Don't rely on the packet buffer (client->packet) being preserved between

attempts to send DISCOVER or REQUEST packets. Some DHCP servers might
NAK the DISCOVER, or other nefarious packets arrive, between attempts
and overwrite the packet being sent. Create and use another buffer
for packets being sent.

Problem encountered by fgsch@, who noticed that once the DISCOVER was
NAK'ed our dhclient sent a bunch of NAK's back to the server. Like it
thought they were DISCOVER messages.

ok beck@

1 files changed, 5 insertions, 5 deletions

diff --git a/sbin/dhclient/bpf.c b/sbin/dhclient/bpf.c
index 50f5a7814b5..900039da52e 100644
--- a/sbin/dhclient/bpf.c
+++ b/sbin/dhclient/bpf.c
@@ -1,4 +1,4 @@
-/*	$OpenBSD: bpf.c,v 1.25 2013/02/03 21:04:19 krw Exp $	*/
+/*	$OpenBSD: bpf.c,v 1.26 2013/02/14 20:39:46 krw Exp $	*/
 
 /* BPF socket interface code, originally contributed by Archie Cobbs. */
 
@@ -274,13 +274,13 @@ send_packet(struct in_addr from, struct sockaddr_in *to,
 
 	assemble_udp_ip_header(buf, &bufp, from.s_addr,
 	    to->sin_addr.s_addr, to->sin_port,
-	    (unsigned char *)&client->packet,
-	    client->packet_length);
+	    (unsigned char *)&client->bootrequest_packet,
+	    client->bootrequest_packet_length);
 
 	iov[0].iov_base = (char *)buf;
 	iov[0].iov_len = bufp;
-	iov[1].iov_base = (char *)&client->packet;
-	iov[1].iov_len = client->packet_length;
+	iov[1].iov_base = (char *)&client->bootrequest_packet;
+	iov[1].iov_len = client->bootrequest_packet_length;
 
 	if (to->sin_addr.s_addr == INADDR_BROADCAST) {
 		result = writev(ifi->wfdesc, iov, IOVCNT);