I am newbie to UNIX programs. I have encountered a situation wherein I have to issue commands to gdb from my C program. I have a C program which invokes another C program by forking a new child process. I need to debug this child C program and hence, I used system command to call gdb process on this C program. But I get a gdb prompt which I do not want. I want to issue commands to the gdb from my parent C program. Is there a way to issue commands to gdb from a C program ?
Please reply.
Thanks a lot.
Esash
If you need to debug the child process, you don't necessarily need to invoke the child with GDB when you fork+exec. As long as you have the PID of the child process, you can use the "attach" command in GDB to attach to the running child process. Basically, you would start GDB like:
$ gdb
(gdb) attach pid-of-child
In the above, replace pid-of-child with the PID of the child process, and there you go, you can debug the child process from interactive GDB, without the parent process needing to deal with GDB at all.
There are several ways to "drive" GDB programmatically.
If you just want to issue one command, e.g. to find out why the child crashed, you can do something like this:
gdb --batch -ex where /path/to/child <pid-of-child>
If there are more commands then you are willing to put on command line, you could write them to a temporary file, and ask gdb to execute them:
gdb --batch -x /path/to/commandfile /path/to/child <pid-of-child>
Neither of the above allows you to perform programmatic (if ... then do-something-in-gdb else do-something-else-in-gdb) control.
For that you may want to either exercise GDB's machine interface (MI), or use the embedded Python interpreter.
Theres also the follow on fork gdb option. This will attach to the child process immediately.
set follow-fork-mode mode
so,
set follow-fork-mode child
Related
Currently debugging a tcsh like remake, I used :
set follow-fork-mode child
to follow a child program after fork but reaching the execve system call GDB exit and I got the following message :
process 11217 is executing new program: /usr/bin/cat
zsh: suspended (tty output) gdb ./mysh
How can I prevent that ?
Thanks you in advance !
Looks like the child got suspended?
You can let the child run with:
set detach-on-fork on
This is the default. You can check with show detach-on-fork whether it's been turned off (perhaps via .gdbinit?).
To follow the exec'ed process, you can use:
set follow-exec-mode new
You can also switch between multiple processes using inferior. First get inferior numbers via:
info inferiors
Then to switch:
inferior <num>
I am trying to write a program , which does a fork and exec a child process and executes it in the back ground .
One approach I would see is to redirect the output to /dev/NULL file and come back to my main program . Any other ideas ?
After a process is started, shell has no more control on process file descriptors so you can not silence it by a shell command i.e. terminal has its stdin, stdout and stderr bound to the terminal and you cannot do anything about it without re-gaining control over that terminal.
There is a tool called retty how you can use it can be seen at this link retty this tool is used to attach processes running on terminals
Beside you can also use the built in disown command to disown the process which will prevent from sending a SIGHUP signal to the program when the shell exits
This link can be helpful Link to a similar problem
I am faced with a situation wherein I have process X executing a
a command ( say /bin/ls ). as soon as the process X executes the command ls
I want to put a breakpoint in a function in ls.
Is there any way to do this ?
An easy solution may be to wrap the binary in question (that is called by process X) in a small shell script that starts the process in a debug session and applies pre-configured breakpoints as well.
I can think of two ways to do it.
Simplest is to set follow-fork-mode child whenever new new client process is created GDB will debug the child. However with this mode you will not be able to debug the parent process any more.
In the child process (ls mentioned above) add some code to wait for a signal say SIGCONT at the very beginning. Whenever child process is created attach GDB (new GDB instance) to it with its PID of child process, issue the singnal SIGCONT to continue.
You can use catch exec [1] to stop on exec calls:
(gdb) catch exec
Catchpoint 1 (exec)
(gdb) r
Starting program: /tmp/a.out
process 7544 is executing new program: /bin/ls
Catchpoint 1 (exec'd /bin/ls), 0x00007ffff7ddfaf0 in _start () from /lib64/ld-linux-x86-64.so.2
Then you can do whatever you want with the new process. See also the link in the comment by dbrank0 for various fork-related options.
Ive got my program in C, 6 source files, and the aim is to copy those files to any other Linux OS computer, and (probably compile, im newbie, so not sure what is needed here) run this program in background. Something like:
user#laptop:~$ program
Program is running in a background. In order to stop Program, type
XXX.
Any tips on this?
Thanks in advance!
Put a daemon(0,0); call in your C program.
stopping it is a bit trickier, I suppose there is only one copy of the program running. Put the program's PID in a file, write another utility (XXX) which reads the PID from the file and kills it.
Important: daemon forks, get the PID of the program after calling daemon.
But maybe you are too newby and just want to execute your program with program& and later kill it.
I completely missunderstood the question. You need shell scripting for this.
For file copying you can use scp. Execute command on the other host with ssh. It should be something like (not tested):
pid=`ssh user#host "make >/dev/null 2>&1; nohup ./program; echo $!`
later you can stop it with
ssh user#host "kill $pid"
First, you should fork().
In parent, you should just exit, in child process - you should handle SIGHUP signal.
In such way - you have daemon.
In general i see the process's pid which is running in the background and start dbx on that process using the command dbx -a <pid>
similarly how could i do it using gdb?
In addition to the previous you can directly use
gdb -p <pid>
There are two ways.
From the command line, include the pid as an argument after the executable name:
gdb /path/to/prog PID
From within gdb, you can use the attach command:
gdb /path/to/prog
gdb> attach PID
While the specifying on the command line is more concise, there is a slight risk that if you have a core file that has a name that is the same as the pid (i.e. for pid 2345, the core file would have to be named "2345") then gdb will open the core file. Admittedly, the chance of this happening is minuscule.
From the gdb man page:
You can, instead, specify a process ID as a second argument, if you want to debug a running process:
gdb program 1234