<< | Thread Index | >> ]    [ << | Date Index | >> ]

Subject: ciped-cb silently dies
From: "digi" <digiwind,AT,hotmail,DOT,com>
Date: Sun, 6 May 2001 03:38:44 +0200

Hi,

I am trying to use ciped-cb (cipe v1.5.1) on a machine with 2.2.16-22smp
kernel. I load cipcb first using modprobe, the cipcb0 interface appears (not
configured) and everything seems to be going smoothly. Then I run ciped-cb
as follows:

# ciped-cb -o /etc/cipe/options
CIPE daemon vers 1.5.1 (c) Olaf Titz 1996-2000
device=(none)
debug=yes
ipaddr=10.1.1.6
ptpaddr=10.1.1.5
mask=
bcast=
mtu=0
metric=0
cttl=0
me=[real_ip_skipped]:9999
peer=[real_ip2_skipped]:8000
key=(secret)
nokey=no
socks=
tokxc=0
tokey=0
ipup=/etc/cipe/ip-up
ipdown=/etc/cipe/ip-down
arg=(none)
maxerr=8
tokxts=0
ping=0
toping=0
dynip=no
hwaddr=(none)
ifconfig=no
checksum=no
opendev: bind

And then it just exits %( and does nothing. Here is what I get from strace:

[skipped]
23690 close(10)                         = -1 EBADF (Bad file descriptor)
23690 close(9)                          = -1 EBADF (Bad file descriptor)
23690 close(8)                          = -1 EBADF (Bad file descriptor)
23690 close(7)                          = -1 EBADF (Bad file descriptor)
23690 close(6)                          = -1 EBADF (Bad file descriptor)
23690 close(5)                          = -1 EBADF (Bad file descriptor)
23690 close(4)                          = -1 EBADF (Bad file descriptor)
23690 close(3)                          = 0
23690 fstat(1, {st_mode=S_IFCHR|0620, st_rdev=makedev(136, 0), ...}) = 0
23690 old_mmap(NULL, 4096, PROT_READ|PROT_WRITE,
MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40018000
23690 ioctl(1, TCGETS, {B38400 opost isig icanon echo ...}) = 0
23690 write(1, "CIPE daemon vers 1.5.1 (c) Olaf "..., 47) = 47
[...]
23690 open("/dev/urandom", O_RDONLY)    = 3
23690 socket(PF_INET, SOCK_DGRAM, IPPROTO_IP) = 4
23690 setsockopt(4, SOL_SOCKET, SO_RCVBUF, [65536], 4) = 0
23690 bind(4, {sin_family=AF_INET, sin_port=htons(9999),
sin_addr=inet_addr("[IP Addr removed]")}}, 16) = 0
23690 connect(4, {sin_family=AF_INET, sin_port=htons(8000),
sin_addr=inet_addr("[peer's IP Addr removed]")}}, 16) = -1 EINVAL (Invalid
argument)
23690 write(2, "opendev: bind", 13)     = 13
23690 write(2, "\n", 1)                 = 1
23690 close(4)                          = 0
23690 munmap(0x40018000, 4096)          = 0
23690 _exit(1)                          = ?

I see that the problem is probably because of the bind but why ? Peer's IP
seems valid, pingable etc. ?

Here is my /etc/cipe/options file:
---
ptpaddr         10.1.1.5
ipaddr          10.1.1.6
me              [real_ip]:9999
peer            [real_ip2]:8000
debug   3
ipup=/etc/cipe/ip-up
ipdown=/etc/cipe/ip-down
key             [skipped]
---

Also, I noticed one strange thing -- when I load cipcb using:
# modprobe cipcb

it writes the following to the /var/log/messages:
kernel: cipcb: CIPE driver vers 20000702 (c) Olaf Titz 1996-2000, 100
channels, debug=1
kernel: cipcb: cipe_alloc_dev 0
kernel: Device not ready.  Make sure there is a disc in the drive.
???

heh, i doubt it has anything to do with cipcb, but heck it seems to popup
everytime i load cipcb.
weird.

Thanks,
Alex





<< | Thread Index | >> ]    [ << | Date Index | >> ]