Merging default into bug23554

14 files changed, 2031 insertions, 17 deletions

diff --git a/docs/rabbitmqctl.1.xml b/docs/rabbitmqctl.1.xml
index a0f03192..cc3d4d2a 100644
--- a/docs/rabbitmqctl.1.xml
+++ b/docs/rabbitmqctl.1.xml
@@ -1361,6 +1361,49 @@
 
       </variablelist>
     </refsect2>
+
+    <refsect2>
+      <title>Mirrored Queue Management</title>
+      <para>
+        Mirrored queues can have slaves dynamically added, and slaves
+        or the master dynamically dropped. Refer to the <ulink
+        url="http://www.rabbitmq.com/ha.html">High Availability
+        guide</ulink> for further details about mirrored queues in
+        general.
+      </para>
+
+      <variablelist>
+        <varlistentry>
+          <term><cmdsynopsis><command>add_queue_mirror</command> <arg choice="req"><replaceable>queue_name</replaceable></arg> <arg choice="req"><replaceable>node</replaceable></arg></cmdsynopsis></term>
+          <listitem>
+            <para>
+              Attempts to add a mirror of the queue
+              <command>queue_name</command> on
+              <command>node</command>. This will only succeed if the
+              queue was declared a mirrored queue and if there is no
+              mirror of the queue already on the node. If it succeeds,
+              the new mirror will start off as an empty slave.
+            </para>
+          </listitem>
+        </varlistentry>
+
+        <varlistentry>
+          <term><cmdsynopsis><command>drop_queue_mirror</command> <arg choice="req"><replaceable>queue_name</replaceable></arg> <arg choice="req"><replaceable>node</replaceable></arg></cmdsynopsis></term>
+          <listitem>
+            <para>
+              Attempts to drop a mirror of the queue
+              <command>queue_name</command> on
+              <command>node</command>. This will only succeed if the
+              queue was declared a mirrored queue and if there is a
+              mirror of the queue already on the node. If the node
+              contains the master of the queue, a slave on some other
+              node will be promoted to become the new master. It is
+              not permitted to drop the only node of a mirrored-queue.
+            </para>
+          </listitem>
+        </varlistentry>
+      </variablelist>
+    </refsect2>
   </refsect1>
 
 </refentry>
diff --git a/include/rabbit.hrl b/include/rabbit.hrl
index b65f048d..0a202c5e 100644
--- a/include/rabbit.hrl
+++ b/include/rabbit.hrl
@@ -46,7 +46,7 @@
 -record(exchange_serial, {name, next}).
 
 -record(amqqueue, {name, durable, auto_delete, exclusive_owner = none,
-                   arguments, pid}).
+                   arguments, pid, mirror_pids}).
 
 %% mnesia doesn't like unary records, so we add a dummy 'value' field
 -record(route, {binding, value = const}).
diff --git a/src/rabbit_amqqueue.erl b/src/rabbit_amqqueue.erl
index c8703740..e5c53620 100644
--- a/src/rabbit_amqqueue.erl
+++ b/src/rabbit_amqqueue.erl
@@ -27,6 +27,8 @@
 -export([notify_sent/2, unblock/2, flush_all/2]).
 -export([commit_all/3, rollback_all/3, notify_down_all/2, limit_all/3]).
 -export([on_node_down/1]).
+-export([store_queue/1]).
+
 
 %% internal
 -export([internal_declare/2, internal_delete/1,
@@ -202,7 +204,8 @@ declare(QueueName, Durable, AutoDelete, Args, Owner) ->
                                       auto_delete     = AutoDelete,
                                       arguments       = Args,
                                       exclusive_owner = Owner,
-                                      pid             = none}),
+                                      pid             = none,
+                                      mirror_pids     = []}),
     case gen_server2:call(Q#amqqueue.pid, {init, false}, infinity) of
         not_found -> rabbit_misc:not_found(QueueName);
         Q1        -> Q1
@@ -257,8 +260,13 @@ lookup(Name) ->
 
 with(Name, F, E) ->
     case lookup(Name) of
-        {ok, Q} -> rabbit_misc:with_exit_handler(E, fun () -> F(Q) end);
-        {error, not_found} -> E()
+        {ok, Q = #amqqueue{mirror_pids = []}} ->
+            rabbit_misc:with_exit_handler(E, fun () -> F(Q) end);
+        {ok, Q} ->
+            E1 = fun () -> timer:sleep(25), with(Name, F, E) end,
+            rabbit_misc:with_exit_handler(E1, fun () -> F(Q) end);
+        {error, not_found} ->
+            E()
     end.
 
 with(Name, F) ->
@@ -295,8 +303,9 @@ with_exclusive_access_or_die(Name, ReaderPid, F) ->
 
 assert_args_equivalence(#amqqueue{name = QueueName, arguments = Args},
                         RequiredArgs) ->
-    rabbit_misc:assert_args_equivalence(Args, RequiredArgs, QueueName,
-                                        [<<"x-expires">>, <<"x-message-ttl">>]).
+    rabbit_misc:assert_args_equivalence(
+      Args, RequiredArgs, QueueName,
+      [<<"x-expires">>, <<"x-message-ttl">>, <<"x-mirror">>]).
 
 check_declare_arguments(QueueName, Args) ->
     [case Fun(rabbit_misc:table_lookup(Args, Key)) of
@@ -307,7 +316,8 @@ check_declare_arguments(QueueName, Args) ->
                              [Key, rabbit_misc:rs(QueueName), Error])
      end || {Key, Fun} <-
                 [{<<"x-expires">>,     fun check_integer_argument/1},
-                 {<<"x-message-ttl">>, fun check_integer_argument/1}]],
+                 {<<"x-message-ttl">>, fun check_integer_argument/1},
+                 {<<"x-mirror">>,      fun check_array_of_longstr_argument/1}]],
     ok.
 
 check_integer_argument(undefined) ->
@@ -320,6 +330,18 @@ check_integer_argument({Type, Val}) when Val > 0 ->
 check_integer_argument({_Type, Val}) ->
     {error, {value_zero_or_less, Val}}.
 
