Date: Thu, 9 Jan 1997 16:20:47 +0100 (MET) From: grog@lemis.de To: peter@spinner.DIALix.COM (Peter Wemm) Cc: FreeBSD-current@FreeBSD.ORG (FreeBSD current users) Subject: Re: What's happened to nfsd and mountd? Message-ID: <199701091520.QAA03172@freebie.lemis.de> In-Reply-To: <199701091502.XAA13553@spinner.DIALix.COM> from Peter Wemm at "Jan 9, 97 11:02:49 pm"
next in thread | previous in thread | raw e-mail | index | archive | help
Peter Wemm writes:
> Greg Lehey wrote:
>> A pointer to the problem, by the look of it:
>>
>> === grog@freebie (/dev/ttypa) ~/src 3 -> gma
>> get_myaddress() returns 0
>> sin_family = 0 (AF_INET = 2)
>> sin_len = 3 (16)
>> sin_port = 0
>> sin_addr = 0.32.0.0
>> SIOCFIGCONF used 1008 bytes of a buffer 1024 long
>
> YIKES! SIOCGIFCONF might be screwing up by the look of it. Either that
> or you
> have records being returned that are not sizeof(ifreq) in length and the
> code that is supposed to track these isn't working. There's 16 spare
> bytes coming from somewhere..
>
> I think get_myaddress() has failed, and didn't return a failure code, so
> the uninitialised data is being used..
No, I've got further than that. There's a bug in get_myaddress, line 91:
if ((ifreq.ifr_flags & IFF_UP) &&
ifr->ifr_addr.sa_family == AF_INET &&
(loopback == 1 && (ifreq.ifr_flags & IFF_LOOPBACK))) { <--- here
*addr = *((struct sockaddr_in *)&ifr->ifr_addr);
addr->sin_port = htons(PMAPPORT);
gotit = 1;
break;
}
This means that it can't find anything in the first pass (where
loopback is 0), and it always takes the loopback interface in the
second pass. I've changed this to:
if ((ifreq.ifr_flags & IFF_UP) &&
ifr->ifr_addr.sa_family == AF_INET &&
(loopback == 1 || ! (ifreq.ifr_flags & IFF_LOOPBACK))) {
Now gma returns:
=== grog@freebie (/dev/ttyp5) ~/src 16 -> gma
get_myaddress() returns 0
sin_family = 2 (AF_INET = 2)
sin_len = 16 (16)
sin_port = 111
sin_addr = 192.109.197.137
SIOCFIGCONF used 1008 bytes of a buffer 1024 long
Unfortunately, mountd still fails (and draws portmap in for sympathy)
with a message which completely baffles me:
Jan 9 12:08:28 freebie portmap[754]: connect from 127.0.0.1 to unset(mountd): request from non-local host
Jan 9 12:08:39 freebie portmap[759]: connect from 127.0.0.1 to unset(mountd): request from non-local host
Jan 9 12:08:39 freebie mountd[749]: Can't register mount
Jan 9 12:08:39 freebie portmap[760]: connect from 127.0.0.1 to set(mountd): request from non-local host
Jan 9 12:10:29 freebie portmap[803]: connect from 127.0.0.1 to unset(mountd): request from non-local host
Jan 9 12:10:29 freebie portmap[804]: connect from 127.0.0.1 to unset(mountd): request from non-local host
Jan 9 12:10:29 freebie mountd[802]: Can't register mount
Jan 9 12:10:29 freebie portmap[805]: connect from 127.0.0.1 to set(mountd): request from non-local host
I've established (with gdb and mountd -d) that it does (now) find the
correct interface (which should be obvious form the sin_addr reported
by gma), but these messages look like they should have come from the
old version, not from this one. Unfortunately, I'm off on a business
trip for 2 weeks tomorrow, and I don't have time to follow through any
further.
>>> One of my systems nearby uses 992 bytes of the SIOCGIFCONF 1024 byte buffer
>>> (as used in get_myaddress). It would be interesting to know what happens
>>> if it is overflowing on the systems that are failing.
>>
>> I'd guess it's probably the wrong family.
>
> Yes. get_myaddress() is returning a family of 0 (AF_UNSPEC), rather than
> AF_INET, so it's no real suprise that it can't send to that address over a
> UDP socket.
>
> Can you give me the result of 'netstat -in' please? I'm curious to see if
> you have large <Link> entries or something on the isdn interfaces.
OK, here goes, though again I don't think this is the problem.
=== root@freebie (/dev/ttyp6) /home/grog 18 -> netstat -in
Name Mtu Network Address Ipkts Ierrs Opkts Oerrs Coll
lp0* 1500 <Link> 0 0 0 0 0
ep0 1500 <Link> 00.a0.24.37.0d.2b 23066 0 17860 1 0
ep0 1500 192.109.197 192.109.197.137 23066 0 17860 1 0
ipi0 1500 <Link> 0 0 0 0 0
ipi0 1500 192.109.197.1 192.109.197.137 0 0 0 0 0
ipi1 1500 <Link> 166 0 208 7 0
ipi1 1500 192.109.197 192.109.197.137 166 0 208 7 0
ipi2 1500 <Link> 0 0 0 0 0
ipi2 1500 192.109.197.1 192.109.197.137 0 0 0 0 0
ipi3 1500 <Link> 0 0 0 0 0
ipi3 1500 192.109.197 192.109.197.37 0 0 0 0 0
ipi4 1500 <Link> 0 0 0 0 0
ipi4 1500 192.109.197 192.109.197.148 0 0 0 0 0
ipi5 1500 <Link> 0 0 0 0 0
ipi5 1500 192.109.197.1 192.109.197.137 0 0 0 0 0
ipi6 1500 <Link> 0 0 0 0 0
ipi6 1500 192.109.197 192.109.197.137 0 0 0 0 0
ipi7* 1500 <Link> 0 0 0 0 0
ipi8* 1500 <Link> 0 0 0 0 0
ipi9 1500 <Link> 413 0 514 0 0
ipi9 1500 194.97.201 194.97.201.66 413 0 514 0 0
ipi10 1500 <Link> 0 0 0 0 0
ipi11 1500 <Link> 0 0 0 0 0
ipi12 1500 <Link> 0 0 0 0 0
ipi13 1500 <Link> 0 0 0 0 0
ipi14 1500 <Link> 0 0 0 0 0
ipi15 1500 <Link> 0 0 0 0 0
tun0* 1500 <Link> 0 0 0 0 0
sl0* 552 <Link> 0 0 0 0 0
ppp0* 1500 <Link> 0 0 0 0 0
ppp1* 1500 <Link> 0 0 0 0 0
lo0 16384 <Link> 1465 0 1465 0 0
lo0 16384 127 127.0.0.1 1465 0 1465 0 0
I've changed my name server config, and at some time in the next 24
hours my lowest MX will point to freefall. When that happens, I may
not get any messages for up to 48 hours. Could you please copy
grog@freebie.lemis.de on any message you send before Friday the 10th,
1000 UTC?
Thanks in advance
Greg
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199701091520.QAA03172>