ike1: Flush active queue when queueing a delete of the IKE_SA

By aborting the active task we don't have to wait for potential
retransmits if the other peer does not respond to the current task.
Since IKEv1 has no sequential message IDs and INFORMATIONALs are no real
exchanges this should not be a problem.

Fixes #1537
References #429, #1410
Closes strongswan/strongswan#48

diff --git a/src/libcharon/sa/ikev1/task_manager_v1.c b/src/libcharon/sa/ikev1/task_manager_v1.c
index b0c4f5f849372295b2124fda8fd6f23759ab220b..117102bce73546f0bcdf870ba441a9d710ad466a 100644 (file)
--- a/src/libcharon/sa/ikev1/task_manager_v1.c
+++ b/src/libcharon/sa/ikev1/task_manager_v1.c
@@ -1661,6 +1661,9 @@ METHOD(task_manager_t, queue_ike_delete, void,
        enumerator_t *enumerator;
        child_sa_t *child_sa;
 
+       /* cancel any currently active task to get the DELETE done quickly */
+       flush_queue(this, TASK_QUEUE_ACTIVE);
+
        enumerator = this->ike_sa->create_child_sa_enumerator(this->ike_sa);
        while (enumerator->enumerate(enumerator, &child_sa))
        {