2 ------------------------
4 - Added ability to match on IPv6 flow label through NXM.
6 - New "fdb/flush" command to flush bridge's MAC learning table.
9 ------------------------
11 - Added an OpenFlow extension which allows the "output" action to accept
13 - Added an OpenFlow extension for flexible learning.
14 - Bumped number of NXM registers from four to five.
16 - New "version" command to determine version of running daemon.
17 - If no argument is provided for "cfm/show", displays detailed
18 information about all interfaces with CFM enabled.
19 - If no argument is provided for "lacp/show", displays detailed
20 information about all ports with LACP enabled.
22 - New "set-if" command to modify a datapath port's configuration.
24 - The software switch now supports 255 OpenFlow tables, instead
25 of just one. By default, only table 0 is consulted, but the
26 new NXAST_RESUBMIT_TABLE action can look up in additional
27 tables. Tables 128 and above are reserved for use by the
28 switch itself; please use only tables 0 through 127.
29 - Add support for 802.1D spanning tree (STP).
30 - Fragment handling extensions:
31 - New OFPC_FRAG_NX_MATCH fragment handling mode, in which L4
32 fields are made available for matching in fragments with
34 - New NXM_NX_IP_FRAG match field for matching IP fragments (usable
35 via "ip_frag" in ovs-ofctl).
36 - New ovs-ofctl "get-frags" and "set-frags" commands to get and set
37 fragment handling policy.
38 - CAPWAP tunneling now supports an extension to transport a 64-key. By
39 default it remains compatible with the old version and other
40 standards-based implementations.
41 - Flow setups are now processed in a round-robin manner across ports
42 to prevent any single client from monopolizing the CPU and conducting
43 a denial of service attack.
44 - Added support for native VLAN tagging. A new "vlan_mode"
45 parameter can be set for "port". Possible values: "access",
46 "trunk", "native-tagged" and "native-untagged".
47 - test-openflowd has been removed. Please use ovs-vswitchd instead.
50 ------------------------
51 - New "ofproto" abstraction layer to ease porting to hardware
53 - Packaging for Red Hat Enterprise Linux 5.6 and 6.0.
54 - Datapath support for Linux kernels up to 3.0.
56 - New "bundle" and "bundle_load" action extensions.
58 - Implement table unique constraints.
59 - Support cooperative locking between callers.
61 - New "-s" option for "show" command prints packet and byte
62 counters for each port.
64 - New "--readd" option for "replace-flows".
66 - New "show" command to print an overview of configuration.
67 - New "comment" command to add remark that explains intentions.
68 - ovs-brcompatd has been rewritten to fix long-standing bugs.
69 - ovs-openflowd has been renamed test-openflowd and moved into the
70 tests directory. Its presence confused too many users. Please
71 use ovs-vswitchd instead.
72 - New ovs-benchmark utility to test flow setup performance.
73 - A new log level "off" has been added. Configuring a log facility
74 "off" prevents any messages from being logged to it. Previously,
75 "emer" was effectively "off" because no messages were ever logged at
76 level "emer". Now, errors that cause a process to exit are logged
78 - "configure" option --with-l26 has been renamed --with-linux, and
79 --with-l26-source has been renamed --with-linux-source. The old
80 names will be removed after the next release, so please update
82 - The "-2.6" suffix has been dropped from the datapath/linux-2.6 and
83 datapath/linux-2.6/compat-2.6 directories.
85 - Dropped support for "tun_id_from_cookie" OpenFlow extension.
86 Please use the extensible match extensions instead.
87 - Removed the Maintenance_Point and Monitor tables in an effort
88 to simplify 802.1ag configuration.
89 - Performance and scalability improvements
93 ------------------------
94 - Ability to define policies over IPv6
97 - Support for extensible match extensions to OpenFlow
99 - Support for HFSC qdisc.
100 - Queue used by in-band control can now be configured.
102 - Kernel<->userspace interface has been reworked and should be
103 close to a stable ABI now.
