Em Tue, Sep 30, 2008 at 06:24:57PM -0300, Leandro Sales escreveu:
> On Tue, Sep 30, 2008 at 4:16 PM, Arnaldo Carvalho de Melo
> <acme@xxxxxxxxxx> wrote:
> > Em Tue, Sep 30, 2008 at 03:32:17PM -0300, Leandro Sales escreveu:
> >> On Mon, Sep 29, 2008 at 3:37 PM, Arnaldo Carvalho de Melo
> >> <acme@xxxxxxxxxx> wrote:
> >> > Em Mon, Sep 29, 2008 at 03:33:44PM -0300, Leandro Sales escreveu:
> >> >> Hello folks,
> >> >> I'm getting this message while using 2.6.27:
> >> >>
> >> >> Can not create DCCP socket (Protocol driver not attached)
> >> >
> >> > What is the source of such messages? VLC? What? Try strace to see what
> >> > is the syscall that fails just before this message is printed, etc.
> >> >
> >> >> "uname -a" output:
> >> >> Linux localhost 2.6.27-rc4-44301-g8269be1 #7 SMP Sun Sep 28 17:02:25
> >> >> BRT 2008 i686 AMD Athlon(tm) 64 Processor 3000+ AuthenticAMD GNU/Linux
> >> >>
> >> >> Anything wrong? I tried compile DCCP using both static and module
> >> >> methods and it produces the same effect.
> >> >
> >> > - Arnaldo
> >> >
> >>
> >> Hello Arnaldo,
> >> I'm just running tests with iperf to later use gstreamer plugin that
> >> I developed. Here it is the output provided by strace while executing
> >> iperf. I'm sorry for the long message. In addition, I'm providing
> >> support for a guy from China that wants to use my gstreamer plugin and
> >> he is facing similar problem, getting a permission denied message
> >> while using the plugin, even then he run the plugin as root.
> >
> > iperf is multithreaded, so you need to run strace with -f to trace child
> > processes, also please try limiting the output to just the networking
> > syscalls:
> >
> > strace -f -e trace=network iperf
> >
> > "net namespaces" blipped in my mind, but it may well be just my brain
> > needing more coffee...
> >
> > - Arnaldo
> >
>
> Arnaldo, thank you for the reply. Please, check the strace output
> below. I'm not sure what it happen, although I have a thought. Any
> clue?
>
> Process 5849 attached (waiting for parent)
> Process 5849 resumed (parent 5848 ready)
> [pid 5849] socket(PF_NETLINK, SOCK_RAW, 0) = 3
> [pid 5849] bind(3, {sa_family=AF_NETLINK, pid=0, groups=00000000}, 12) = 0
> [pid 5849] getsockname(3, {sa_family=AF_NETLINK, pid=5848,
> groups=00000000}, [12]) = 0
> [pid 5849] sendto(3, "\24\0\0\0\26\0\1\3&\230\342H\0\0\0\0\0\0\0\0",
> 20, 0, {sa_family=AF_NETLINK, pid=0, groups=00000000}, 12) = 20
> [pid 5849] recvmsg(3, {msg_name(12)={sa_family=AF_NETLINK, pid=0,
> groups=00000000},
> msg_iov(1)=[{"8\0\0\0\24\0\2\0&\230\342H\330\26\0\0\2\10\200\376\1\0\0\0\10\0\1\0\177\0\0\1"...,
> 4096}], msg_controllen=0, msg_flags=0}, 0) = 116
> [pid 5849] recvmsg(3, {msg_name(12)={sa_family=AF_NETLINK, pid=0,
> groups=00000000},
> msg_iov(1)=[{"@\0\0\0\24\0\2\0&\230\342H\330\26\0\0\n\200\200\376\1\0\0\0\24\0\1\0\0\0\0\0"...,
> 4096}], msg_controllen=0, msg_flags=0}, 0) = 128
> [pid 5849] recvmsg(3, {msg_name(12)={sa_family=AF_NETLINK, pid=0,
> groups=00000000},
> msg_iov(1)=[{"\24\0\0\0\3\0\2\0&\230\342H\330\26\0\0\0\0\0\0\1\0\0\0\24\0\1\0\0\0\0\0"...,
> 4096}], msg_controllen=0, msg_flags=0}, 0) = 20
> [pid 5849] socket(PF_INET6, SOCK_DGRAM, IPPROTO_IP) = 3
> [pid 5849] connect(3, {sa_family=AF_INET6, sin6_port=htons(5001),
> inet_pton(AF_INET6, "::", &sin6_addr), sin6_flowinfo=0,
> sin6_scope_id=0}, 28) = 0
> [pid 5849] getsockname(3, {sa_family=AF_INET6,
> sin6_port=htons(56372), inet_pton(AF_INET6, "::1", &sin6_addr),
> sin6_flowinfo=0, sin6_scope_id=0}, [28]) = 0
> [pid 5849] socket(PF_INET, SOCK_DGRAM, IPPROTO_IP) = 3
> [pid 5849] connect(3, {sa_family=AF_INET, sin_port=htons(5001),
> sin_addr=inet_addr("0.0.0.0")}, 16) = 0
> [pid 5849] getsockname(3, {sa_family=AF_INET, sin_port=htons(37587),
> sin_addr=inet_addr("127.0.0.1")}, [16]) = 0
> [pid 5849] socket(PF_INET6, 0x6 /* SOCK_??? */, 0x21 /* IPPROTO_???
> */) = -1 EUNATCH (Protocol driver not attached)
Here we gett the EUNATACH when trying SOCK_DCCP (6), IPPROTO_DCCP (33),
but I couldn't find any EUNATACH return in the sock_create path, do you
have selinux running?
It tries several times PF_INET6 then falls back to PF_INET. Is this
using glibc? dietlibc? ARM? lsmod output can also help.
- Arnaldo
--
To unsubscribe from this list: send the line "unsubscribe dccp" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
