aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorVlad Yasevich <vladislav.yasevich@hp.com>2009-08-26 09:36:25 -0400
committerVlad Yasevich <vladislav.yasevich@hp.com>2009-09-04 18:20:58 -0400
commitb9f8478682445c2a3e0b87718a0563ef543ad94e (patch)
tree9bc0351c0a329a0780c6f2f7291e9207727de686
parentd71a09ed555e52299b5d3ad8945bdf84f65423a6 (diff)
sctp: Fix error count increments that were results of HEARTBEATS
SCTP RFC 4960 states that unacknowledged HEARTBEATS count as errors agains a given transport or endpoint. As such, we should increment the error counts for only for unacknowledged HB, otherwise we detect failure too soon. This goes for both the overall error count and the path error count. Now, there is a difference in how the detection is done between the two. The path error detection is done after the increment, so to detect it properly, we actually need to exceed the path threshold. The overall error detection is done _BEFORE_ the increment. Thus to detect the failure, it's enough for the error count to match the threshold. This is why all the state functions use '>=' to detect failure, while path detection uses '>'. Thanks goes to Chunbo Luo <chunbo.luo@windriver.com> who first proposed patches to fix this issue and made me re-read the spec and the code to figure out how this cruft really works. Signed-off-by: Vlad Yasevich <vladislav.yasevich@hp.com>
-rw-r--r--net/sctp/sm_sideeffect.c20
-rw-r--r--net/sctp/sm_statefuns.c2
2 files changed, 17 insertions, 5 deletions
diff --git a/net/sctp/sm_sideeffect.c b/net/sctp/sm_sideeffect.c
index 238adf7978e..694f7491731 100644
--- a/net/sctp/sm_sideeffect.c
+++ b/net/sctp/sm_sideeffect.c
@@ -440,14 +440,26 @@ static void sctp_do_8_2_transport_strike(struct sctp_association *asoc,
/* The check for association's overall error counter exceeding the
* threshold is done in the state function.
*/
- /* When probing UNCONFIRMED addresses, the association overall
- * error count is NOT incremented
+ /* We are here due to a timer expiration. If the timer was
+ * not a HEARTBEAT, then normal error tracking is done.
+ * If the timer was a heartbeat, we only increment error counts
+ * when we already have an outstanding HEARTBEAT that has not
+ * been acknowledged.
+ * Additionaly, some tranport states inhibit error increments.
*/
- if (transport->state != SCTP_UNCONFIRMED)
+ if (!is_hb) {
asoc->overall_error_count++;
+ if (transport->state != SCTP_INACTIVE)
+ transport->error_count++;
+ } else if (transport->hb_sent) {
+ if (transport->state != SCTP_UNCONFIRMED)
+ asoc->overall_error_count++;
+ if (transport->state != SCTP_INACTIVE)
+ transport->error_count++;
+ }
if (transport->state != SCTP_INACTIVE &&
- (transport->error_count++ >= transport->pathmaxrxt)) {
+ (transport->error_count > transport->pathmaxrxt)) {
SCTP_DEBUG_PRINTK_IPADDR("transport_strike:association %p",
" transport IP: port:%d failed.\n",
asoc,
diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c
index 7fb08a69491..45b8bcafd82 100644
--- a/net/sctp/sm_statefuns.c
+++ b/net/sctp/sm_statefuns.c
@@ -971,7 +971,7 @@ sctp_disposition_t sctp_sf_sendbeat_8_3(const struct sctp_endpoint *ep,
{
struct sctp_transport *transport = (struct sctp_transport *) arg;
- if (asoc->overall_error_count > asoc->max_retrans) {
+ if (asoc->overall_error_count >= asoc->max_retrans) {
sctp_add_cmd_sf(commands, SCTP_CMD_SET_SK_ERR,
SCTP_ERROR(ETIMEDOUT));
/* CMD_ASSOC_FAILED calls CMD_DELETE_TCB. */