public inbox for passt-dev@passt.top
 help / color / mirror / code / Atom feed
From: David Gibson <david@gibson.dropbear.id.au>
To: passt-dev@passt.top, Stefano Brivio <sbrivio@redhat.com>
Cc: David Gibson <david@gibson.dropbear.id.au>
Subject: [PATCH 11/16] conf: Treat --dns addresses as guest visible addresses
Date: Wed, 14 Aug 2024 14:30:45 +1000	[thread overview]
Message-ID: <20240814043050.4177037-12-david@gibson.dropbear.id.au> (raw)
In-Reply-To: <20240814043050.4177037-1-david@gibson.dropbear.id.au>

Although it's not 100% explicit in the man page, addresses given to the
--dns option are intended to be addresses as seen by the guest.  This
differs from addresses taken from the host's /etc/resolv.conf, which must
be translated to to guest accessible versions in some cases.

Our implementation is currently inconsistent on this: when using
--dns-forward, you must usually also give --dns with the matching address,
which is meaningful only in the guest's address view.  However if you give
--dns with a loopback addres, it will be translated like a host view
address.

Move the remapping logic for DNS addresses out of add_dns4() and add_dns6()
into add_dns_resolv() so that it is only applied for host nameserver
addresses, not for nameservers given explicitly with --dns.

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
---
 conf.c  | 84 ++++++++++++++++++++++++++++-----------------------------
 passt.1 | 14 ++++++----
 2 files changed, 50 insertions(+), 48 deletions(-)

diff --git a/conf.c b/conf.c
index cb462b09..fa6d90ef 100644
--- a/conf.c
+++ b/conf.c
@@ -361,29 +361,11 @@ bind_all_fail:
 static unsigned add_dns4(struct ctx *c, const struct in_addr *addr,
 			 unsigned idx)
 {
-	unsigned added = 0;
-
 	if (idx >= ARRAY_SIZE(c->ip4.dns))
 		return 0;
 
-	/* Guest or container can only access local addresses via redirect */
-	if (IN4_IS_ADDR_LOOPBACK(addr)) {
-		if (!c->no_map_gw) {
-			c->ip4.dns[idx] = c->ip4.gw;
-			added++;
-
-			if (IN4_IS_ADDR_UNSPECIFIED(&c->ip4.dns_match))
-				c->ip4.dns_match = c->ip4.gw;
-		}
-	} else {
-		c->ip4.dns[idx] = *addr;
-		added++;
-	}
-
-	if (IN4_IS_ADDR_UNSPECIFIED(&c->ip4.dns_host))
-		c->ip4.dns_host = *addr;
-
-	return added;
+	c->ip4.dns[idx] = *addr;
+	return 1;
 }
 
 /**
@@ -394,31 +376,14 @@ static unsigned add_dns4(struct ctx *c, const struct in_addr *addr,
  *
  * Return: Number of entries added (0 or 1)
  */
-static unsigned add_dns6(struct ctx *c, struct in6_addr *addr, unsigned idx)
+static unsigned add_dns6(struct ctx *c, const struct in6_addr *addr,
+			 unsigned idx)
 {
-	unsigned added = 0;
-
 	if (idx >= ARRAY_SIZE(c->ip6.dns))
 		return 0;
 
-	/* Guest or container can only access local addresses via redirect */
-	if (IN6_IS_ADDR_LOOPBACK(addr)) {
-		if (!c->no_map_gw) {
-			c->ip6.dns[idx] = c->ip6.gw;
-			added++;
-
-			if (IN6_IS_ADDR_UNSPECIFIED(&c->ip6.dns_match))
-				c->ip6.dns_match = c->ip6.gw;
-		}
-	} else {
-		c->ip6.dns[idx] = *addr;
-		added++;
-	}
-
-	if (IN6_IS_ADDR_UNSPECIFIED(&c->ip6.dns_host))
-		c->ip6.dns_host = *addr;
-
-	return added;
+	c->ip6.dns[idx] = *addr;
+	return 1;
 }
 
 /**
@@ -437,11 +402,44 @@ static void add_dns_resolv(struct ctx *c, const char *nameserver,
 	struct in6_addr ns6;
 	struct in_addr ns4;
 
-	if (idx4 && inet_pton(AF_INET, nameserver, &ns4))
+	if (idx4 && inet_pton(AF_INET, nameserver, &ns4)) {
+		if (IN4_IS_ADDR_UNSPECIFIED(&c->ip4.dns_host))
+			c->ip4.dns_host = ns4;
+
+		/* Guest or container can only access local addresses via
+		 * redirect
+		 */
+		if (IN4_IS_ADDR_LOOPBACK(&ns4)) {
+			if (c->no_map_gw)
+				return;
+
+			ns4 = c->ip4.gw;
+			if (IN4_IS_ADDR_UNSPECIFIED(&c->ip4.dns_match))
+				c->ip4.dns_match = c->ip4.gw;
+		}
+
 		*idx4 += add_dns4(c, &ns4, *idx4);
+	}
+
+	if (idx6 && inet_pton(AF_INET6, nameserver, &ns6)) {
+		if (IN6_IS_ADDR_UNSPECIFIED(&c->ip6.dns_host))
+			c->ip6.dns_host = ns6;
+
+		/* Guest or container can only access local addresses via
+		 * redirect
+		 */
+		if (IN6_IS_ADDR_LOOPBACK(&ns6)) {
+			if (c->no_map_gw)
+				return;
+
+			ns6 = c->ip6.gw;
+
+			if (IN6_IS_ADDR_UNSPECIFIED(&c->ip6.dns_match))
+				c->ip6.dns_match = c->ip6.gw;
+		}
 
-	if (idx6 && inet_pton(AF_INET6, nameserver, &ns6))
 		*idx6 += add_dns6(c, &ns6, *idx6);
+	}
 }
 
 /**
diff --git a/passt.1 b/passt.1
index 3062b719..dca433b6 100644
--- a/passt.1
+++ b/passt.1
@@ -236,11 +236,15 @@ interface will be chosen instead.
 