+check_array_of_longstr_argument(undefined) ->
+    ok;
+check_array_of_longstr_argument({array, Array}) ->
+    case lists:all(fun ({longstr, _NodeName}) -> true;
+                       (_)                    -> false
+                   end, Array) of
+        true  -> ok;
+        false -> {error, {array_contains_non_longstrs, Array}}
+    end;
+check_array_of_longstr_argument({Type, _Val}) ->
+    {error, {unacceptable_type, Type}}.
+
 list(VHostPath) ->
     mnesia:dirty_match_object(
       rabbit_queue,
@@ -474,7 +496,8 @@ drop_expired(QPid) ->
 on_node_down(Node) ->
     rabbit_misc:execute_mnesia_tx_with_tail(
       fun () -> Dels = qlc:e(qlc:q([delete_queue(QueueName) ||
-                                       #amqqueue{name = QueueName, pid = Pid}
+                                       #amqqueue{name = QueueName, pid = Pid,
+                                                 mirror_pids = []}
                                            <- mnesia:table(rabbit_queue),
                                        node(Pid) == Node])),
                 rabbit_binding:process_deletions(
@@ -491,7 +514,8 @@ pseudo_queue(QueueName, Pid) ->
               durable     = false,
               auto_delete = false,
               arguments   = [],
-              pid         = Pid}.
+              pid         = Pid,
+              mirror_pids = []}.
 
 safe_delegate_call_ok(F, Pids) ->
     case delegate:invoke(Pids, fun (Pid) ->
diff --git a/src/rabbit_amqqueue_process.erl b/src/rabbit_amqqueue_process.erl
index 1e5ad349..e7cb67a2 100644
--- a/src/rabbit_amqqueue_process.erl
+++ b/src/rabbit_amqqueue_process.erl
@@ -33,6 +33,8 @@
          handle_info/2, handle_pre_hibernate/1, prioritise_call/3,
          prioritise_cast/2, prioritise_info/2]).
 
+-export([init_with_backing_queue_state/7]).
+
 %% Queue's state
 -record(q, {q,
             exclusive_consumer,
@@ -72,7 +74,8 @@
          messages,
          consumers,
          memory,
-         backing_queue_status
+         backing_queue_status,
+         mirror_pids
         ]).
 
 -define(CREATION_EVENT_KEYS,
@@ -114,6 +117,34 @@ init(Q) ->
             msg_id_to_channel   = dict:new()}, hibernate,
      {backoff, ?HIBERNATE_AFTER_MIN, ?HIBERNATE_AFTER_MIN, ?DESIRED_HIBERNATE}}.
 
+init_with_backing_queue_state(Q = #amqqueue{exclusive_owner = Owner}, BQ, BQS,
+                              RateTRef, AckTags, Deliveries, MTC) ->
+    ?LOGDEBUG("Queue starting - ~p~n", [Q]),
+    case Owner of
+        none -> ok;
+        _    -> erlang:monitor(process, Owner)
+    end,
+    State = requeue_and_run(
+              AckTags,
+              process_args(
+                #q{q                   = Q,
+                   exclusive_consumer  = none,
+                   has_had_consumers   = false,
+                   backing_queue       = BQ,
+                   backing_queue_state = BQS,
+                   active_consumers    = queue:new(),
+                   blocked_consumers   = queue:new(),
+                   expires             = undefined,
+                   sync_timer_ref      = undefined,
+                   rate_timer_ref      = RateTRef,
+                   expiry_timer_ref    = undefined,
+                   ttl                 = undefined,
+                   stats_timer         = rabbit_event:init_stats_timer(),
+                   msg_id_to_channel   = MTC})),
+    lists:foldl(
+      fun (Delivery, StateN) -> deliver_or_enqueue(Delivery, StateN) end,
+      State, Deliveries).
+
 terminate(shutdown = R,      State = #q{backing_queue = BQ}) ->
     terminate_shutdown(fun (BQS) -> BQ:terminate(R, BQS) end, State);
 terminate({shutdown, _} = R, State = #q{backing_queue = BQ}) ->
@@ -225,9 +256,12 @@ next_state(State = #q{backing_queue = BQ, backing_queue_state = BQS}) ->
         timed -> {ensure_sync_timer(State1), 0        }
     end.
 
-backing_queue_module(#amqqueue{}) ->
-    {ok, BQM} = application:get_env(backing_queue_module),
-    BQM.
+backing_queue_module(#amqqueue{arguments = Args}) ->
+    case rabbit_misc:table_lookup(Args, <<"x-mirror">>) of
+        undefined -> {ok, BQM} = application:get_env(backing_queue_module),
+                     BQM;
+        _Nodes    -> rabbit_mirror_queue_master
+    end.
 
 ensure_sync_timer(State = #q{sync_timer_ref = undefined}) ->
     {ok, TRef} = timer:apply_after(
@@ -769,6 +803,9 @@ i(memory, _) ->
     M;
 i(backing_queue_status, #q{backing_queue_state = BQS, backing_queue = BQ}) ->
     BQ:status(BQS);
+i(mirror_pids, #q{q = #amqqueue{name = Name}}) ->
+    {ok, #amqqueue{mirror_pids = MPids}} = rabbit_amqqueue:lookup(Name),
+    MPids;
 i(Item, _) ->
     throw({bad_argument, Item}).
 
diff --git a/src/rabbit_control.erl b/src/rabbit_control.erl
index 9eef384a..9194a45b 100644
--- a/src/rabbit_control.erl
+++ b/src/rabbit_control.erl
@@ -274,6 +274,18 @@ action(list_queues, Node, Args, Opts, Inform) ->
                                [VHostArg, ArgAtoms]),
                       ArgAtoms);
 
