We use NXTRequest for Switch-to-Controller messages so let's rename it
to an appropriate name.
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Reviewed-by: Isaku Yamahata <yamahata@valinux.co.jp>
NXT_ROLE_REQUEST/REPLY use the same on-wire format (NX_ROLE_PACK_STR).
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Reviewed-by: Isaku Yamahata <yamahata@valinux.co.jp>
OFPStatsReply needs to handle variable-length bodys so having two
parsers (parser_stats_body_array and parser_stats_body) doesn't make
sense. Just kill parser_stats_body_array().
cls_stats_body_cls_size is also pointless since OFPStatsReply needs to
handle variable-length bodys. OFPStatsReply class needs to know if a
stats class is array or not. So register_stats_type takes
'body_single_struct' instead of body_cls_size. We need to change this
scheme if we need to handle VendorStats in the same way (both array
and single struct). But currently we don't even have any VendorStats
implementation so let's invent something more complicated when it
becomes necessary.
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Reviewed-by: Isaku Yamahata <yamahata@valinux.co.jp>
OFPStatsReply parser_stats_body_array needs body classes to have
length field.
We could set length as class-wide constants for classes having a fixed
length however we have two classes that need a variable length so
let's set the length in the same way for all the stats classes.
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
- __str__ method: needs to pass a tuple to msg_str_attr
- parser method: rate shouldn't be a tuple.
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
ryu crashes when it gets OFPT_STATS_REPLY including OFP actions.
ofp_flow_stats can include ofp_action_header(s) so
parser_stats_body_array method can't use the entry_size argument (the
minimum size, 88 bytes). It needs to use the size that body_cls's
parser returns.
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
frame data can contain padding of ofp packet. Discard padding.
Otherwise packet parser can get confused which can rely on frame length.
This is a preparation for LLDP TLV parser.
Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
0xffffffff == -1 as uint32_t.
However python integer representation isn't sized integer.
So assert buffer_id == -1 doesn't hold. use 0xffffffff explicitly instead
of -1.
Traceback (most recent call last):
File "/usr/lib/pymodules/python2.7/gevent/greenlet.py", line 390, in run
result = self._run(*self.args, **self.kwargs)
File "/ryu/bin/ryu/app/discovery.py", line 421, in lldp_loop
self.send_lldp_packet(dp, port_no)
File "/ryu/bin/ryu/app/discovery.py", line 390, in send_lldp_packet
dp.send_packet_out(actions=actions, data=port_data.data)
File "/ryu/bin/ryu/controller/controller.py", line 211, in send_packet_out
self.send_msg(packet_out)
File "/ryu/bin/ryu/controller/controller.py", line 181, in send_msg
self.serialize_msg(msg)
File "/ryu/bin/ryu/controller/controller.py", line 178, in serialize_msg
msg.serialize()
File "/ryu/bin/ryu/ofproto/ofproto_parser.py", line 119, in serialize
self._serialize_body()
File "/ryu/bin/ryu/ofproto/ofproto_v1_0_parser.py", line 1210, in _serialize_body
assert self.buffer_id == -1
AssertionError
Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
The step to create quantum db was missing.
Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Use '6s' for a mac address. It's consistent with nx code and works
with mac.haddr_to_bin helper function.
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Reviewed-by: Simon Horman <horms@verge.net.au>
OFPActionHeader class should have 'serialize' method.
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Reviewed-by: Simon Horman <horms@verge.net.au>
format is defined in builtin, so it should be avoid for argument.
use flow_format instead.
Signed-off-by: Isaku Yamahata <yamahata@valinux.co.jp>
Reviewed-by: Simon Horman <simon@horms.net>
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