 .TP
 .BR \-D ", " \-\-dns " " \fIaddr
-Use \fIaddr\fR (IPv4 or IPv6) for DHCP, DHCPv6, NDP or DNS forwarding, as
-configured (see options \fB--no-dhcp-dns\fR, \fB--dhcp-dns\fR,
-\fB--dns-forward\fR) instead of reading addresses from \fI/etc/resolv.conf\fR.
-This option can be specified multiple times.  Specifying \fB-D none\fR disables
-usage of DNS addresses altogether.
+Instruct the guest (via DHCP, DHVPv6 or NDP) to use \fIaddr\fR (IPv4
+or IPv6) as a nameserver, as configured (see options
+\fB--no-dhcp-dns\fR, \fB--dhcp-dns\fR) instead of reading addresses
+from \fI/etc/resolv.conf\fR.  This option can be specified multiple
+times.  Specifying \fB-D none\fR disables usage of DNS addresses
+altogether.  Unlike addresses from \fI/etc/resolv.conf\fR, \fIaddr\fR
+is given to the guest without remapping.  For example \fB--dns
+127.0.0.1\fR will instruct the guest to use itself as nameserver, not
+the host.
 
 .TP
 .BR \-\-dns-forward " " \fIaddr
-- 
@@ -236,11 +236,15 @@ interface will be chosen instead.
 
 .TP
 .BR \-D ", " \-\-dns " " \fIaddr
-Use \fIaddr\fR (IPv4 or IPv6) for DHCP, DHCPv6, NDP or DNS forwarding, as
-configured (see options \fB--no-dhcp-dns\fR, \fB--dhcp-dns\fR,
-\fB--dns-forward\fR) instead of reading addresses from \fI/etc/resolv.conf\fR.
-This option can be specified multiple times.  Specifying \fB-D none\fR disables
-usage of DNS addresses altogether.
+Instruct the guest (via DHCP, DHVPv6 or NDP) to use \fIaddr\fR (IPv4
+or IPv6) as a nameserver, as configured (see options
+\fB--no-dhcp-dns\fR, \fB--dhcp-dns\fR) instead of reading addresses
+from \fI/etc/resolv.conf\fR.  This option can be specified multiple
+times.  Specifying \fB-D none\fR disables usage of DNS addresses
+altogether.  Unlike addresses from \fI/etc/resolv.conf\fR, \fIaddr\fR
+is given to the guest without remapping.  For example \fB--dns
+127.0.0.1\fR will instruct the guest to use itself as nameserver, not
+the host.
 
 .TP
 .BR \-\-dns-forward " " \fIaddr
-- 
2.46.0


  parent reply	other threads:[~2024-08-14  4:31 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-08-14  4:30 [PATCH 00/16] Cleanups and fixes related to "our" addresses David Gibson
2024-08-14  4:30 ` [PATCH 01/16] conf: Don't ignore -t and -u options after -D David Gibson
2024-08-14 12:18   ` Stefano Brivio
2024-08-14  4:30 ` [PATCH 02/16] treewide: Use "our address" instead of "forwarding address" David Gibson
2024-08-14  4:30 ` [PATCH 03/16] util: Helper for formatting MAC addresses David Gibson
2024-08-14  4:30 ` [PATCH 04/16] treewide: Rename MAC address fields for clarity David Gibson
2024-08-14  4:30 ` [PATCH 05/16] treewide: Use struct assignment instead of memcpy() for IP addresses David Gibson
2024-08-14  4:30 ` [PATCH 06/16] conf: Use array indices rather than pointers for DNS array slots David Gibson
2024-08-14  4:30 ` [PATCH 07/16] conf: More accurately count entries added in get_dns() David Gibson
2024-08-14  4:30 ` [PATCH 08/16] conf: Move DNS array bounds checks into add_dns[46] David Gibson
2024-08-14  4:30 ` [PATCH 09/16] conf: Move adding of a nameserver from resolv.conf into subfunction David Gibson
2024-08-14  4:30 ` [PATCH 10/16] conf: Correct setting of dns_match address in add_dns6() David Gibson
2024-08-14  4:30 ` David Gibson [this message]
2024-08-14  4:30 ` [PATCH 12/16] conf: Remove incorrect initialisation of addr_ll_seen David Gibson
2024-08-14  4:30 ` [PATCH 13/16] util: Correct sock_l4() binding for link local addresses David Gibson
2024-08-14  4:30 ` [PATCH 14/16] treewide: Change misleading 'addr_ll' name David Gibson
2024-08-14  4:30 ` [PATCH 15/16] Clarify which addresses in ip[46]_ctx are meaningful where David Gibson
2024-08-14  4:30 ` [PATCH 16/16] Initialise our_tap_ll to ip6.gw when suitable David Gibson
2024-08-14  4:39   ` David Gibson
2024-08-14  5:03 ` [PATCH 00/16] Cleanups and fixes related to "our" addresses David Gibson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240814043050.4177037-12-david@gibson.dropbear.id.au \
    --to=david@gibson.dropbear.id.au \
    --cc=passt-dev@passt.top \
    --cc=sbrivio@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this public inbox

	https://passt.top/passt

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for IMAP folder(s).