+action(add_queue_mirror, Node, [Queue, MirrorNode], Opts, Inform) ->
+    Inform("Adding mirror of queue ~p on node ~p~n", [Queue, MirrorNode]),
+    VHostArg = list_to_binary(proplists:get_value(?VHOST_OPT, Opts)),
+    rpc_call(Node, rabbit_mirror_queue_misc, add_slave,
+             [VHostArg, list_to_binary(Queue), list_to_atom(MirrorNode)]);
+
+action(drop_queue_mirror, Node, [Queue, MirrorNode], Opts, Inform) ->
+    Inform("Dropping mirror of queue ~p on node ~p~n", [Queue, MirrorNode]),
+    VHostArg = list_to_binary(proplists:get_value(?VHOST_OPT, Opts)),
+    rpc_call(Node, rabbit_mirror_queue_misc, drop_slave,
+             [VHostArg, list_to_binary(Queue), list_to_atom(MirrorNode)]);
+
 action(list_exchanges, Node, Args, Opts, Inform) ->
     Inform("Listing exchanges", []),
     VHostArg = list_to_binary(proplists:get_value(?VHOST_OPT, Opts)),
diff --git a/src/rabbit_mirror_queue_coordinator.erl b/src/rabbit_mirror_queue_coordinator.erl
new file mode 100644
index 00000000..2727c1d0
--- /dev/null
+++ b/src/rabbit_mirror_queue_coordinator.erl
@@ -0,0 +1,424 @@
+%% The contents of this file are subject to the Mozilla Public License
+%% Version 1.1 (the "License"); you may not use this file except in
+%% compliance with the License. You may obtain a copy of the License at
+%% http://www.mozilla.org/MPL/
+%%
+%% Software distributed under the License is distributed on an "AS IS"
+%% basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See the
+%% License for the specific language governing rights and limitations
+%% under the License.
+%%
+%% The Original Code is RabbitMQ.
+%%
+%% The Initial Developer of the Original Code is VMware, Inc.
+%% Copyright (c) 2007-2010 VMware, Inc.  All rights reserved.
+%%
+
+-module(rabbit_mirror_queue_coordinator).
+
+-export([start_link/3, get_gm/1, ensure_monitoring/2]).
+
+-export([init/1, handle_call/3, handle_cast/2, handle_info/2, terminate/2,
+         code_change/3]).
+
+-export([joined/2, members_changed/3, handle_msg/3]).
+
+-behaviour(gen_server2).
+-behaviour(gm).
+
+-include("rabbit.hrl").
+-include("gm_specs.hrl").
+
+-record(state, { q,
+                 gm,
+                 monitors,
+                 death_fun
+               }).
+
+-define(ONE_SECOND, 1000).
+
+%%----------------------------------------------------------------------------
+%%
+%% Mirror Queues
+%%
+%% A queue with mirrors consists of the following:
+%%
+%%  #amqqueue{ pid, mirror_pids }
+%%             |    |
+%%  +----------+    +-------+--------------+-----------...etc...
+%%  |                       |              |
+%%  V                       V              V
+%% amqqueue_process---+    slave-----+    slave-----+  ...etc...
+%% | BQ = master----+ |    | BQ = vq |    | BQ = vq |
+%% |      | BQ = vq | |    +-+-------+    +-+-------+
+%% |      +-+-------+ |      |              |
+%% +-++-----|---------+      |              |  (some details elided)
+%%   ||     |                |              |
+%%   ||   coordinator-+      |              |
+%%   ||   +-+---------+      |              |
+%%   ||     |                |              |
+%%   ||     gm-+ -- -- -- -- gm-+- -- -- -- gm-+- -- --...etc...
+%%   ||     +--+             +--+           +--+
+%%   ||
+%%  consumers
+%%
+%% The master is merely an implementation of bq, and thus is invoked
+%% through the normal bq interface by the amqqueue_process. The slaves
+%% meanwhile are processes in their own right (as is the
+%% coordinator). The coordinator and all slaves belong to the same gm
+%% group. Every member of a gm group receives messages sent to the gm
+%% group. Because the master is the bq of amqqueue_process, it doesn't
+%% have sole control over its mailbox, and as a result, the master
+%% itself cannot be passed messages directly (well, it could by via
+%% the amqqueue:run_backing_queue_async callback but that would induce
+%% additional unnecessary loading on the master queue process), yet it
+%% needs to react to gm events, such as the death of slaves. Thus the
+%% master creates the coordinator, and it is the coordinator that is
+%% the gm callback module and event handler for the master.
+%%
+%% Consumers are only attached to the master. Thus the master is
+%% responsible for informing all slaves when messages are fetched from
+%% the bq, when they're acked, and when they're requeued.
+%%
+%% The basic goal is to ensure that all slaves performs actions on
+%% their bqs in the same order as the master. Thus the master
+%% intercepts all events going to its bq, and suitably broadcasts
+%% these events on the gm. The slaves thus receive two streams of
+%% events: one stream is via the gm, and one stream is from channels
+%% directly. Whilst the stream via gm is guaranteed to be consistently
+%% seen by all slaves, the same is not true of the stream via
+%% channels. For example, in the event of an unexpected death of a
+%% channel during a publish, only some of the mirrors may receive that
+%% publish. As a result of this problem, the messages broadcast over
+%% the gm contain published content, and thus slaves can operate
+%% successfully on messages that they only receive via the gm. The key
+%% purpose of also sending messages directly from the channels to the
+%% slaves is that without this, in the event of the death of the
+%% master, messages could be lost until a suitable slave is promoted.
+%%
+%% However, that is not the only reason. For example, if confirms are
+%% in use, then there is no guarantee that every slave will see the
+%% delivery with the same msg_seq_no. As a result, the slaves have to
+%% wait until they've seen both the publish via gm, and the publish
+%% via the channel before they have enough information to be able to
+%% perform the publish to their own bq, and subsequently issue the
+%% confirm, if necessary. Either form of publish can arrive first, and
+%% a slave can be upgraded to the master at any point during this
+%% process. Confirms continue to be issued correctly, however.
+%%
+%% Because the slave is a full process, it impersonates parts of the
+%% amqqueue API. However, it does not need to implement all parts: for
+%% example, no ack or consumer-related message can arrive directly at
+%% a slave from a channel: it is only publishes that pass both
+%% directly to the slaves and go via gm.
+%%
+%% Slaves can be added dynamically. When this occurs, there is no
+%% attempt made to sync the current contents of the master with the
+%% new slave, thus the slave will start empty, regardless of the state
+%% of the master. Thus the slave needs to be able to detect and ignore
