[mvapich-discuss] mpirun: "bind: address family not supported by protocol" "

Bernward Platz bernward.platz at googlemail.com
Sat Feb 23 16:05:31 EST 2013


Hi,

in an infiniband-cluster when I execute


mpirun_rsh -ssh -np <cpus> -hostfile hostfile <program>

I always get a

bind: Address family not supported by protocol

I get the same error-message with all  binaries also from the
mvapich-test-suite.

ssh is configured and tested correctly  (no prompting for password or
something else).
I tried it with hostnames, ip-adresses, localhost. Always the same message.

mpirun from openmpi is working fine.

Platform: Linux x86_64
OS: SLES 11p2 +
mpvapich: mvapich from MLNX_OFED_LINUX-1.5.3-3.1.0-sles11sp2-x86_64

Output of strace:

execve("/usr/mpi/gcc/mvapich-1.2.0/bin/mpirun_rsh",
["/usr/mpi/gcc/mvapich-1.2.0/bin/m"..., "-ssh", "-np", "4", "-hostfile",
"hostfile", "./osu_alltoall", "cd", "/CFD/fs1"], [/* 60 vars */]) = 0
brk(0)                                  = 0x613000
mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) =
0x7f7aab703000
access("/etc/ld.so.preload", R_OK)      = -1 ENOENT (No such file or
directory)
open("/usr/mpi/gcc/mvapich-1.2.0/lib/shared/tls/x86_64/libm.so.6",
O_RDONLY) = -1 ENOENT (No such file or directory)
stat("/usr/mpi/gcc/mvapich-1.2.0/lib/shared/tls/x86_64", 0x7fff29e5edd0) =
-1 ENOENT (No such file or directory)
open("/usr/mpi/gcc/mvapich-1.2.0/lib/shared/tls/libm.so.6", O_RDONLY) = -1
ENOENT (No such file or directory)
stat("/usr/mpi/gcc/mvapich-1.2.0/lib/shared/tls", 0x7fff29e5edd0) = -1
ENOENT (No such file or directory)
open("/usr/mpi/gcc/mvapich-1.2.0/lib/shared/x86_64/libm.so.6", O_RDONLY) =
-1 ENOENT (No such file or directory)
stat("/usr/mpi/gcc/mvapich-1.2.0/lib/shared/x86_64", 0x7fff29e5edd0) = -1
ENOENT (No such file or directory)
open("/usr/mpi/gcc/mvapich-1.2.0/lib/shared/libm.so.6", O_RDONLY) = -1
ENOENT (No such file or directory)
stat("/usr/mpi/gcc/mvapich-1.2.0/lib/shared", {st_mode=S_IFDIR|0755,
st_size=4096, ...}) = 0
open("/etc/ld.so.cache", O_RDONLY)      = 3
fstat(3, {st_mode=S_IFREG|0644, st_size=222401, ...}) = 0
mmap(NULL, 222401, PROT_READ, MAP_PRIVATE, 3, 0) = 0x7f7aab6cc000
close(3)                                = 0
open("/lib64/libm.so.6", O_RDONLY)      = 3
read(3,
"\177ELF\2\1\1\0\0\0\0\0\0\0\0\0\3\0>\0\1\0\0\0\200E\0\0\0\0\0\0"..., 832)
= 832
fstat(3, {st_mode=S_IFREG|0755, st_size=541821, ...}) = 0
mmap(NULL, 2592072, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) =
0x7f7aab26d000
fadvise64(3, 0, 2592072, POSIX_FADV_WILLNEED) = 0
mprotect(0x7f7aab2c8000, 2093056, PROT_NONE) = 0
mmap(0x7f7aab4c7000, 126976, PROT_READ|PROT_WRITE,
MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x5a000) = 0x7f7aab4c7000
close(3)                                = 0
open("/usr/mpi/gcc/mvapich-1.2.0/lib/shared/libc.so.6", O_RDONLY) = -1
ENOENT (No such file or directory)
open("/lib64/libc.so.6", O_RDONLY)      = 3
read(3,
"\177ELF\2\1\1\0\0\0\0\0\0\0\0\0\3\0>\0\1\0\0\0P\355\1\0\0\0\0\0"..., 832)
= 832
fstat(3, {st_mode=S_IFREG|0755, st_size=1754140, ...}) = 0
mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) =
0x7f7aab6cb000
mmap(NULL, 3619016, PROT_READ|PROT_EXEC, MAP_PRIVATE|MAP_DENYWRITE, 3, 0) =
