sagfault with p2p socket

[zeromq]

include <zmq.h>

include <stdio.h>

include <stdlib.h>

int main (int argc, char *argv [])
{
const char *bind_to;
int message_count;
int message_size;
void *ctx;
void *s;
int rc;
int i;
zmq_msg_t msg;
void *watch;
unsigned long elapsed;
unsigned long throughput;
double megabits;

if (argc != 4) {
    printf ("usage: local_thr <bind-to> <message-size> <message-count>\n");
    return 1;
}
bind_to = argv [1];
message_size = atoi (argv [2]);
message_count = atoi (argv [3]);

ctx = zmq_init (1, 1, 0);
if (!ctx) {
    printf ("error in zmq_send: %s\n", zmq_strerror (errno));
    return -1;
}

s = zmq_socket (ctx, ZMQ_P2P);
if (!s) {
    printf ("error in zmq_socket: %s\n", zmq_strerror (errno));
    return -1;
}

//rc = zmq_setsockopt (s, ZMQ_SUBSCRIBE , "", 0);
//if (rc != 0) {
//    printf ("error in zmq_setsockopt: %s\n", zmq_strerror (errno));
//    return -1;
//}

//  Add your socket options here.
//  For example ZMQ_RATE, ZMQ_RECOVERY_IVL and ZMQ_MCAST_LOOP for PGM.

rc = zmq_bind (s, bind_to);
if (rc != 0) {
    printf ("error in zmq_bind: %s\n", zmq_strerror (errno));
    return -1;
}

rc = zmq_msg_init (&msg);
if (rc != 0) {
    printf ("error in zmq_msg_init: %s\n", zmq_strerror (errno));
    return -1;
}

rc = zmq_recv (s, &msg, 0);
if (rc != 0) {
    printf ("error in zmq_recv: %s\n", zmq_strerror (errno));
    return -1;
}
if (zmq_msg_size (&msg) != message_size) {
    printf ("message of incorrect size received\n");
    return -1;
}

watch = zmq_stopwatch_start ();

for (i = 0; i != message_count - 1; i++) {
    rc = zmq_recv (s, &msg, 0);
    if (rc != 0) {
        printf ("error in zmq_recv: %s\n", zmq_strerror (errno));
        return -1;
    }
    if (zmq_msg_size (&msg) != message_size) {
        printf ("message of incorrect size received\n");
        return -1;
    }
}

elapsed = zmq_stopwatch_stop (watch);
if (elapsed == 0)
    elapsed = 1;

rc = zmq_msg_close (&msg);
if (rc != 0) {
    printf ("error in zmq_msg_close: %s\n", zmq_strerror (errno));
    return -1;
}
//rc = zmq_msg_init_size(&msg, 2);
//rc = zmq_send(s, &msg, 0);
//rc = zmq_msg_close(&msg);

throughput = (unsigned long)
    ((double) message_count / (double) elapsed * 1000000);
megabits = (double) (throughput * message_size * 8) / 1000000;

printf ("message size: %d [B]\n", (int) message_size);
printf ("message count: %d\n", (int) message_count);
printf ("mean throughput: %d [msg/s]\n", (int) throughput);
printf ("mean throughput: %.3f [Mb/s]\n", (double) megabits);

rc = zmq_close (s);
if (rc != 0) {
    printf ("error in zmq_close: %s\n", zmq_strerror (errno));
    return -1;
}

rc = zmq_term (ctx);
if (rc != 0) {
    printf ("error in zmq_term: %s\n", zmq_strerror (errno));
    return -1;
}

return 0;

}

output:
(gdb) run
Starting program: /home/tcip/src/zeromq/foo/local_thr tcp://192.168.2.2:20000 1000000 100
[Thread debugging using libthread_db enabled]
[New Thread 0x7fefeca12750 (LWP 9184)]
[New Thread 0x7fefeb68d950 (LWP 9187)]

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fefeca12750 (LWP 9184)]
zmq_msg_close (msg_=0x7fffb76ff780) at zmq.cpp:168
168 if (content->ffn)
Current language: auto; currently c++
(gdb) bt
#0 zmq_msg_close (msg_=0x7fffb76ff780) at zmq.cpp:168
#1 0x00007fefec5e9fec in zmq::p2p_t::xrecv (this=0x1782fe0,

msg_=0x7fffb76ff780, flags_=0) at p2p.cpp:115

#2 0x00007fefec5f208f in zmq::socket_base_t::recv (this=0x1782fe0,

msg_=0x7fffb76ff780, flags_=0) at socket_base.cpp:393

#3 0x0000000000400dc7 in main ()

(gdb)

[sophacles]