+%% operations which are for messages it has not received: because of
+%% the strict FIFO nature of queues in general, this is
+%% straightforward - all new publishes that the new slave receives via
+%% gm should be processed as normal, but fetches which are for
+%% messages the slave has never seen should be ignored. Similarly,
+%% acks for messages the slave never fetched should be
+%% ignored. Eventually, as the master is consumed from, the messages
+%% at the head of the queue which were there before the slave joined
+%% will disappear, and the slave will become fully synced with the
+%% state of the master. The detection of the sync-status of a slave is
+%% done entirely based on length: if the slave and the master both
+%% agree on the length of the queue after the fetch of the head of the
+%% queue, then the queues must be in sync. The only other possibility
+%% is that the slave's queue is shorter, and thus the fetch should be
+%% ignored.
+%%
+%% Because acktags are issued by the bq independently, and because
+%% there is no requirement for the master and all slaves to use the
+%% same bq, all references to msgs going over gm is by msg_id. Thus
+%% upon acking, the master must convert the acktags back to msg_ids
+%% (which happens to be what bq:ack returns), then sends the msg_ids
+%% over gm, the slaves must convert the msg_ids to acktags (a mapping
+%% the slaves themselves must maintain).
+%%
+%% When the master dies, a slave gets promoted. This will be the
+%% eldest slave, and thus the hope is that that slave is most likely
+%% to be sync'd with the master. The design of gm is that the
+%% notification of the death of the master will only appear once all
+%% messages in-flight from the master have been fully delivered to all
+%% members of the gm group. Thus at this point, the slave that gets
+%% promoted cannot broadcast different events in a different order
+%% than the master for the same msgs: there is no possibility for the
+%% same msg to be processed by the old master and the new master - if
+%% it was processed by the old master then it will have been processed
+%% by the slave before the slave was promoted, and vice versa.
+%%
+%% Upon promotion, all msgs pending acks are requeued as normal, the
+%% slave constructs state suitable for use in the master module, and
+%% then dynamically changes into an amqqueue_process with the master
+%% as the bq, and the slave's bq as the master's bq. Thus the very
+%% same process that was the slave is now a full amqqueue_process.
+%%
+%% It is important that we avoid memory leaks due to the death of
+%% senders (i.e. channels) and partial publications. A sender
+%% publishing a message may fail mid way through the publish and thus
+%% only some of the mirrors will receive the message. We need the
+%% mirrors to be able to detect this and tidy up as necessary to avoid
+%% leaks. If we just had the master monitoring all senders then we
+%% would have the possibility that a sender appears and only sends the
+%% message to a few of the slaves before dying. Those slaves would
+%% then hold on to the message, assuming they'll receive some
+%% instruction eventually from the master. Thus we have both slaves
+%% and the master monitor all senders they become aware of. But there
+%% is a race: if the slave receives a DOWN of a sender, how does it
+%% know whether or not the master is going to send it instructions
+%% regarding those messages?
+%%
+%% Whilst the master monitors senders, it can't access its mailbox
+%% directly, so it delegates monitoring to the coordinator. When the
+%% coordinator receives a DOWN message from a sender, it informs the
+%% master via a callback. This allows the master to do any tidying
+%% necessary, but more importantly allows the master to broadcast a
+%% sender_death message to all the slaves, saying the sender has
+%% died. Once the slaves receive the sender_death message, they know
+%% that they're not going to receive any more instructions from the gm
+%% regarding that sender, thus they throw away any publications from
+%% the sender pending publication instructions. However, it is
+%% possible that the coordinator receives the DOWN and communicates
+%% that to the master before the master has finished receiving and
+%% processing publishes from the sender. This turns out not to be a
+%% problem: the sender has actually died, and so will not need to
+%% receive confirms or other feedback, and should further messages be
+%% "received" from the sender, the master will ask the coordinator to
+%% set up a new monitor, and will continue to process the messages
+%% normally. Slaves may thus receive publishes via gm from previously
+%% declared "dead" senders, but again, this is fine: should the slave
+%% have just thrown out the message it had received directly from the
+%% sender (due to receiving a sender_death message via gm), it will be
+%% able to cope with the publication purely from the master via gm.
+%%
+%% When a slave receives a DOWN message for a sender, if it has not
+%% received the sender_death message from the master via gm already,
+%% then it will wait 20 seconds before broadcasting a request for
+%% confirmation from the master that the sender really has died.
+%% Should a sender have only sent a publish to slaves, this allows
+%% slaves to inform the master of the previous existence of the
+%% sender. The master will thus monitor the sender, receive the DOWN,
+%% and subsequently broadcast the sender_death message, allowing the
+%% slaves to tidy up. This process can repeat for the same sender:
+%% consider one slave receives the publication, then the DOWN, then
+%% asks for confirmation of death, then the master broadcasts the
+%% sender_death message. Only then does another slave receive the
+%% publication and thus set up its monitoring. Eventually that slave
+%% too will receive the DOWN, ask for confirmation and the master will
+%% monitor the sender again, receive another DOWN, and send out
+%% another sender_death message. Given the 20 second delay before
+%% requesting death confirmation, this is highly unlikely, but it is a
+%% possibility.
+%%
+%% When the 20 second timer expires, the slave first checks to see
+%% whether it still needs confirmation of the death before requesting
+%% it. This prevents unnecessary traffic on gm as it allows one
+%% broadcast of the sender_death message to satisfy many slaves.
+%%
+%% If we consider the promotion of a slave at this point, we have two
+%% possibilities: that of the slave that has received the DOWN and is
+%% thus waiting for confirmation from the master that the sender
+%% really is down; and that of the slave that has not received the
