Why does waitpid hang and not reap all child processes? - c

I'm attempting to simulate a shell processing foreground commands but I'm running into trouble when it comes to reaping child processes. I spawn multiple processes using fork in the following code.
while (currCmdListNode != list_end(&currentPipeline->commands)) {
if (fork() == 0)
{
printf("Child created with pid %d \n", getpid());
currentCmd = list_entry(currCmdListNode, struct esh_command, elem);
execvp(currentCmd->argv[0], currentCmd->argv);
}
currCmdListNode = list_next(currCmdListNode);
}
Then, I try to reap all these processes from the parent process using the following code.
while((pid = waitpid(-1, &status, WUNTRACED)) > 0)
{
if (WIFEXITED(status))
{
printf("child %d terminated normally with exit status=%d\n", pid, WEXITSTATUS(status));
}
else
{
printf("child %d terminated abnormally\n", pid);
}
}
Let's say I run this code so that two child processes spawn from my initial fork. My reap functionality will return saying it reaped the first child. However, it will hang once it hits waitpid the second time, and I will never get a message saying it reaped the second child. If I spawn four child processes, then maybe two will get reaped, then it will hang. What is going on here? Do I have to do some sort of blocking of the SIGCHLD signal? If so, where should I be blocking it?

Related

Execute background process and check status from parent

I'm currently trying to:
Execute a background process with a fork() call and execvp() inside the children.
If the call to execvp() was successful, adding the PID of the child process to a list.
If the call to execvp() returned -1 (error), do not add the PID to the list.
My problem is: if execvp() returns -1 (for instance, if the executable isn't found) I can't "communicate" with the parent proccess since I'm inside the child process in the fork(). And if I use waitpid() to wait for the child, I'm not running a background process (since I lose the concurrency).
I tried using WNOHANG and checking *status with (WEXITSTATUS(status) == 1) but it doesn't work (since the parent process executes "instantly" and doesn't know if the child has exited yet, or that's what I think)
My current code:
void background(char *vector[]) {
pid_t childp;
int status;
childp=fork();
if (childp==-1){
perror("Error using fork \n");
return;
}
if (childp==0) { //Child process
if (execvp(vector[0],vector)==-1) {
perror("exec");
exit(1);
}
}
waitpid(-1, &status, WNOHANG);
if(WIFEXITED(status)) {
//Child exited
if (WEXITSTATUS(status) == 1) {
//Child exited with error
return;
}
}
if (insertProcess(childp,vector)==-1)
perror("Full list");
else
updateList();
}
The only solution I've found at the moment is inserting a sleep(1) call before waitpid (Source: https://stackoverflow.com/a/26617152/1339354 ), but this looks more like a hack than something well done. Any other ideas?
If you are using waitpid() with WNOHANG, be sure to especify the child process. If you don't, waitpid() returns status of the last child dead.
I'm not sure if you are trying to save on a list all process that was success. If you are trying it, why not instert always and then delete the process when they die? Control the signal SIGCHLD. In that, use the waitpid for look what process died and delete it in the list. Just deactivate the signal SIGCHLD when you create the process, insert the process on the list and then active again the signal.

Linux Fork: pid reuse

I wrote the following program to understand the way fork works when called without wait() or waitpid().
int main()
{
pid_t childpid;
int retval = 0;
int i;
while(1){
//usleep(1);
childpid = fork();
if (childpid >= 0)
{
i++;
if (childpid == 0)
{
exit(retval);
}
else
{
//printf("childpid is %d\n", childpid);
}
}
else
{
printf("total no. of processes created = %d\n", i);
perror("fork");
exit(0);
}
}
}
Here's the output I get->
total no. of processes created = 64901
fork: Cannot allocate memory
I expected the program to go on as I'm exiting the child process instantly and fork() should reuse the pids after pid > pid_max. Why doesn't this happen?
The exited child processes do remain in the process table as zombies. Zombie processes exist until their parent calls wait or waitpid to obtain their exit status. Also, the corresponding process id is kept, to prevent other newly created processes of duplicating it.
In your case, the process table becomes too large and the system rejects the creation of new processes.
Forking processes and then not retrieving their exit status can be regarded as a resource leak. When the parent exits, they will be adopted by the init process and then reaped, but if the parent stays alive for too long, there is no way for the system to just remove some of the zombies, because it is assumed that the parent should get interested in them at some point via wait or waitpid.
Child processes also hold some resource like memory. But they are not released because parent process can not process SIGCHLD signal, which will be sent by child processes when they exit.
Those child processes will become zombie.
You can use "ps -aux" to dump those fd.

How can parents waits all child processes finish then exit when it receives SIGINT?

I'm writing a multi process program in C.
I hope parents process can wait all child processes finish then exit when
it receives SIGINT.
I have two questions.
How can parents record each pid of child process it forked.
Child process may finish before recording function run on
main process.
If parents has no idea about how many child processes it has. How can he wait
all child process finish.
Thanks in advance.
You record the pid of child processes as you fork them (if required).
call waitpid in a loop with pid = 0 it will either return a pid of a process that exited or return -1 and if errno = ECHILD you have no slaves left.
Keep calling wait(2) in a loop. Every time wait() returns, you'll get back the PID of the exited child and its status. The status will tell you, whether it exited normally (with an exit code) or due to a signal. Something like this (not tested):
#include <sys/types.h>
#include <sys/wait.h>
...
pid_t pid;
int status;
...
while ((pid = wait(&status)) > 0) {
printf("Child %lu ", (unsigned long)pid);
if (WIFEXITED(status))
printf("exited with status %d\n", WEXITSTATUS(status));
else if (WIFSIGNALED(status))
printf("killed by signal %d\n", WTERMSIG(status));
else if (WIFSTOPPED(status))
printf("stopped by signal %d\n", WSTOPSIG(status));
else if (WIFCONTINUED(status))
printf("resumed\n");
else
warnx("wait(2) returned for no discernible reason");
}

