The OpenFlow 1.0 specification says:
Emergency flow mod messages must have timeout values set to zero.
Otherwise, the switch must refuse the addition and respond with an
ofp_error_msg with OFPET_FLOW_MOD_FAILED type and
OFPFMFC_BAD_EMERG_TIMEOUT code.
but Open vSwitch reported OFPFMFC_TABLE_FULL in this case. This commit
fixes the problem.
Fixes detailed_contr_sw_messages.EmerFlowTimeout failure in OFTest.
Signed-off-by: Ben Pfaff <b...@nicira.com>
---
ofproto/ofproto.c | 10 +++++++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/ofproto/ofproto.c b/ofproto/ofproto.c
index 2fb2fc8..444717e 100644
--- a/ofproto/ofproto.c
+++ b/ofproto/ofproto.c
@@ -3302,9 +3302,13 @@ handle_flow_mod(struct ofconn *ofconn, const struct
ofp_header *oh)
if (fm.flags & OFPFF10_EMERG) {
/* We do not support the OpenFlow 1.0 emergency flow cache, which
* is not required in OpenFlow 1.0.1 and removed from OpenFlow 1.1.
- * There is no good error code, so just state that the flow table
- * is full. */
- error = OFPERR_OFPFMFC_TABLE_FULL;
+ *
+ * OpenFlow 1.0 specifies the error code to use when idle_timeout or
+ * hard_timeout is nonzero. Otherwise, there is no good error code, so
+ * just state that the flow table is full. */
+ error = (fm.hard_timeout || fm.idle_timeout
+ ? OFPERR_OFPFMFC_BAD_EMERG_TIMEOUT
+ : OFPERR_OFPFMFC_TABLE_FULL);
}
if (!error) {
error = ofpacts_check(fm.ofpacts, fm.ofpacts_len,
--
1.7.10.4
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