+%% DOWN. In the first case, in the act of promotion to master, the new
+%% master will monitor again the dead sender, and after it has
+%% finished promoting itself, it should find another DOWN waiting,
+%% which it will then broadcast. This will allow slaves to tidy up as
+%% normal. In the second case, we have the possibility that
+%% confirmation-of-sender-death request has been broadcast, but that
+%% it was broadcast before the master failed, and that the slave being
+%% promoted does not know anything about that sender, and so will not
+%% monitor it on promotion. Thus a slave that broadcasts such a
+%% request, at the point of broadcasting it, recurses, setting another
+%% 20 second timer. As before, on expiry of the timer, the slaves
+%% checks to see whether it still has not received a sender_death
+%% message for the dead sender, and if not, broadcasts a death
+%% confirmation request. Thus this ensures that even when a master
+%% dies and the new slave has no knowledge of the dead sender, it will
+%% eventually receive a death confirmation request, shall monitor the
+%% dead sender, receive the DOWN and broadcast the sender_death
+%% message.
+%%
+%% The preceding commentary deals with the possibility of slaves
+%% receiving publications from senders which the master does not, and
+%% the need to prevent memory leaks in such scenarios. The inverse is
+%% also possible: a partial publication may cause only the master to
+%% receive a publication. It will then publish the message via gm. The
+%% slaves will receive it via gm, will publish it to their BQ and will
+%% set up monitoring on the sender. They will then receive the DOWN
+%% message and the master will eventually publish the corresponding
+%% sender_death message. The slave will then be able to tidy up its
+%% state as normal.
+%%
+%% We don't support transactions on mirror queues. To do so is
+%% challenging. The underlying bq is free to add the contents of the
+%% txn to the queue proper at any point after the tx.commit comes in
+%% but before the tx.commit-ok goes out. This means that it is not
+%% safe for all mirrors to simply issue the bq:tx_commit at the same
+%% time, as the addition of the txn's contents to the queue may
+%% subsequently be inconsistently interwoven with other actions on the
+%% bq. The solution to this is, in the master, wrap the PostCommitFun
+%% and do the gm:broadcast in there: at that point, you're in the bq
+%% (well, there's actually nothing to stop that function being invoked
+%% by some other process, but let's pretend for now: you could always
+%% use run_backing_queue to ensure you really are in the queue process
+%% (the _async variant would be unsafe from an ordering pov)), the
+%% gm:broadcast is safe because you don't have to worry about races
+%% with other gm:broadcast calls (same process). Thus this signal
+%% would indicate sufficiently to all the slaves that they must insert
+%% the complete contents of the txn at precisely this point in the
+%% stream of events.
+%%
+%% However, it's quite difficult for the slaves to make that happen:
+%% they would be forced to issue the bq:tx_commit at that point, but
+%% then stall processing any further instructions from gm until they
+%% receive the notification from their bq that the tx_commit has fully
+%% completed (i.e. they need to treat what is an async system as being
+%% fully synchronous). This is not too bad (apart from the
+%% vomit-inducing notion of it all): just need a queue of instructions
+%% from the GM; but then it gets rather worse when you consider what
+%% needs to happen if the master dies at this point and the slave in
+%% the middle of this tx_commit needs to be promoted.
+%%
+%% Finally, we can't possibly hope to make transactions atomic across
+%% mirror queues, and it's not even clear that that's desirable: if a
+%% slave fails whilst there's an open transaction in progress then
+%% when the channel comes to commit the txn, it will detect the
+%% failure and destroy the channel. However, the txn will have
+%% actually committed successfully in all the other mirrors (including
+%% master). To do this bit properly would require 2PC and all the
+%% baggage that goes with that.
+%%
+%% Recovery of mirrored queues is straightforward: as nodes die, the
+%% remaining nodes record this, and eventually a situation is reached
+%% in which only one node is alive, which is the master. This is the
+%% only node which, upon recovery, will resurrect a mirrored queue:
+%% nodes which die and then rejoin as a slave will start off empty as
+%% if they have no mirrored content at all. This is not surprising: to
+%% achieve anything more sophisticated would require the master and
+%% recovering slave to be able to check to see whether they agree on
+%% the last seen state of the queue: checking length alone is not
+%% sufficient in this case.
+%%
+%% For more documentation see the comments in bug 23554.
+%%
+%%----------------------------------------------------------------------------
+
+start_link(Queue, GM, DeathFun) ->
+    gen_server2:start_link(?MODULE, [Queue, GM, DeathFun], []).
+
+get_gm(CPid) ->
+    gen_server2:call(CPid, get_gm, infinity).
+
+ensure_monitoring(CPid, Pids) ->
+    gen_server2:cast(CPid, {ensure_monitoring, Pids}).
+
+%% ---------------------------------------------------------------------------
+%% gen_server
+%% ---------------------------------------------------------------------------
+
+init([#amqqueue { name = QueueName } = Q, GM, DeathFun]) ->
+    GM1 = case GM of
+              undefined ->
+                  {ok, GM2} = gm:start_link(QueueName, ?MODULE, [self()]),
+                  receive {joined, GM2, _Members} ->
+                          ok
+                  end,
+                  GM2;
+              _ ->
+                  true = link(GM),
+                  GM
+          end,
+    {ok, _TRef} =
+        timer:apply_interval(?ONE_SECOND, gm, broadcast, [GM1, heartbeat]),
+    {ok, #state { q         = Q,
+                  gm        = GM1,
+                  monitors  = dict:new(),
+                  death_fun = DeathFun },
+     hibernate,
+     {backoff, ?HIBERNATE_AFTER_MIN, ?HIBERNATE_AFTER_MIN, ?DESIRED_HIBERNATE}}.
+
+handle_call(get_gm, _From, State = #state { gm = GM }) ->
+    reply(GM, State).
+
+handle_cast({gm_deaths, Deaths},
+            State = #state { q  = #amqqueue { name = QueueName } }) ->
+    rabbit_log:info("Mirrored-queue (~s): Master ~s saw deaths of mirrors ~s~n",
+                    [rabbit_misc:rs(QueueName),
+                     rabbit_misc:pid_to_string(self()),