I double checked this against a build from the latest trunk (above was against the 2.0.6 in Adrian's hg repo) and the backtrace becomes (note a couple of line number changes):

#0 zmq_msg_close (msg_=0x7fffe5061e10) at zmq.cpp:170
#1 0x00007f9afd1a0f4c in zmq::p2p_t::xrecv (this=0x20143f0,
msg_=0x7fffe5061e10, flags_=0) at p2p.cpp:115
#2 0x00007f9afd1a8ef9 in zmq::socket_base_t::recv (this=0x20143f0,
msg_=0x7fffe5061e10, flags_=0) at socket_base.cpp:442
#3 0x0000000000400dc7 in main ()

[zeromq]

I have run the remote_thr supplied and it simply stays still, waiting for messages. What do you use as the other peer?

[sophacles]

The orignially linked code is the local_thr code, this code below is the remote_thr code that causes local_thr to segfault (atho it is literally a s/ZMQ_SUB/ZMQ_P2P/):

#include "../include/zmq.h"
#include 
#include 

int main (int argc, char *argv [])
{
    const char *connect_to;
    int message_count;
    int message_size;
    void *ctx;
    void *s;
    int rc;
    int i;
    int stime;
    zmq_msg_t msg;

    if (argc != 5) {
        printf ("usage: remote_thr   "
            " \n");
        return 1;
    }
    connect_to = argv [1];
    message_size = atoi (argv [2]);
    message_count = atoi (argv [3]);
    stime = atoi (argv[4]);

    ctx = zmq_init (1, 1, 0);
    if (!ctx) {
        printf ("error in zmq_recv: %s\n", zmq_strerror (errno));
        return -1;
    }

    s = zmq_socket (ctx, ZMQ_P2P);
    if (!s) {
        printf ("error in zmq_socket: %s\n", zmq_strerror (errno));
        return -1;
    }

    //  Add your socket options here.
    //  For example ZMQ_RATE, ZMQ_RECOVERY_IVL and ZMQ_MCAST_LOOP for PGM.

    rc = zmq_connect (s, connect_to);
    if (rc != 0) {
        printf ("error in zmq_connect: %s\n", zmq_strerror (errno));
        return -1;
    }

    for (i = 0; i != message_count; i++) {
        rc = zmq_msg_init_size (&msg, message_size);
        if (rc != 0) {
            printf ("error in zmq_msg_init_size: %s\n", zmq_strerror (errno));
            return -1;
        }
        rc = zmq_send (s, &msg, 0);
        if (rc != 0) {
            printf ("error in zmq_send: %s\n", zmq_strerror (errno));
            return -1;
        }
        rc = zmq_msg_close (&msg);
        if (rc != 0) {
            printf ("error in zmq_msg_close: %s\n", zmq_strerror (errno));
            return -1;
        }
    }

    zmq_sleep (stime);

    rc = zmq_close (s);
    if (rc != 0) {
        printf ("error in zmq_close: %s\n", zmq_strerror (errno));
        return -1;
    }

    rc = zmq_term (ctx);
    if (rc != 0) {
        printf ("error in zmq_term: %s\n", zmq_strerror (errno));
        return -1;
    }

    return 0;
}

[sophacles]

Also, I did a fresh build of everything involved, there is still a segfault, but excitingly not every time. Sometimes everything works fine, and sometimes I get the following glibc backtrace:

