Checksum Mismatch! Could not connect to windbg Segmentation fault

[visce]

Work environment

Questions	Answers
OS/arch/bits (mandatory)	4.17.9-1-ARCH arm 64
File format of the file you reverse (mandatory)	sudo r2 -a x86 -b 32 -D windbg windbg:///tmp/win10.pipe
Architecture/bits of the file (mandatory)	PPC, x86/32, x86/64 etc.
r2 -v full output, not truncated (mandatory)	radare2 -v radare2 2.8.0-git 18851 @ linux-x86-64 git.2.7.0-164-g79bfcd8f6 commit: 79bfcd8 build: 2018-07-27__11:23:10

Expected behavior

I expected a prompt for debugging. This is also a fresh install.

Actual behavior

$ sudo r2 -a x86 -b 32 -D windbg windbg:///tmp/win10.pipe
[sudo] password for *****
Opened pipe /tmp/win10.pipe with fd 0x6
Checksum mismatch!
Could not connect to windbg
Segmentation fault

Steps to reproduce the behavior

Start up virtual machine, and use sudo r2 -a x86 -b 32 -D windbg windbg:///tmp/win10.pipe to connect to the pipe. Debugging is enabled.

Additional Logs, screenshots, source-code, configuration dump, ...

[radare]

Wheres the crashlog? You can just use -d windbg://... no need to use -D plugname plugname://path

…

On 27 Jul 2018, at 18:02, visce ***@***.***> wrote: Work environment Questions Answers OS/arch/bits (mandatory) 4.17.9-1-ARCH arm 64 File format of the file you reverse (mandatory) sudo r2 -a x86 -b 32 -D windbg windbg:///tmp/win10.pipe Architecture/bits of the file (mandatory) PPC, x86/32, x86/64 etc. r2 -v full output, not truncated (mandatory) radare2 -v radare2 2.8.0-git 18851 @ linux-x86-64 git.2.7.0-164-g79bfcd8f6 commit: 79bfcd8 build: 2018-07-27__11:23:10 Expected behavior I expected a prompt for debugging. This is also a fresh install. Actual behavior $ sudo r2 -a x86 -b 32 -D windbg windbg:///tmp/win10.pipe [sudo] password for ***** Opened pipe /tmp/win10.pipe with fd 0x6 Checksum mismatch! Could not connect to windbg Segmentation fault Steps to reproduce the behavior Start up virtual machine, and use sudo r2 -a x86 -b 32 -D windbg windbg:///tmp/win10.pipe to connect to the pipe. Debugging is enabled. Additional Logs, screenshots, source-code, configuration dump, ... — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

[visce]

What's the command to check the crashlog?

[radare]

Compile r2 with asan or run it with valgrind or gdb to show the backtrace, regstateetc

…

On 27 Jul 2018, at 20:20, visce ***@***.***> wrote: What's the command to check the crashlog? — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.

[visce]

I left the $ sudo r2 -a x86 -b 32 -d windbg:///tmp/win10.pipe
going for about an hour, and a half. It still didn't finish to the error whilst being debugged. Here is the only bt I could get before the error. Thanks for the help!
(gdb) bt
#0 0x00007fba5f550197 in select () from /usr/lib/libc.so.6
#1 0x00007fba61e4ee2f in iob_pipe_read (p=0x6, buf=0x556af0cf5041 "",
count=49103, timeout=100) at iob_pipe.c:84
#2 0x00007fba61e4c36f in iob_read (fp=0x6, buf=0x556af0cf2010 "",
buf_len=61440) at transport.c:87
#3 0x00007fba61e4f243 in kd_read_packet (fp=0x6, p=0x7ffc6e775590) at kd.c:97
#4 0x00007fba61e4c894 in windbg_wait_packet (ctx=0x556af0cd9a90, type=6,
p=0x0) at windbg.c:217
#5 0x00007fba61e4db7e in windbg_sync (ctx=0x556af0cd9a90) at windbg.c:655
#6 0x00007fba62ee6224 in r_debug_windbg_attach (dbg=0x556af0c68220, pid=-6)
at p/debug_windbg.c:116
#7 0x00007fba62ed3e65 in r_debug_attach (dbg=0x556af0c68220, pid=-6)
at debug.c:429
#8 0x00007fba63d6e703 in cmd_debug_pid (core=0x556aeeb71580 ,
input=0x556af0cd80d1 "pa -6") at cmd_debug.c:1050
#9 0x00007fba63d874b3 in cmd_debug (data=0x556aeeb71580 ,
input=0x556af0cd80d1 "pa -6") at cmd_debug.c:4665
#10 0x00007fba63e43ee1 in r_cmd_call (cmd=0x556af0c37fd0,
input=0x556af0cd80d0 "dpa -6") at cmd_api.c:237
#11 0x00007fba63dfa336 in r_core_cmd_subst_i (core=0x556aeeb71580 ,
cmd=0x556af0cd80d0 "dpa -6", colon=0x0, tmpseek=0x7ffc6e776c16)
at cmd.c:2911
#12 0x00007fba63df6e96 in r_core_cmd_subst (core=0x556aeeb71580 ,
cmd=0x556af0cd80d0 "dpa -6") at cmd.c:1940
#13 0x00007fba63dfca9b in r_core_cmd (core=0x556aeeb71580 ,
cstr=0x7ffc6e776d20 "dpa -6", log=0) at cmd.c:3615
#14 0x00007fba63dfd1a3 in r_core_cmdf (user=0x556aeeb71580 ,
fmt=0x7fba63ecb3f7 "dpa %d") at cmd.c:3773
#15 0x00007fba63e19bc9 in r_core_setup_debugger (r=0x556aeeb71580 ,
debugbackend=0x556af0cbc360 "windbg", attach=true) at cio.c:24
#16 0x0000556aee96dd11 in main (argc=7, argv=0x7ffc6e778238,
envp=0x7ffc6e778278) at radare2.c:1216

