i usb
[ 3043.503064] usb 2-1: device descriptor read/64, error -110
[ 3043.752586] usb 2-1: New USB device found, idVendor=2d2d, idProduct=504d
[ 3043.752592] usb 2-1: New USB device strings: Mfr=1, Product=0, SerialNumber=0
VID_20170505_102650.mp4
[ 3043.752595] usb 2-1: Manufacturer: proxmark.org
[ 3043.753416] cdc_acm 2-1:1.0: ttyACM0: USB ACM device
[ 3062.698751] usb 2-1: new full-speed USB device number 27 using xhci_hcd
[ 3079.066422] usb 2-1: new full-speed USB device number 28 using xhci_hcd
[ 3095.434218] usb 2-1: new full-speed USB device number 29 using xhci_hcd
root@kali-system:~/proxmark3/client#


Now, when I hold the button and run the client mount commands, the script will launch, but none of the commands respond. Either something becomes unstable and the script collapses with a segmentation fault or I receive the same segmentation fault as the program exits:

root@kali-system:~/proxmark3/client# ./proxmark3 /dev/ttyACM0
proxmark3> hw tune

Measuring antenna characteristics, please wait...........
No response from Proxmark. Aborting...         
proxmark3> quit
Segmentation fault
root@kali-system:~/proxmark3/client#

Offline

#2 2017-05-11 19:21:20

iceman
Administrator
Registered: 2013-04-25
Posts: 9,497
Website

Re: Bootloop?

Try a usb2.0 port.

Offline

#3 2017-05-11 19:56:57

WitchDoctor
Contributor
Registered: 2017-05-10
Posts: 7

Re: Bootloop?

It was working fine before I flashed the new firmware. I've tried both 2.0 and 3.0 ports on this machine.

Thank you

Offline

#4 2017-05-11 21:07:33

WitchDoctor
Contributor
Registered: 2017-05-10
Posts: 7

Re: Bootloop?

Got 'em, coach. maybe not the appropriate way to do it, but I was tired of looking insane. From the client folder, I ran:

./flasher /dev/ttyACM0 -b ../bootrom/obj/bootrom.elf

That got weird, but it began to appear on my dmesg without the button being depressed.

Then I went to the hid-flasher folder and ran the normal bootrom codes:

./flasher -b ../../bootrom/obj/bootrom.elf

Kind of weird output (your bootloader doesn't understand START-NEW or something like that).

cd ../

./flasher /dev/ttyACM0 ../armsrc/obj/fullimage.elf

Success. Kind of.

Unstable as hell, but it's functioning and reading/writing/simulating.

Offline

#5 2017-05-12 07:10:15

iceman
Administrator
Registered: 2013-04-25
Posts: 9,497
Website

Re: Bootloop?

Glad to hear that your device works.

you don't need / shouldn't use the hid-flasher if you are on a new bootrom.

you only need to flash fullimage 99% of the time. 

So,  as a reference to futher readers,

DON'T use hid-flasher.

Offline

#6 2017-05-12 08:16:28

ntk
Contributor
Registered: 2015-05-24
Posts: 701

Re: Bootloop?

I second iceman. DON'T use hid-flasher.

very surprised s.o could you hid-flasher when the PM3 is already a turned CDC USB device.

the segmentation fault on linux system is most likely something's gone wrong during driver setup. In opposite to window OS where an re-install can take time re install on linux system could be 30min, ubuntu pre R14 many linux users chose this way and sort out problem without reporting that back. even with the same USB\CD each linux installation could result different

Offline

#7 2017-05-18 03:44:40

WitchDoctor
Contributor
Registered: 2017-05-10
Posts: 7

Re: Bootloop?

I think the bootrom flashing guidelines are a little unclear then. Everything I have read told me to use hid-flasher for the bootrom. Here I was thinking I'm moving out of the script kiddie realm, but I just did exactly what I was supposed to do (and then ruined that by following the Kali instructions on the wiki). Haha. I had some problems during my test in Miami and abandoned the proxmark as a viable option for the physical security portion, but I was still able to get in. We go again in six months, so hopefully I'll have it ironed out by then. Excellent tool, I just wish I had come back to the forum before using something in the field before it was ready. Anyway, I appreciate the input. I'm going to reflash it tomorrow and see if that produces a more stable result. Thanks again for the help, guys!

