>>>>> On Thu, 20 Oct 2005 14:23:59 -0500, Steve Greenland <[EMAIL PROTECTED]>
>>>>> said:
Steve> So I battled my way through AIX configure/build process[1] and got
Steve> bacula-fd built.
Steve> However, the FD won't bind a specified address. It gets stuck in the
Steve> bind() loop in bnet_server.c with EADDRNOTAVAIL, claiming that the
Steve> address is not available.
Steve> Now, I'm pretty sure that 127.0.0.1 is available, and I've tried
various
Steve> other addresses that were on the machine, no luck. I've run it under
Steve> the debugger to confirm that it is, in fact, attempting to bind to the
Steve> specified address. I modified the error message to include the
Steve> address like this:
Steve> berrno be;
Steve> struct sockaddr_in sa;
Steve> sa = *(struct sockaddr_in *) p->get_sockaddr();
Steve> if (tlog <= 0) {
Steve> tlog = 10; /* Complain every 10 seconds */
Steve> Emsg3(M_WARNING,0, _("Cannot bind port %s:%d: ERR=%s. Retrying
...\n"),
Steve> inet_ntoa(sa.sin_addr), ntohs(fd_ptr->port),
be.strerror());
Steve> }
Steve> And got the following message repeated:
Steve> Warning: Cannot bind port 127.0.0.1:9102: ERR=Can't assign
requested address. Retrying ...
Steve> Letting it bind to INADDR_ANY (i.e. not specifying FDAddress) works
Steve> fine, but isn't acceptable for my application.
Steve> I see in the list archives that someone else ran into this on Windows,
Steve> and people seemed to assume that it was the usual Windows brokeness.
Steve> However, I can guarantee that AIX supports binding to a particular
Steve> address, my code does it all the time. What's weird, of course, is that
Steve> I use pretty much the same code that bnet_server does, so no obvious
Steve> hint there.
Steve> Anybody else run into this and, hopefully, fixed it? I can't provide
Steve> access to this machine, but I'd be happy to run whatever
tests/debugging
Steve> I can do.
A common reason for this problem is that the padding bytes at the end of the
sockaddr_in object (called something like sin_zero) are not actually zero.
Typically this happens when the code fails to initialize them.
Unfortunately, there has been an outbreak of C++ in this code so it is
impossible to be sure whether it does this :-(
If you can look at the raw bytes passed to bind() in the debugger, then you
might be able to see what they are.
__Martin
-------------------------------------------------------
This SF.Net email is sponsored by:
Power Architecture Resource Center: Free content, downloads, discussions,
and more. http://solutions.newsforge.com/ibmarch.tmpl
_______________________________________________
Bacula-users mailing list
Bacula-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bacula-users
