Date: Tue, 23 Nov 1999 21:26:19 +0100 From: Eric Masson <e-masson@kisoft-services.com> To: Giorgos Keramidas <keramida@ceid.upatras.gr> Cc: freebsd-questions@FreeBSD.ORG Subject: Re: ktrace/kdump output interpreting, help needed. Message-ID: <383AF86B.A0FA3CF3@kisoft-services.com> References: <383967E9.E8A2936@kisoft-services.com> <86wvranegk.fsf@localhost.hell.gr>
next in thread | previous in thread | raw e-mail | index | archive | help
Giorgos Keramidas a écrit :
>
> Eric Masson <e-masson@kisoft-services.com> writes:
>
> A few comments on the code before the real answer a few lines below ;)
>
> > sock_id=rexec(ahost,port,user,password,cmd,&fd2p);
>
> >From the manual page of rexec(3) we read:
>
> The port inport specifies which well-known DARPA Internet port to
> use for the connection; the call `getservbyname("exec", "tcp")' (see
> getservent(3)) will return a pointer to a structure, which contains
> the necessary port.
>
> In /etc/services we can read that 512 is indeed the rexec port. So
> you're right on using port 512, but to be as portable as one can be, try
> to follow the manuals more closely. Don't mind the few extra lines of
> code that are destined to eventually prove very useful when a weird set
> of circumstances comes forth.
>
> > buffer = (void *)malloc(BUF_SIZE);
>
> What if this fails for some obscure reason? I would really prefer it
> written like:
>
> if ((buffer = malloc(BUF_SIZE*sizeof(char))) == NULL)
> err(0, "whoops! memory exchausted");
>
> > while(count = read(sock_id,buffer,BUF_SIZE)) {
>
> The read() system call can return a negative number too, in case of
> failure, which will pass your while() test, and have the for loop that
> follows it barf on unread and probably broken data. Try using:
>
> while ((count = read(sock_id, buffer, BUF_SIZE)) > 0)
> ... blah ...
>
Right, I should have said that this code snippet was intented for test
purposes only :), Thanks for improvements.
> > :
> >
> > 403 rexec CALL nanosleep(0xbfbfdb28,0xbfbfdb20)
> > 403 rexec RET nanosleep 0
> > 403 rexec CALL socket(0x2,0x1,0)
> > 403 rexec RET socket 3
> > 403 rexec CALL connect(0x3,0xbfbfdb94,0x10)
> > 403 rexec RET connect -1 errno 61 Connection refused
>
> "Connection refused" this says a lot. Probably rexec is not allowed
> from your host to the one you're trying to connect to.
>
> [large snip]
>
> > 403 rexec CALL connect(0x3,0xbfbfdb94,0x10)
> > 403 rexec RET connect -1 errno 61 Connection refused
>
> Ditto. You are refused the attempted connection. A few lines below you
> can see your rexec retrying to establish the connection, but you're
> always refused from the destination host. Finally, the error message is
> very informative of why you failed to get any meaningful output:
>
> > 403 rexec CALL writev(0x2,0xbfbfdb14,0x4)
> > 403 rexec GIO fd 2 wrote 60 bytes
> > "srvas4nanssv.nantes.kisoft-services.com: Connection refused
> > "
>
> Regards.
Right, twice. I know rexec says connection is denied, I submitted this
just because when i telnet the host on port 512, it doesn't deny
connection. So I've done the test on my FBSD workstation after i have
enabled rexecd in inetd.conf ( and of course sent a HUP to inetd). Same
result, connection refused. That's why I posted ;).
Regards
Eric
>
> --
> Giorgos Keramidas, <keramida@ceid.upatras.gr>
> "What we have to learn to do, we learn by doing." [Aristotle]
>
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with "unsubscribe freebsd-questions" in the body of the message
--
Any opinions expressed above | Murphy's Law Corollary :
are my own, not Kisoft's | Murphy was an optimist.
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?383AF86B.A0FA3CF3>