+                     [[rabbit_misc:pid_to_string(Pid), $ ] || Pid <- Deaths]]),
+    case rabbit_mirror_queue_misc:remove_from_queue(QueueName, Deaths) of
+        {ok, Pid} when node(Pid) =:= node() ->
+            noreply(State);
+        {error, not_found} ->
+            {stop, normal, State}
+    end;
+
+handle_cast({ensure_monitoring, Pids},
+            State = #state { monitors = Monitors }) ->
+    Monitors1 =
+        lists:foldl(fun (Pid, MonitorsN) ->
+                            case dict:is_key(Pid, MonitorsN) of
+                                true  -> MonitorsN;
+                                false -> MRef = erlang:monitor(process, Pid),
+                                         dict:store(Pid, MRef, MonitorsN)
+                            end
+                    end, Monitors, Pids),
+    noreply(State #state { monitors = Monitors1 }).
+
+handle_info({'DOWN', _MonitorRef, process, Pid, _Reason},
+            State = #state { monitors  = Monitors,
+                             death_fun = Fun }) ->
+    noreply(
+      case dict:is_key(Pid, Monitors) of
+          false -> State;
+          true  -> ok = Fun(Pid),
+                   State #state { monitors = dict:erase(Pid, Monitors) }
+      end);
+
+handle_info(Msg, State) ->
+    {stop, {unexpected_info, Msg}, State}.
+
+terminate(_Reason, #state{}) ->
+    %% gen_server case
+    ok;
+terminate([_CPid], _Reason) ->
+    %% gm case
+    ok.
+
+code_change(_OldVsn, State, _Extra) ->
+    {ok, State}.
+
+%% ---------------------------------------------------------------------------
+%% GM
+%% ---------------------------------------------------------------------------
+
+joined([CPid], Members) ->
+    CPid ! {joined, self(), Members},
+    ok.
+
+members_changed([_CPid], _Births, []) ->
+    ok;
+members_changed([CPid], _Births, Deaths) ->
+    ok = gen_server2:cast(CPid, {gm_deaths, Deaths}).
+
+handle_msg([_CPid], _From, heartbeat) ->
+    ok;
+handle_msg([CPid], _From, {ensure_monitoring, _Pids} = Msg) ->
+    ok = gen_server2:cast(CPid, Msg);
+handle_msg([_CPid], _From, _Msg) ->
+    ok.
+
+%% ---------------------------------------------------------------------------
+%% Others
+%% ---------------------------------------------------------------------------
+
+noreply(State) ->
+    {noreply, State, hibernate}.
+
+reply(Reply, State) ->
+    {reply, Reply, State, hibernate}.
diff --git a/src/rabbit_mirror_queue_master.erl b/src/rabbit_mirror_queue_master.erl
new file mode 100644
index 00000000..da12ea82
--- /dev/null
+++ b/src/rabbit_mirror_queue_master.erl
@@ -0,0 +1,380 @@
+%% The contents of this file are subject to the Mozilla Public License
+%% Version 1.1 (the "License"); you may not use this file except in
+%% compliance with the License. You may obtain a copy of the License at
+%% http://www.mozilla.org/MPL/
+%%
+%% Software distributed under the License is distributed on an "AS IS"
+%% basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See the
+%% License for the specific language governing rights and limitations
+%% under the License.
+%%
+%% The Original Code is RabbitMQ.
+%%
+%% The Initial Developer of the Original Code is VMware, Inc.
+%% Copyright (c) 2007-2010 VMware, Inc.  All rights reserved.
+%%
+
+-module(rabbit_mirror_queue_master).
+
+-export([init/4, terminate/2, delete_and_terminate/2,
+         purge/1, publish/4, publish_delivered/5, fetch/2, ack/2,
+         tx_publish/5, tx_ack/3, tx_rollback/2, tx_commit/4,
+         requeue/3, len/1, is_empty/1, drain_confirmed/1, dropwhile/2,
+         set_ram_duration_target/2, ram_duration/1,
+         needs_timeout/1, timeout/1, handle_pre_hibernate/1,
+         status/1, invoke/3, is_duplicate/3, discard/3]).
+
+-export([start/1, stop/0]).
+
+-export([promote_backing_queue_state/6, sender_death_fun/0]).
+
+-behaviour(rabbit_backing_queue).
+
+-include("rabbit.hrl").
+
+-record(state, { gm,
+                 coordinator,
+                 backing_queue,
+                 backing_queue_state,
+                 set_delivered,
+                 seen_status,
+                 confirmed,
+                 ack_msg_id,
+                 known_senders
+               }).
+
+%% For general documentation of HA design, see
+%% rabbit_mirror_queue_coordinator
+
+%% ---------------------------------------------------------------------------
+%% Backing queue
+%% ---------------------------------------------------------------------------
+
+start(_DurableQueues) ->
+    %% This will never get called as this module will never be
+    %% installed as the default BQ implementation.
+    exit({not_valid_for_generic_backing_queue, ?MODULE}).
+
+stop() ->
+    %% Same as start/1.
+    exit({not_valid_for_generic_backing_queue, ?MODULE}).
+
+sender_death_fun() ->
+    Self = self(),
+    fun (DeadPid) ->
+            rabbit_amqqueue:run_backing_queue_async(
+              Self, ?MODULE,
+              fun (?MODULE, State = #state { gm = GM, known_senders = KS }) ->
+                      ok = gm:broadcast(GM, {sender_death, DeadPid}),
+                      KS1 = sets:del_element(DeadPid, KS),
+                      State #state { known_senders = KS1 }
+              end)
+    end.
+
+init(#amqqueue { arguments = Args, name = QName } = Q, Recover,
+     AsyncCallback, SyncCallback) ->
+    {ok, CPid} = rabbit_mirror_queue_coordinator:start_link(
+                   Q, undefined, sender_death_fun()),
+    GM = rabbit_mirror_queue_coordinator:get_gm(CPid),
+    {_Type, Nodes} = rabbit_misc:table_lookup(Args, <<"x-mirror">>),
+    Nodes1 = (case Nodes of
+                  [] -> rabbit_mnesia:all_clustered_nodes();
+                  _  -> [list_to_atom(binary_to_list(Node)) ||
+                            {longstr, Node} <- Nodes]
+              end) -- [node()],
+    [rabbit_mirror_queue_misc:add_slave(QName, Node) || Node <- Nodes1],
+    {ok, BQ} = application:get_env(backing_queue_module),
+    BQS = BQ:init(Q, Recover, AsyncCallback, SyncCallback),
+    #state { gm                  = GM,
+             coordinator         = CPid,
+             backing_queue       = BQ,
+             backing_queue_state = BQS,
+             set_delivered       = 0,
+             seen_status         = dict:new(),
+             confirmed           = [],
+             ack_msg_id          = dict:new(),
+             known_senders       = sets:new() }.
+
+promote_backing_queue_state(CPid, BQ, BQS, GM, SeenStatus, KS) ->
+    #state { gm                  = GM,
+             coordinator         = CPid,
+             backing_queue       = BQ,
+             backing_queue_state = BQS,
+             set_delivered       = BQ:len(BQS),
