[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[GNUnet-SVN] r17702 - gnunet/src/core
From: |
gnunet |
Subject: |
[GNUnet-SVN] r17702 - gnunet/src/core |
Date: |
Mon, 24 Oct 2011 10:39:44 +0200 |
Author: grothoff
Date: 2011-10-24 10:39:44 +0200 (Mon, 24 Oct 2011)
New Revision: 17702
Modified:
gnunet/src/core/gnunet-service-core_clients.c
Log:
fixing 1845
Modified: gnunet/src/core/gnunet-service-core_clients.c
===================================================================
--- gnunet/src/core/gnunet-service-core_clients.c 2011-10-24 08:29:21 UTC
(rev 17701)
+++ gnunet/src/core/gnunet-service-core_clients.c 2011-10-24 08:39:44 UTC
(rev 17702)
@@ -443,9 +443,16 @@
tc.car = GNUNET_CONTAINER_multihashmap_get (c->requests,
&sm->peer.hashPubKey);
if (NULL == tc.car)
{
- /* client did not request transmission first! */
- GNUNET_break (0);
- GNUNET_SERVER_receive_done (client, GNUNET_SYSERR);
+ /* Must have been that we first approved the request, then got disconnected
+ (which triggered removal of the 'car') and now the client gives us a
message
+ just *before* the client learns about the disconnect. Theoretically, we
+ might also now be *again* connected. So this can happen (but should be
+ rare). If it does happen, the message is discarded. */
+ GNUNET_STATISTICS_update (GSC_stats,
+ gettext_noop ("# messages discarded (session
disconnected)"),
+ 1,
+ GNUNET_NO);
+ GNUNET_SERVER_receive_done (client, GNUNET_OK);
return;
}
GNUNET_assert (GNUNET_YES ==
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- [GNUnet-SVN] r17702 - gnunet/src/core,
gnunet <=