[radare]

This seems more like a sigpipe than a sigsegv. Run cont and wait for the segfault to happen or build with sys/asan.sh or sys/meson.sh —asan

…

On 27 Jul 2018, at 23:33, visce ***@***.***> wrote: I left the $ sudo r2 -a x86 -b 32 -d windbg:///tmp/win10.pipe going for about an hour, and a half. It still didn't finish to the error whilst being debugged. Here is the only bt I could get before the error. Thanks for the help! (gdb) bt #0 0x00007fba5f550197 in select () from /usr/lib/libc.so.6 #1 0x00007fba61e4ee2f in iob_pipe_read (p=0x6, buf=0x556af0cf5041 "", count=49103, timeout=100) at iob_pipe.c:84 #2 0x00007fba61e4c36f in iob_read (fp=0x6, buf=0x556af0cf2010 "", buf_len=61440) at transport.c:87 #3 0x00007fba61e4f243 in kd_read_packet (fp=0x6, p=0x7ffc6e775590) at kd.c:97 #4 0x00007fba61e4c894 in windbg_wait_packet (ctx=0x556af0cd9a90, type=6, p=0x0) at windbg.c:217 #5 0x00007fba61e4db7e in windbg_sync (ctx=0x556af0cd9a90) at windbg.c:655 #6 0x00007fba62ee6224 in r_debug_windbg_attach (dbg=0x556af0c68220, pid=-6) at p/debug_windbg.c:116 #7 0x00007fba62ed3e65 in r_debug_attach (dbg=0x556af0c68220, pid=-6) at debug.c:429 #8 0x00007fba63d6e703 in cmd_debug_pid (core=0x556aeeb71580 , input=0x556af0cd80d1 "pa -6") at cmd_debug.c:1050 #9 0x00007fba63d874b3 in cmd_debug (data=0x556aeeb71580 , input=0x556af0cd80d1 "pa -6") at cmd_debug.c:4665 #10 0x00007fba63e43ee1 in r_cmd_call (cmd=0x556af0c37fd0, input=0x556af0cd80d0 "dpa -6") at cmd_api.c:237 #11 0x00007fba63dfa336 in r_core_cmd_subst_i (core=0x556aeeb71580 , cmd=0x556af0cd80d0 "dpa -6", colon=0x0, tmpseek=0x7ffc6e776c16) at cmd.c:2911 #12 0x00007fba63df6e96 in r_core_cmd_subst (core=0x556aeeb71580 , cmd=0x556af0cd80d0 "dpa -6") at cmd.c:1940 #13 0x00007fba63dfca9b in r_core_cmd (core=0x556aeeb71580 , cstr=0x7ffc6e776d20 "dpa -6", log=0) at cmd.c:3615 #14 0x00007fba63dfd1a3 in r_core_cmdf (user=0x556aeeb71580 , fmt=0x7fba63ecb3f7 "dpa %d") at cmd.c:3773 #15 0x00007fba63e19bc9 in r_core_setup_debugger (r=0x556aeeb71580 , debugbackend=0x556af0cbc360 "windbg", attach=true) at cio.c:24 #16 0x0000556aee96dd11 in main (argc=7, argv=0x7ffc6e778238, envp=0x7ffc6e778278) at radare2.c:1216 — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.

[visce]

What would be the process of debugging with meson? I've never used it to debug before. Unless you're talking about building radare2 again with meson, then grabbing the debug build, and posting it. Just need a little more elucidation on the issue, thanks!

[radare]

Meson is a build system, alternative to make. Not a debugger. Just build with asan like i said and paste the crashlog after connecting to the windbg target

…

On 28 Jul 2018, at 01:02, visce ***@***.***> wrote: What would be the process of debugging with meson? I've never used it to debug before. Unless you're talking about building radare2 again with meson, then grabbing the debug build, and posting it. Just need a little more elucidation on the issue, thanks! — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.

[XVilka]

@visce ping?

[radare]

PING

[visce]

Thanks for your help I've gotten the issue resolved now. I had to install vmware-workstation instead of virtualbox, and it works.

[yssource]

It is the problem about virtualbox, isn't it?
The same issue happened to me also, and I ran it with virtualbox.

My guest machine is winXP 32bit.

[Maijin]

@yssource This has been closed on August 3rd, open a new issue please.