+             seen_status         = SeenStatus,
+             confirmed           = [],
+             ack_msg_id          = dict:new(),
+             known_senders       = sets:from_list(KS) }.
+
+terminate({shutdown, dropped} = Reason,
+          State = #state { backing_queue = BQ, backing_queue_state = BQS }) ->
+    %% Backing queue termination - this node has been explicitly
+    %% dropped. Normally, non-durable queues would be tidied up on
+    %% startup, but there's a possibility that we will be added back
+    %% in without this node being restarted. Thus we must do the full
+    %% blown delete_and_terminate now, but only locally: we do not
+    %% broadcast delete_and_terminate.
+    State #state { backing_queue_state = BQ:delete_and_terminate(Reason, BQS),
+                   set_delivered       = 0 };
+terminate(Reason,
+          State = #state { backing_queue = BQ, backing_queue_state = BQS }) ->
+    %% Backing queue termination. The queue is going down but
+    %% shouldn't be deleted. Most likely safe shutdown of this
+    %% node. Thus just let some other slave take over.
+    State #state { backing_queue_state = BQ:terminate(Reason, BQS) }.
+
+delete_and_terminate(Reason, State = #state { gm                  = GM,
+                                              backing_queue       = BQ,
+                                              backing_queue_state = BQS }) ->
+    ok = gm:broadcast(GM, {delete_and_terminate, Reason}),
+    State #state { backing_queue_state = BQ:delete_and_terminate(Reason, BQS),
+                   set_delivered       = 0 }.
+
+purge(State = #state { gm                  = GM,
+                       backing_queue       = BQ,
+                       backing_queue_state = BQS }) ->
+    ok = gm:broadcast(GM, {set_length, 0}),
+    {Count, BQS1} = BQ:purge(BQS),
+    {Count, State #state { backing_queue_state = BQS1,
+                           set_delivered       = 0 }}.
+
+publish(Msg = #basic_message { id = MsgId }, MsgProps, ChPid,
+        State = #state { gm                  = GM,
+                         seen_status         = SS,
+                         backing_queue       = BQ,
+                         backing_queue_state = BQS }) ->
+    false = dict:is_key(MsgId, SS), %% ASSERTION
+    ok = gm:broadcast(GM, {publish, false, ChPid, MsgProps, Msg}),
+    BQS1 = BQ:publish(Msg, MsgProps, ChPid, BQS),
+    ensure_monitoring(ChPid, State #state { backing_queue_state = BQS1 }).
+
+publish_delivered(AckRequired, Msg = #basic_message { id = MsgId }, MsgProps,
+                  ChPid, State = #state { gm                  = GM,
+                                          seen_status         = SS,
+                                          backing_queue       = BQ,
+                                          backing_queue_state = BQS,
+                                          ack_msg_id          = AM }) ->
+    false = dict:is_key(MsgId, SS), %% ASSERTION
+    %% Must use confirmed_broadcast here in order to guarantee that
+    %% all slaves are forced to interpret this publish_delivered at
+    %% the same point, especially if we die and a slave is promoted.
+    ok = gm:confirmed_broadcast(
+           GM, {publish, {true, AckRequired}, ChPid, MsgProps, Msg}),
+    {AckTag, BQS1} =
+        BQ:publish_delivered(AckRequired, Msg, MsgProps, ChPid, BQS),
+    AM1 = maybe_store_acktag(AckTag, MsgId, AM),
+    {AckTag,
+     ensure_monitoring(ChPid, State #state { backing_queue_state = BQS1,
+                                             ack_msg_id          = AM1 })}.
+
+dropwhile(Fun, State = #state { gm                  = GM,
+                                backing_queue       = BQ,
+                                backing_queue_state = BQS,
+                                set_delivered       = SetDelivered }) ->
+    Len = BQ:len(BQS),
+    BQS1 = BQ:dropwhile(Fun, BQS),
+    Dropped = Len - BQ:len(BQS1),
+    SetDelivered1 = lists:max([0, SetDelivered - Dropped]),
+    ok = gm:broadcast(GM, {set_length, BQ:len(BQS1)}),
+    State #state { backing_queue_state = BQS1,
+                   set_delivered       = SetDelivered1 }.
+
+drain_confirmed(State = #state { backing_queue       = BQ,
+                                 backing_queue_state = BQS,
+                                 seen_status         = SS,
+                                 confirmed           = Confirmed }) ->
+    {MsgIds, BQS1} = BQ:drain_confirmed(BQS),
+    {MsgIds1, SS1} =
+        lists:foldl(
+          fun (MsgId, {MsgIdsN, SSN}) ->
+                  %% We will never see 'discarded' here
+                  case dict:find(MsgId, SSN) of
+                      error ->
+                          {[MsgId | MsgIdsN], SSN};
+                      {ok, published} ->
+                          %% It was published when we were a slave,
+                          %% and we were promoted before we saw the
+                          %% publish from the channel. We still
+                          %% haven't seen the channel publish, and
+                          %% consequently we need to filter out the
+                          %% confirm here. We will issue the confirm
+                          %% when we see the publish from the channel.
+                          {MsgIdsN, dict:store(MsgId, confirmed, SSN)};
+                      {ok, confirmed} ->
+                          %% Well, confirms are racy by definition.
+                          {[MsgId | MsgIdsN], SSN}
+                  end
+          end, {[], SS}, MsgIds),
+    {Confirmed ++ MsgIds1, State #state { backing_queue_state = BQS1,
+                                          seen_status         = SS1,
+                                          confirmed           = [] }}.
+
+fetch(AckRequired, State = #state { gm                  = GM,
+                                    backing_queue       = BQ,
+                                    backing_queue_state = BQS,
+                                    set_delivered       = SetDelivered,
+                                    ack_msg_id          = AM }) ->
+    {Result, BQS1} = BQ:fetch(AckRequired, BQS),
+    State1 = State #state { backing_queue_state = BQS1 },
+    case Result of
+        empty ->
+            {Result, State1};
+        {#basic_message { id = MsgId } = Message, IsDelivered, AckTag,
+         Remaining} ->
+            ok = gm:broadcast(GM, {fetch, AckRequired, MsgId, Remaining}),
+            IsDelivered1 = IsDelivered orelse SetDelivered > 0,
+            SetDelivered1 = lists:max([0, SetDelivered - 1]),
+            AM1 = maybe_store_acktag(AckTag, MsgId, AM),
+            {{Message, IsDelivered1, AckTag, Remaining},
