Home/ Questions/Q 9008245
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-06-16T01:53:50+00:00

I have a working code running on a RedHat installed Linux with kernel 2.6.18-194.el5

  • 0

I have a working code running on a RedHat installed Linux with kernel 2.6.18-194.el5 #1 SMP x86_64.

When I moved the code to a newly installed machine with CentOS 6.3 the same code failed with

Error assigning socket option: Numerical argument out of domain

The kernel version of the latter machine is 2.6.32-279.el6.x86_64 #1 SMP

Below is the code of what’s both working and failing in the machines, respectively.

    struct timeval          tv;

    tv.tv_sec       = 0;
    tv.tv_usec      = 1500000;

    if (setsockopt(sockfd, SOL_SOCKET, SO_RCVTIMEO, &tv, sizeof(struct timeval)) != 0)
    {
            LM_ERR("Error assigning socket option: %s", strerror( errno ));
            return FALSE;
    }
Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    Try using 

    tv.tv_sec       = 1;
tv.tv_usec      = 500000;

    I’ve seen some implementations not accepting tv_usec above 10^6.

    Edit: The issue is interesting enough to dig a little bit. Looking for SO_RCVTIMEO in the kernel source code I’ve found the following piece of code in net/core/sock.c:

    int sock_setsockopt(struct socket *sock, int level, int optname,
            char __user *optval, unsigned int optlen)
{

   / ... /    
   switch (optname) {

   / ... /    

   case SO_RCVTIMEO:
        ret = sock_set_timeout(&sk->sk_rcvtimeo, optval, optlen);
        break;

   / ... /    
}

    The beginning of the sock_set_timeout() function indeed contains some range checking:

    static int sock_set_timeout(long *timeo_p, char __user *optval, int optlen)
{
    struct timeval tv;

    if (optlen < sizeof(tv))
        return -EINVAL;
    if (copy_from_user(&tv, optval, sizeof(tv)))
        return -EFAULT;
    if (tv.tv_usec < 0 || tv.tv_usec >= USEC_PER_SEC)
        return -EDOM;

    /* ... */
}

    Now we know enough to do some git blame-ing 🙂 The change was introduced with the following changeset:

    commit ba78073e6f70cd9c64a478a9bd901d7c8736cfbc 
Author: Vasily Averin <vvs@sw.ru> 
Date:   Thu May 24 16:58:54 2007 -0700

[NET]: "wrong timeout value" in sk_wait_data() v2

sys_setsockopt() do not check properly timeout values for 
SO_RCVTIMEO/SO_SNDTIMEO, for example it's possible to set negative timeout
values. POSIX do not defines behaviour for sys_setsockopt in case negative
timeouts, but requires that setsockopt() shall fail with -EDOM if the send and
receive timeout values are too big to fit into the timeout fields in the socket
structure. 
In current implementation negative timeout can lead to error messages like
"schedule_timeout: wrong timeout value".

Proposed patch:
- checks tv_usec and returns -EDOM if it is wrong
- do not allows to set negative timeout values (sets 0 instead) and outputs
  ratelimited information message about such attempts.

Signed-off-By: Vasily Averin <vvs@sw.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>

    I believe that the commit comment explains everything. As far as I can see this change was included in 2.6.22-rc3.

    • 0