104 - "Port group" concept has been dropped.
105 - GRE over IPSEC tunnels
107 - New active backup bonding mode.
108 - New L4 hashing support when LACP is enabled.
109 - Source MAC hash now includes VLAN field also.
111 - Greatly improved handling of large flow tables
113 - "show" command now prints full vport configuration.
114 - "dump-groups" command removed since kernel support for
115 port groups was dropped.
117 - New commands for working with the new Managers table.
118 - "list" command enhanced with new formatting options and --columns
120 - "get" command now accepts new --id option.
121 - New "find" command.
123 - New "queue-stats" command for printing queue stats.
124 - New commands "replace-flows" and "diff-flows".
125 - Commands to add and remove flows can now read from files.
126 - New --flow-format option to enable or disable NXM.
127 - New --more option to increase OpenFlow message verbosity.
128 - Removed "tun-cookie" command, which is no longer useful.
129 - ovs-controller enhancements for testing various features.
130 - New ovs-vlan-test command for testing for Linux kernel driver VLAN
131 bugs. New ovs-vlan-bug-workaround command for enabling and
132 disabling a workaround for these driver bugs.
134 - "Resubmit" actions now update flow statistics.
135 - New "register" extension for use in matching and actions, via NXM.
136 - New "multipath" experimental action extension.
137 - New support for matching multicast Ethernet frames, via NXM.
138 - New extension for OpenFlow vendor error codes.
139 - New extension to set the QoS output queue without actually
140 sending to an output port.
141 - Open vSwitch now reports a single flow table, instead of
142 separate hash and wildcard tables. This better models the
143 current implementation.
144 - New experimental "note" action.
145 - New "ofproto/trace" ovs-appctl command and associated utilities
146 to ease debugging complex flow tables.
148 - Schema documentation now includes an entity-relationship diagram.
149 - The database is now garbage collected. In most tables,
150 unreferenced rows will be deleted automatically.
151 - Many tables now include statistics updated periodically by
152 ovs-vswitchd or ovsdb-server.
153 - Every table now has an "external-ids" column for use by OVS
155 - There is no default controller anymore. Each bridge must have its
156 controller individually specified.
157 - The "fail-mode" is now a property of a Bridge instead of a Controller.
158 - New versioning and checksum features.
159 - New Managers table and manager_options column in Open_vSwitch table
160 for specifying managers. The old "managers" column in the
161 Open_vSwitch table has been removed.
162 - Many "name" columns are now immutable.
164 - Dropped support for XenServer pre-5.6.100.
165 - Dropped support for Linux pre-2.6.18.
166 - Dropped controller discovery support.
167 - Dropped "ovs-ofctl status" and the OpenFlow extension that it used.
168 Statistics reporting in the database is a rough equivalent.
169 - Dropped the "corekeeper" package (now separate, at
170 http://openvswitch.org/cgi-bin/gitweb.cgi?p=corekeeper).
171 - Performance and scalability improvements
174 v1.1.0pre2 - 13 Sep 2010
175 ------------------------
178 v1.1.0pre1 - 31 Aug 2010
179 ------------------------
180 - OpenFlow 1.0 slicing (QoS) functionality
181 - Python bindings for configuration database (no write support)
182 - Performance and scalability improvements
187 - New "patch" interface type
192 - Configuration database with remote management
195 - Support for XenServer 5.5 and 5.6
196 - Performance and scalability improvements
199 v0.99.2 - 18 Feb 2010
200 ---------------------
203 v0.99.1 - 25 Jan 2010
204 ---------------------
205 - Add support for sFlow(R)
206 - Make headers compatible with C++
209 v0.99.0 - 14 Jan 2010
210 ---------------------
211 - User-space forwarding engine
214 v0.90.7 - 29 Nov 2009
215 ---------------------
216 - Add support for NetFlow active timeouts
223 v0.90.5 - 21 Sep 2009
224 ---------------------
225 - Generalize in-band control to more diverse network setups