Offline

#8 2017-05-18 08:07:27

iceman
Administrator
Registered: 2013-04-25
Posts: 9,497
Website

Re: Bootloop?

Feel free to update the wiki.  It is always room for improvements in the instructions.

Many ppl who follow the instructions will manage to make a working dev env setup. Those who fails usually hasn't read it and did not understood what they read. Not sure if this an indication of unclear instructions or if says something about the person in general.
While we can re-write instructions...

Offline

#9 2017-05-18 13:20:57

ntk
Contributor
Registered: 2015-05-24
Posts: 701

Re: Bootloop?

The problem many new users have faced with no-operative PM3 is not caused by Proxmark environment type of communication HID/CDC USB device type or setup instructions; but by the specifics of system you use, particular culprit is that "raised standard of integrity check" compared to XP window version; in this case window2000, window7, window8 window10.  the integrity check of those OS, and the many different methods to help user conveniently overcome the integrity check, mix together can cause mayhem

The windows user has a lot of problems the Mac and linux would not care about Just think only about the virus problem Mac Os and Linux do not care much about, but on window OS how many different types of antivirus programs you could install to protect your machine,  java-weakness, flash weakness, adware, bad ware, malware, viren, worm,trojan, rookit, backdoor, NSA, GOV stuffs etc and indirectly you block yourself easily making un-informed changes to system too.

The key works are:
1/ window driver installation requirement
2/ unsigned/signed driver
3/ test mode
4/ you must know in details what to do with your system

I am not confident in English to write up solution, but just stay 1 minute calm, patiently spread the cards and logically think from those starting points and you see where lay problems and solutions.

Last edited by ntk (2017-05-18 21:02:56)

Offline

#10 2017-05-22 20:15:47

WitchDoctor
Contributor
Registered: 2017-05-10
Posts: 7

Re: Bootloop?

Yeah I wasn't complaining about the instructions or like whining about something. I know that people working their way through the learning curve isn't something most devs tolerate well, but I'm not a 14 year-old in my mom's basement. The instructions are unclear. I haven't updated them, but it's not for a lack of trying. I'm finally back in the Ukraine after several days of travel. I hope I didn't come across to any of you as having some attitude about it. Text makes it difficult to convey tone. We want to work together to create a community of like-minded contributors, but I don't think passive insults are the way to do that.

Offline

#11 2017-05-22 20:38:51

iceman
Administrator
Registered: 2013-04-25
Posts: 9,497
Website

Re: Bootloop?

as mentioned before, feel free to update the wiki-instructions. 

I will be looking forward to your contributions to the community.

Offline

#12 2017-05-22 20:57:47

akileos
Contributor
Registered: 2017-05-17
Posts: 23

Re: Bootloop?

WitchDoctor wrote:

Text makes it difficult to convey tone. We want to work together to create a community of like-minded contributors, but I don't think passive insults are the way to do that.

Haven't seen any passive insults here, no worries. I'm new aswell here and to my understanding all the info is already here if you know what to look for. You won't get prebaked answers unless you ask for something noone asked before.
Carry on and please update the docs with the bits you felt missing wink

Offline

#13 2017-05-23 10:15:27

WitchDoctor
Contributor
Registered: 2017-05-10
Posts: 7

Re: Bootloop?

Changed. Anyone with anything to add/debate, please let me know. I'm new to RFID and I look forward to learning and contributing more. I apologize for being a bit abrasive as well. I let the jet lag get the better of me for a moment and I don't feel it was necessary to snap at everyone.

Edited to follow up with a statement- I'm still producing a segmentation fault upon exiting the client module via CLI. The device seems to be functioning without any problem since flashing through the client module. I saw someone mentioned this is likely an issue with the driver setup from the beginning. I'm going to leave it alone for now. Thanks for the input, guys. I appreciate it.