*** glibc detected *** ./lthrp2p: double free or corruption (!prev): 0x0000000000614100 ***
======= Backtrace: =========
/lib/libc.so.6[0x2ac24a608d16]
/lib/libc.so.6(cfree+0x6c)[0x2ac24a60d9bc]
/usr/lib/libzmq.so.0(zmq_msg_close+0x2e)[0x2ac24a34e82e]
/usr/lib/libzmq.so.0(_ZN3zmq5p2p_t5xrecvEP9zmq_msg_ti+0x1c)[0x2ac24a34145c]
/usr/lib/libzmq.so.0(_ZN3zmq13socket_base_t4recvEP9zmq_msg_ti+0x86)[0x2ac24a348af6]
./lthrp2p[0x400dd1]
/lib/libc.so.6(__libc_start_main+0xfd)[0x2ac24a5b6abd]
./lthrp2p[0x400ad9]
======= Memory map: ========
00400000-00402000 r-xp 00000000 03:01 336666                             /home/tcip/src/zeromq/foo/lthrp2p
00601000-00602000 rw-p 00001000 03:01 336666                             /home/tcip/src/zeromq/foo/lthrp2p
00602000-00623000 rw-p 00602000 00:00 0                                  [heap]
2ac24a0f7000-2ac24a114000 r-xp 00000000 03:01 50190                      /lib/ld-2.10.2.so
2ac24a114000-2ac24a117000 rw-p 2ac24a114000 00:00 0 
2ac24a313000-2ac24a314000 r--p 0001c000 03:01 50190                      /lib/ld-2.10.2.so
2ac24a314000-2ac24a315000 rw-p 0001d000 03:01 50190                      /lib/ld-2.10.2.so
2ac24a315000-2ac24a394000 r-xp 00000000 03:01 24998                      /usr/lib/libzmq.so.0.0.0
2ac24a394000-2ac24a594000 ---p 0007f000 03:01 24998                      /usr/lib/libzmq.so.0.0.0
2ac24a594000-2ac24a598000 rw-p 0007f000 03:01 24998                      /usr/lib/libzmq.so.0.0.0
2ac24a598000-2ac24a6e2000 r-xp 00000000 03:01 50558                      /lib/libc-2.10.2.so
2ac24a6e2000-2ac24a8e2000 ---p 0014a000 03:01 50558                      /lib/libc-2.10.2.so
2ac24a8e2000-2ac24a8e6000 r--p 0014a000 03:01 50558                      /lib/libc-2.10.2.so
2ac24a8e6000-2ac24a8e7000 rw-p 0014e000 03:01 50558                      /lib/libc-2.10.2.so
2ac24a8e7000-2ac24a8ec000 rw-p 2ac24a8e7000 00:00 0 
2ac24a8ec000-2ac24a9de000 r-xp 00000000 03:01 24806                      /usr/lib/libstdc++.so.6.0.13
2ac24a9de000-2ac24abde000 ---p 000f2000 03:01 24806                      /usr/lib/libstdc++.so.6.0.13
2ac24abde000-2ac24abe5000 r--p 000f2000 03:01 24806                      /usr/lib/libstdc++.so.6.0.13
2ac24abe5000-2ac24abe7000 rw-p 000f9000 03:01 24806                      /usr/lib/libstdc++.so.6.0.13
2ac24abe7000-2ac24abfc000 rw-p 2ac24abe7000 00:00 0 
2ac24abfc000-2ac24ac00000 r-xp 00000000 03:01 25550                      /usr/lib/libgthread-2.0.so.0.2200.4
2ac24ac00000-2ac24adff000 ---p 00004000 03:01 25550                      /usr/lib/libgthread-2.0.so.0.2200.4
2ac24adff000-2ac24ae00000 rw-p 00003000 03:01 25550                      /usr/lib/libgthread-2.0.so.0.2200.4
2ac24ae00000-2ac24ae01000 rw-p 2ac24ae00000 00:00 0 
2ac24ae01000-2ac24ae08000 r-xp 00000000 03:01 50548                      /lib/librt-2.10.2.so
2ac24ae08000-2ac24b007000 ---p 00007000 03:01 50548                      /lib/librt-2.10.2.so
2ac24b007000-2ac24b008000 r--p 00006000 03:01 50548                      /lib/librt-2.10.2.so
2ac24b008000-2ac24b009000 rw-p 00007000 03:01 50548                      /lib/librt-2.10.2.so
2ac24b009000-2ac24b0ce000 r-xp 00000000 03:01 50573                      /lib/libglib-2.0.so.0.2200.4
2ac24b0ce000-2ac24b2cd000 ---p 000c5000 03:01 50573                      /lib/libglib-2.0.so.0.2200.4
2ac24b2cd000-2ac24b2cf000 rw-p 000c4000 03:01 50573                      /lib/libglib-2.0.so.0.2200.4
2ac24b2cf000-2ac24b2d2000 r-xp 00000000 03:01 49245                      /lib/libuuid.so.1.3.0
2ac24b2d2000-2ac24b4d2000 ---p 00003000 03:01 49245                      /lib/libuuid.so.1.3.0
2ac24b4d2000-2ac24b4d3000 rw-p 00003000 03:01 49245                      /lib/libuuid.so.1.3.0
2ac24b4d3000-2ac24b4d4000 rw-p 2ac24b4d3000 00:00 0 
2ac24b4d4000-2ac24b4ea000 r-xp 00000000 03:01 50563                      /lib/libpthread-2.10.2.so
2ac24b4ea000-2ac24b6ea000 ---p 00016000 03:01 50563                      /lib/libpthread-2.10.2.so
2ac24b6ea000-2ac24b6eb000 r--p 00016000 03:01 50563                      /lib/libpthread-2.10.2.so
2ac24b6eb000-2ac24b6ec000 rw-p 00017000 03:01 50563                      /lib/libpthread-2.10.2.so
2ac24bAborted

[zeromq]

The terminal zero of the string is not part of the message.

[sustrik]

User error AFAIU. Closing the issue.