using fork to do something while the child is running

So basically what i need is:
pid = fork();
if (pid == -1)
exit(1);
if (pid == 0)
{
// do stuff in child
}
else
{
// ONLY do stuff while child is running
}
would I need to create a tmp file right before the child exits saying that it is no longer running so the parent knows the child has exited when that file exists, or is there a simpler way to do this?
You can use waitpid to know if a child process is still running:
int status;
if (waitpid(pid, &status, WNOHANG) == 0) {
// still running
}
With WNOHANG, waitpid returns immediately so that the program can do something else.
When you have nothing to do other than waiting for the child process to terminate, call waitpid without WNOHANG.
The standard way to know that the child has terminated (and get its exit code) is to use the waitpid() system call.
Check wait () and waitpid () : http://linux.die.net/man/2/wait
Here is some more resource: http://users.actcom.co.il/~choo/lupg/tutorials/multi-process/multi-process.html#child_death_wait_syscall
There's a bunch of ways to do it. If you don't need to do anything with the child output, you can set a SIGCHLD handler to reap the child when it exits, and then forget about it in your main thread of execution. You can use the SIGCHLD handler to flag the exit of the child process via an IPC mechanism.
Or you can add a while loop that checks waitpid in your else clause. You would be doing discrete units of work between polls of the child state and you wouldn't get interrupted immediately on child exit.
Use the system wait() call if you just need to check if the child has stopped running.
int pid;
int status;
while (true)
{
pid = wait(&status);
if (pid < 0)
//keep waiting
else if (pid == 0)
//child is done
}

What exactly does fork return?

On success, the PID of the child
process is returned in the parent’s
thread of execution, and a 0 is
returned in the child’s thread of execution.
p = fork();
I'm confused at its manual page,is p equal to 0 or PID?
I'm not sure how the manual can be any clearer! fork() creates a new process, so you now have two identical processes. To distinguish between them, the return value of fork() differs. In the original process, you get the PID of the child process. In the child process, you get 0.
So a canonical use is as follows:
p = fork();
if (0 == p)
{
// We're the child process
}
else if (p > 0)
{
// We're the parent process
}
else
{
// We're the parent process, but child couldn't be created
}
p = fork();
/* assume no errors */
/* you now have two */
/* programs running */
--------------------
if (p > 0) { | if (p == 0) {
printf("parent\n"); | printf("child\n");
... | ...
Processes are structured in a directed tree where you only know your single-parent (getppid()). In short, fork() returns -1 on error like many other system functions, non-zero value is useful for initiator of the fork call (the parent) to know its new-child pid.
Nothing is as good as example:
/* fork/getpid test */
#include <sys/types.h>
#include <unistd.h> /* fork(), getpid() */
#include <stdio.h>
int main(int argc, char* argv[])
{
int pid;
printf("Entry point: my pid is %d, parent pid is %d\n",
getpid(), getppid());
pid = fork();
if (pid == 0) {
printf("Child: my pid is %d, parent pid is %d\n",
getpid(), getppid());
}
else if (pid > 0) {
printf("Parent: my pid is %d, parent pid is %d, my child pid is %d\n",
getpid(), getppid(), pid);
}
else {
printf("Parent: oops! can not create a child (my pid is %d)\n",
getpid());
}
return 0;
}
And the result (bash is pid 2249, in this case):
Entry point: my pid is 16051, parent pid is 2249
Parent: my pid is 16051, parent pid is 2249, my child pid is 16052
Child: my pid is 16052, parent pid is 16051
If you need to share some resources (files, parent pid, etc.) between parent and child, look at clone() (for GNU C library, and maybe others)
Once fork is executed, you have two processes. The call returns different values to each process.
If you do something like this
int f;
f = fork();
if (f == 0) {
printf("I am the child\n");
} else {
printf("I am the parent and the childs pid is %d\n",f);
}
You will see both the messages printed. They're being printed by two separate processes. This is they way you can differentiate between the two processes created.
This is the cool part. It's equal to BOTH.
Well, not really. But once fork returns, you now have two copies of your program running! Two processes. You can sort of think of them as alternate universes. In one, the return value is 0. In the other, it's the ID of the new process!
Usually you will have something like this:
p = fork();
if (p == 0){
printf("I am a child process!\n");
//Do child things
}
else {
printf("I am the parent process! Child is number %d\n", p);
//Do parenty things
}
In this case, both strings will get printed, but by different processes!
fork() is invoked in the parent process. Then a child process is spawned. By the time the child process spawns, fork() has finished its execution.
At this point, fork() is ready to return, but it returns a different value depending on whether it's in the parent or child. In the child process, it returns 0, and in the parent process/thread, it returns the child's process ID.
Fork creates a duplicate process and a new process context. When it returns a 0 value it means that a child process is running, but when it returns another value that means a parent process is running. We usually use wait statement so that a child process completes and parent process starts executing.
I think that it works like this:
when pid = fork(), the code should be executed two times, one is in current process, one is in child process.
So it explains why if/else both execute.
And the order is, first current process, and then execute the child.

Resources