0x7f7aaaef9000
fadvise64(3, 0, 3619016, POSIX_FADV_WILLNEED) = 0
mprotect(0x7f7aab064000, 2093056, PROT_NONE) = 0
mmap(0x7f7aab263000, 20480, PROT_READ|PROT_WRITE,
MAP_PRIVATE|MAP_FIXED|MAP_DENYWRITE, 3, 0x16a000) = 0x7f7aab263000
mmap(0x7f7aab268000, 18632, PROT_READ|PROT_WRITE,
MAP_PRIVATE|MAP_FIXED|MAP_ANONYMOUS, -1, 0) = 0x7f7aab268000
close(3)                                = 0
mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) =
0x7f7aab6ca000
mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) =
0x7f7aab6c9000
arch_prctl(ARCH_SET_FS, 0x7f7aab6ca700) = 0
mprotect(0x7f7aab263000, 16384, PROT_READ) = 0
mprotect(0x7f7aab4c7000, 4096, PROT_READ) = 0
mprotect(0x610000, 4096, PROT_READ)     = 0
mprotect(0x7f7aab704000, 4096, PROT_READ) = 0
munmap(0x7f7aab6cc000, 222401)          = 0
brk(0)                                  = 0x613000
brk(0x634000)                           = 0x634000
access("/usr/mpi/gcc/mvapich-1.2.0/etc/mvapich.conf", R_OK) = 0
open("/usr/mpi/gcc/mvapich-1.2.0/etc/mvapich.conf", O_RDONLY) = 3
fstat(3, {st_mode=S_IFREG|0644, st_size=27877, ...}) = 0
mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) =
0x7f7aab702000
read(3, "#-------------------------------"..., 4096) = 4096
read(3, "TAL_SIZE=9216\n# For a PCI Expres"..., 4096) = 4096
read(3, "-------------------------\n# VIAD"..., 4096) = 4096
read(3, "n\n#-----------------------------"..., 4096) = 4096
read(3, "epost_depth + viadev_prepost_ren"..., 4096) = 4096
read(3, "#\n# Default value: 1<<10\n# The t"..., 4096) = 4096
read(3, "efault value: 1 (enabled)\n# This"..., 4096) = 3301
read(3, "", 4096)                       = 0
close(3)                                = 0
munmap(0x7f7aab702000, 4096)            = 0
open("hostfile", O_RDONLY)              = 3
fstat(3, {st_mode=S_IFREG|0644, st_size=40, ...}) = 0
mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) =
0x7f7aab702000
read(3, "127.0.0.1\n127.0.0.1\n127.0.0.1\n12"..., 4096) = 40
close(3)                                = 0
munmap(0x7f7aab702000, 4096)            = 0
stat(".", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
stat("/usr/mpi/gcc/mvapich-1.2.0/tests/osu_benchmarks-3.1.1",
{st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
uname({sys="Linux", node="compute01", ...}) = 0
socket(PF_INET, SOCK_STREAM, IPPROTO_TCP) = 3
bind(3, {sa_family=0xf390 /* AF_??? */,
sa_data="\0\0\0\0\0\0.UO\253z\177\0\0"}, 16) = -1 EAFNOSUPPORT (Address
family not supported by protocol)
dup(2)                                  = 4
fcntl(4, F_GETFL)                       = 0x8002 (flags O_RDWR|O_LARGEFILE)
fstat(4, {st_mode=S_IFCHR|0620, st_rdev=makedev(136, 2), ...}) = 0
mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) =
0x7f7aab702000
lseek(4, 0, SEEK_CUR)                   = -1 ESPIPE (Illegal seek)
write(4, "bind: Address family not support"..., 47bind: Address family not
supported by protocol
) = 47
close(4)                                = 0
munmap(0x7f7aab702000, 4096)            = 0
exit_group(1)

Thanks

Bernward
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.cse.ohio-state.edu/pipermail/mvapich-discuss/attachments/20130223/c2b3ae12/attachment-0001.html


More information about the mvapich-discuss mailing list