Currently on a migration target, we create then immediately bind() new
sockets for the TCP connections we're reconstructing. Mostly, this works,
since a socket() that is bound but hasn't had listen() or connect() called
is essentially passive. However, this bind() is subject to the usual
address conflict checking. In particular that means if we already have
a listening socket on that port, we'll get an EADDRINUSE. This will happen
for every connection we try to migrate that was initiated from outside to
the guest, since we necessarily created a listening socket for that case.
We set SO_REUSEADDR on the socket in an attempt to avoid this, but that's
not sufficient; even with SO_REUSEADDR address conflicts are still
prohibited for listening sockets. Of course once these incoming sockets
are fully repaired and connect()ed they'll no longer conflict, but that
doesn't help us if we fail at the bind().
We can avoid this by not calling bind() until we're already in repair mode
which suppresses this transient conflict. Because of the batching of
setting repair mode, to do that we need to move the bind to a step in
tcp_flow_migrate_target_ext().
Signed-off-by: David Gibson <david(a)gibson.dropbear.id.au>
---
tcp.c | 38 +++++++++++++++++++++++++++-----------
1 file changed, 27 insertions(+), 11 deletions(-)
diff --git a/tcp.c b/tcp.c
index fa1d8857..35626c91 100644
--- a/tcp.c
+++ b/tcp.c
@@ -3414,13 +3414,8 @@ fail:
static int tcp_flow_repair_socket(struct ctx *c, struct tcp_tap_conn *conn)
{
sa_family_t af = CONN_V4(conn) ? AF_INET : AF_INET6;
- const struct flowside *sockside = HOSTFLOW(conn);
- union sockaddr_inany a;
- socklen_t sl;
int s, rc;
- pif_sockaddr(c, &a, &sl, PIF_HOST, &sockside->oaddr,
sockside->oport);
-
if ((conn->sock = socket(af, SOCK_STREAM | SOCK_NONBLOCK | SOCK_CLOEXEC,
IPPROTO_TCP)) < 0) {
rc = -errno;
@@ -3435,12 +3430,6 @@ static int tcp_flow_repair_socket(struct ctx *c, struct
tcp_tap_conn *conn)
tcp_sock_set_nodelay(s);
- if (bind(s, &a.sa, sizeof(a))) {
- rc = -errno;
- flow_perror(conn, "Failed to bind socket for migrated flow");
- goto err;
- }
-
if ((rc = tcp_flow_repair_on(c, conn)))
goto err;
@@ -3452,6 +3441,30 @@ err:
return rc;
}
+/**
+ * tcp_flow_repair_bind() - Bind socket in repair mode
+ * @c: Execution context
+ * @conn: Pointer to the TCP connection structure
+ *
+ * Return: 0 on success, negative error code on failure
+ */
+static int tcp_flow_repair_bind(const struct ctx *c, struct tcp_tap_conn *conn)
+{
+ const struct flowside *sockside = HOSTFLOW(conn);
+ union sockaddr_inany a;
+ socklen_t sl;
+
+ pif_sockaddr(c, &a, &sl, PIF_HOST, &sockside->oaddr,
sockside->oport);
+
+ if (bind(conn->sock, &a.sa, sizeof(a))) {
+ int rc = -errno;
+ flow_perror(conn, "Failed to bind socket for migrated flow");
+ return rc;
+ }
+
+ return 0;
+}
+
/**
* tcp_flow_repair_connect() - Connect socket in repair mode, then turn it off
* @c: Execution context
@@ -3618,6 +3631,9 @@ int tcp_flow_migrate_target_ext(struct ctx *c, struct tcp_tap_conn
*conn, int fd
/* We weren't able to create the socket, discard flow */
goto fail;
+ if (tcp_flow_repair_bind(c, conn))
+ goto fail;
+
if (tcp_flow_repair_timestamp(conn, &t))
goto fail;
--
2.49.0