fork() can fail

#1

Post by dantiberian » Wed, Aug 20, 2014, 7:15 PM UTC

Re: fork() can fail

#2

Post by emo_tards_on_hn » Wed, Aug 20, 2014, 7:42 PM UTC

This is an important point, along similar lines to something we discussed at work today in a code review. The programmer in question assumed that all and every request for the resources consumed would always work. Literally, no testing or consideration for what if one of those request fails. Will be emailing them this article tomorrow!

Re: fork() can fail

#3

Post by japaget » Wed, Aug 20, 2014, 7:48 PM UTC

See also https://news.ycombinator.com/item?id=8189968 for another UNIX trap for the unwary.

Re: fork() can fail

#4

Post by arbitrage » Wed, Aug 20, 2014, 7:49 PM UTC

emo_tards_on_hn wrote: post #2

This is an important point, along similar lines to something we discussed at work today in a code review. The programmer in question assumed that all and every request for the resources consumed would always work. Literally, no testing or consideration for what if one of those request fails. Will be emailing them this article tomorrow!

Your username is offensive. I'd like to take what you say seriously, and perhaps even engage in further conversation about the topic at hand, but ... you know ... you look stupid. Grow up. You should be embarrassed.

Re: fork() can fail

#5

Post by dasmithii » Wed, Aug 20, 2014, 7:50 PM UTC

I'm uncertain that I've ever checked for errors after calling fork.

Thanks for providing an impetus to do so.

Re: fork() can fail

#6

Post by jgrahamc » Wed, Aug 20, 2014, 7:52 PM UTC

Quietly goes to check the last piece of C I wrote containing a fork():

    if (daemon && !test_mode) {
      int pid = fork();
      if (pid == -1) {
        fatal_error("Failed to fork");
      }
      if (pid != 0) {
        write_pid(pid_file, pid, !test_mode);
        exit(0);
      }
    } else {
      write_pid(pid_file, getpid(), !test_mode);
    }

Phew!

Re: fork() can fail

#7

Post by cperciva » Wed, Aug 20, 2014, 7:53 PM UTC

This reminds me of one of the most epic bugs I've ever run into:

    mkdir("/foo", 0700);
    chdir("/foo");
    recursively_delete_everything_in_current_directory();

Running as root, this usually worked fine: It would create a directory, move into it, and clean out any garbage left behind by a previous run before doing anything new.

Running as non-root, the mkdir failed, the chdir failed, and it started eating my home directory.

Re: fork() can fail

#8

Post by runarb » Wed, Aug 20, 2014, 7:53 PM UTC

Easy to test out to. In C:

  #include 
  
  int main(void)
  {
         while(1) {
                 fork();
         }
  }

Re: fork() can fail

#9

Post by dang » Wed, Aug 20, 2014, 7:53 PM UTC

arbitrage wrote: post #4

emo_tards_on_hn wrote: post #2

This is an important point, along similar lines to something we discussed at work today in a code review. The programmer in question assumed that all and every request for the resources consumed would always work. Literally, no testing or consideration for what if one of those request fails. Will be emailing them this article tomorrow!

Your username is offensive. I'd like to take what you say seriously, and perhaps even engage in further conversation about the topic at hand, but ... you know ... you look stupid. Grow up. You should be embarrassed.

That's understandable, but please don't feed the trolls.

When you see a comment that is truly bad for HN, it's best to flag it. You can do this by clicking on "link" to go to the item page for the comment, then "flag". We monitor those flags and take action based on them.

Re: fork() can fail

#10

Post by Aurel1us » Wed, Aug 20, 2014, 7:54 PM UTC

Just as a reminder: "So, malloc on Linux only fails if there isn’t enough memory for its control structures. It does not fail if there isn’t enough memory to fulfill the request." - http://scvalex.net/posts/6/