Last edited by WitchDoctor (2017-05-23 10:19:39)

Offline

#14 2017-05-23 10:55:14

ntk
Contributor
Registered: 2015-05-24
Posts: 701

Re: Bootloop?

Case may be closed already, but pls allow my last input, could be useful for new users.

I see now that you, @witchdoctor, use Kali or Ubuntu, running live session. with persistence or not? To compile Proxmark make sure you have enough room, >20GB,  to install the extra required SW components, the project code and temporary files. Linux can react strangely when it has to be squeezed in a confined space. U could has problem running on USB stick which ppl running HD-installed linux, or on VM box won't see. Even you know when you chose different options on your startup menu, your system behaves very differently too. I would be careful to use the live system, it is only for tasting a bit of how linux would look alike. I don't trust it, and can only say it is too inconvenient. 

In future I propose people posting problem with sudden issue on driver setup flashing of proxmark3 board etc... they should  also giving info on system type configuration, it is easier to compare to follow and giving help, a fix in window not conform to a fix in linux (again ubuntu, mint, kde, kali and all those options either live or persistence or permanent installation type etc.) or Mac OS 

I read all your post again and see this "When I hold the button, I can see it in dmesg. I can't get the bootroom (hid-flasher) flasher to recognize the device in any situation."

this ""When I hold the button, I can see it..." can be fix, contact JAYC and work with him. JAYC has the similar fault in window 10 env., you in is linux, the principle to resolve this fault is similar.

not sure about segmentation fault, you need give info on your system intel/AMD/NVIDIA to prove your installation is in general error free when run extensive tests like furemark speed test endurance etc, a fault in graphic driver can cause related problem.

"Segmentation faults can arise from similar conditions. A buffer overflow, such as trying to reach outside the bounds of an array, can cause a segfault, or trying to access memory that has not been allocated or has been deleted. Trying to write to memory that is read-only can also cause a memory error...." or it could be that you faces segmentation fault similar case I report here
  Crash in graph tool in recent SW version 

Last edited by ntk (2017-05-23 16:22:53)

Offline

#15 2017-05-28 11:37:37

ntk
Contributor
Registered: 2015-05-24
Posts: 701

Re: Bootloop?

@WitchDoctor
Not sure it is the right place where can we discuss about the modification on wiki.
I believe the change  you made on wiki is not fully correct.
the client/hid-flasher executable will not be created, when you run make in /pm3/client directory.
As the directory name expressed, the two flasher types are different.

If your device, very rarely, somehow comes up as a HID device, you would need to rebuild and to use the /client/hid-flasher ( I always rename it hid-flasher.exe or flasher OLD.exe for not to mixing that with the for CDC device flasher. This hid-flasher is never created when you run make i n /pm3/client directory. when thsi case happened and when you especially need the hid-flasher or the flasher OLD.exe you must go in there, run the make, and only there run your flasher with the -b option, and in this case the ../../ meaning is correct.

what you change is assuming that the PM3 device is already a CDC, here dmesg will have to confirm first. you would, only and only build the new flasher. In this case your instruction  ./flasher /dev/ttyACM0 -b ../bootrom/obj/bootrom.elf make sense (but not necessary to run on new CDC already enable device each time you change SW version)

your are right only here in this case, but then the following command "cd ../" in the modification:"
Release the button.  And go back to the main Proxmark3 directory.
- cd ../..
+ cd ../"
is unnecessary, because you can stay here to run your firmware update and should stay in the client directory to test out your sw updated PM3. why doing
"
  cd client
  ./flasher /dev/ttyACM0 ../armsrc/obj/fullimage.elf
  cd ..
"? jumping out, later then jumping back in, it makes me dizzy, doctor.

Not want to be too critic on your change, just want to show how difficult it is to explain and make right for every OS, every case, every user, now you have sensed the gist of it too, am I right?

Nice to see PM3 works for you.

Last edited by ntk (2017-05-28 12:40:56)

Offline

Board footer

Powered by FluxBB