[Openswan dev] [PATCH] Fix race condition between pluto start and whack

[Mattias Walstrom]

Fix race condition between pluto start and whack, which can result in pluto becoming a zombie that never returns.

When sending any whack message during the time pluto starts, whack will just hang for a long time.
And if pluto should die during this hang, it will become a Zombie. This patch makes sure that we do not open the ctl socket until we actually can receive messages (all pluto initialization is done).

Signed-off-by: Mattias Walström <lazzer at vmlinux.org>

--
 plutomain.c |   26 +++++++++++---------------
 1 file changed, 11 insertions(+), 15 deletions(-)

diff --git a/programs/pluto/plutomain.c b/programs/pluto/plutomain.c
index 1597dc4..0d19d2c 100644
--- a/programs/pluto/plutomain.c
+++ b/programs/pluto/plutomain.c
@@ -725,21 +725,6 @@ main(int argc, char **argv)
 #endif
 #endif
 
-    /** create control socket.
-     * We must create it before the parent process returns so that
-     * there will be no race condition in using it.  The easiest
-     * place to do this is before the daemon fork.
-     */
-    {
-	err_t ugh = init_ctl_socket();
-
-	if (ugh != NULL)
-	{
-	    fprintf(stderr, "pluto: %s", ugh);
-	    exit_pluto(1);
-	}
-    }
-
 #ifdef IPSECPOLICY
     /* create info socket. */
     {
@@ -1051,6 +1036,17 @@ main(int argc, char **argv)
     load_authcerts_from_nss("CA cert",  AUTH_CA);
 #endif
 
+    /* create control socket. */
+    {
+	err_t ugh = init_ctl_socket();
+
+	if (ugh != NULL)
+	{
+	    fprintf(stderr, "pluto: %s", ugh);
+	    exit_pluto(1);
+	}
+    }
+
     daily_log_event();
     call_server();
     return -1;	/* Shouldn't ever reach this */