summaryrefslogtreecommitdiffstats
path: root/net/sctp/sm_sideeffect.c
diff options
context:
space:
mode:
authorThomas Graf <tgraf@infradead.org>2011-07-07 02:28:35 +0200
committerDavid S. Miller <davem@davemloft.net>2011-07-07 23:08:44 +0200
commitf8d9605243280f1870dd2c6c37a735b925c15f3c (patch)
tree2d6a3ce33c503bce8fca71489d4c4dc266579469 /net/sctp/sm_sideeffect.c
parentMerge branch 'pm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/rafa... (diff)
downloadlinux-f8d9605243280f1870dd2c6c37a735b925c15f3c.tar.xz
linux-f8d9605243280f1870dd2c6c37a735b925c15f3c.zip
sctp: Enforce retransmission limit during shutdown
When initiating a graceful shutdown while having data chunks on the retransmission queue with a peer which is in zero window mode the shutdown is never completed because the retransmission error count is reset periodically by the following two rules: - Do not timeout association while doing zero window probe. - Reset overall error count when a heartbeat request has been acknowledged. The graceful shutdown will wait for all outstanding TSN to be acknowledged before sending the SHUTDOWN request. This never happens due to the peer's zero window not acknowledging the continuously retransmitted data chunks. Although the error counter is incremented for each failed retransmission, the receiving of the SACK announcing the zero window clears the error count again immediately. Also heartbeat requests continue to be sent periodically. The peer acknowledges these requests causing the error counter to be reset as well. This patch changes behaviour to only reset the overall error counter for the above rules while not in shutdown. After reaching the maximum number of retransmission attempts, the T5 shutdown guard timer is scheduled to give the receiver some additional time to recover. The timer is stopped as soon as the receiver acknowledges any data. The issue can be easily reproduced by establishing a sctp association over the loopback device, constantly queueing data at the sender while not reading any at the receiver. Wait for the window to reach zero, then initiate a shutdown by killing both processes simultaneously. The association will never be freed and the chunks on the retransmission queue will be retransmitted indefinitely. Signed-off-by: Thomas Graf <tgraf@infradead.org> Acked-by: Vlad Yasevich <vladislav.yasevich@hp.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/sctp/sm_sideeffect.c')
-rw-r--r--net/sctp/sm_sideeffect.c20
1 files changed, 18 insertions, 2 deletions
diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c
index 534c2e5feb05..6e0f88295aaf 100644
--- a/net/sctp/sm_sideeffect.c
+++ b/net/sctp/sm_sideeffect.c
@@ -670,10 +670,19 @@ static void sctp_cmd_transport_on(sctp_cmd_seq_t *cmds,
/* 8.3 Upon the receipt of the HEARTBEAT ACK, the sender of the
* HEARTBEAT should clear the error counter of the destination
* transport address to which the HEARTBEAT was sent.
- * The association's overall error count is also cleared.
*/
t->error_count = 0;
- t->asoc->overall_error_count = 0;
+
+ /*
+ * Although RFC4960 specifies that the overall error count must
+ * be cleared when a HEARTBEAT ACK is received, we make an
+ * exception while in SHUTDOWN PENDING. If the peer keeps its
+ * window shut forever, we may never be able to transmit our
+ * outstanding data and rely on the retransmission limit be reached
+ * to shutdown the association.
+ */
+ if (t->asoc->state != SCTP_STATE_SHUTDOWN_PENDING)
+ t->asoc->overall_error_count = 0;
/* Clear the hb_sent flag to signal that we had a good
* acknowledgement.
@@ -1437,6 +1446,13 @@ static int sctp_cmd_interpreter(sctp_event_t event_type,
sctp_cmd_setup_t2(commands, asoc, cmd->obj.ptr);
break;
+ case SCTP_CMD_TIMER_START_ONCE:
+ timer = &asoc->timers[cmd->obj.to];
+
+ if (timer_pending(timer))
+ break;
+ /* fall through */
+
case SCTP_CMD_TIMER_START:
timer = &asoc->timers[cmd->obj.to];
timeout = asoc->timeouts[cmd->obj.to];