+             State1 #state { set_delivered = SetDelivered1,
+                             ack_msg_id    = AM1 }}
+    end.
+
+ack(AckTags, State = #state { gm                  = GM,
+                              backing_queue       = BQ,
+                              backing_queue_state = BQS,
+                              ack_msg_id          = AM }) ->
+    {MsgIds, BQS1} = BQ:ack(AckTags, BQS),
+    AM1 = lists:foldl(fun dict:erase/2, AM, AckTags),
+    case MsgIds of
+        [] -> ok;
+        _  -> ok = gm:broadcast(GM, {ack, MsgIds})
+    end,
+    {MsgIds, State #state { backing_queue_state = BQS1,
+                            ack_msg_id          = AM1 }}.
+
+tx_publish(_Txn, _Msg, _MsgProps, _ChPid, State) ->
+    %% We don't support txns in mirror queues
+    State.
+
+tx_ack(_Txn, _AckTags, State) ->
+    %% We don't support txns in mirror queues
+    State.
+
+tx_rollback(_Txn, State) ->
+    {[], State}.
+
+tx_commit(_Txn, PostCommitFun, _MsgPropsFun, State) ->
+    PostCommitFun(), %% Probably must run it to avoid deadlocks
+    {[], State}.
+
+requeue(AckTags, MsgPropsFun, State = #state { gm                  = GM,
+                                               backing_queue       = BQ,
+                                               backing_queue_state = BQS }) ->
+    {MsgIds, BQS1} = BQ:requeue(AckTags, MsgPropsFun, BQS),
+    ok = gm:broadcast(GM, {requeue, MsgPropsFun, MsgIds}),
+    {MsgIds, State #state { backing_queue_state = BQS1 }}.
+
+len(#state { backing_queue = BQ, backing_queue_state = BQS }) ->
+    BQ:len(BQS).
+
+is_empty(#state { backing_queue = BQ, backing_queue_state = BQS }) ->
+    BQ:is_empty(BQS).
+
+set_ram_duration_target(Target, State = #state { backing_queue       = BQ,
+                                                 backing_queue_state = BQS }) ->
+    State #state { backing_queue_state =
+                       BQ:set_ram_duration_target(Target, BQS) }.
+
+ram_duration(State = #state { backing_queue = BQ, backing_queue_state = BQS }) ->
+    {Result, BQS1} = BQ:ram_duration(BQS),
+    {Result, State #state { backing_queue_state = BQS1 }}.
+
+needs_timeout(#state { backing_queue = BQ, backing_queue_state = BQS }) ->
+    BQ:needs_timeout(BQS).
+
+timeout(State = #state { backing_queue = BQ, backing_queue_state = BQS }) ->
+    State #state { backing_queue_state = BQ:timeout(BQS) }.
+
+handle_pre_hibernate(State = #state { backing_queue       = BQ,
+                                      backing_queue_state = BQS }) ->
+    State #state { backing_queue_state = BQ:handle_pre_hibernate(BQS) }.
+
+status(#state { backing_queue = BQ, backing_queue_state = BQS }) ->
+    BQ:status(BQS).
+
+invoke(?MODULE, Fun, State) ->
+    Fun(?MODULE, State);
+invoke(Mod, Fun, State = #state { backing_queue       = BQ,
+                                  backing_queue_state = BQS }) ->
+    State #state { backing_queue_state = BQ:invoke(Mod, Fun, BQS) }.
+
+is_duplicate(none, Message = #basic_message { id = MsgId },
+             State = #state { seen_status         = SS,
+                              backing_queue       = BQ,
+                              backing_queue_state = BQS,
+                              confirmed           = Confirmed }) ->
+    %% Here, we need to deal with the possibility that we're about to
+    %% receive a message that we've already seen when we were a slave
+    %% (we received it via gm). Thus if we do receive such message now
+    %% via the channel, there may be a confirm waiting to issue for
+    %% it.
+
+    %% We will never see {published, ChPid, MsgSeqNo} here.
+    case dict:find(MsgId, SS) of
+        error ->
+            %% We permit the underlying BQ to have a peek at it, but
+            %% only if we ourselves are not filtering out the msg.
+            {Result, BQS1} = BQ:is_duplicate(none, Message, BQS),
+            {Result, State #state { backing_queue_state = BQS1 }};
+        {ok, published} ->
+            %% It already got published when we were a slave and no
+            %% confirmation is waiting. amqqueue_process will have, in
+            %% its msg_id_to_channel mapping, the entry for dealing
+            %% with the confirm when that comes back in (it's added
+            %% immediately after calling is_duplicate). The msg is
+            %% invalid. We will not see this again, nor will we be
+            %% further involved in confirming this message, so erase.
+            {published, State #state { seen_status = dict:erase(MsgId, SS) }};
+        {ok, confirmed} ->
+            %% It got published when we were a slave via gm, and
+            %% confirmed some time after that (maybe even after
+            %% promotion), but before we received the publish from the
+            %% channel, so couldn't previously know what the
+            %% msg_seq_no was (and thus confirm as a slave). So we
+            %% need to confirm now. As above, amqqueue_process will
+            %% have the entry for the msg_id_to_channel mapping added
+            %% immediately after calling is_duplicate/2.
+            {published, State #state { seen_status = dict:erase(MsgId, SS),
+                                       confirmed = [MsgId | Confirmed] }};
+        {ok, discarded} ->
+            %% Don't erase from SS here because discard/2 is about to
+            %% be called and we need to be able to detect this case
+            {discarded, State}
+    end;
+is_duplicate(_Txn, _Msg, State) ->
+    %% In a transaction. We don't support txns in mirror queues. But
+    %% it's probably not a duplicate...
+    {false, State}.
+
+discard(Msg = #basic_message { id = MsgId }, ChPid,
+        State = #state { gm                  = GM,
+                         backing_queue       = BQ,
+                         backing_queue_state = BQS,
+                         seen_status         = SS }) ->
+    %% It's a massive error if we get told to discard something that's
+    %% already been published or published-and-confirmed. To do that
+    %% would require non FIFO access. Hence we should not find
+    %% 'published' or 'confirmed' in this dict:find.
+    case dict:find(MsgId, SS) of
+        error ->
+            ok = gm:broadcast(GM, {discard, ChPid, Msg}),
+            State #state { backing_queue_state = BQ:discard(Msg, ChPid, BQS),
+                           seen_status         = dict:erase(MsgId, SS) };
+        {ok, discarded} ->
+            State
+    end.
+
+maybe_store_acktag(undefined, _MsgId, AM) ->
+    AM;
+maybe_store_acktag(AckTag, MsgId, AM) ->
+    dict:store(AckTag, MsgId, AM).
+
+ensure_monitoring(ChPid, State = #state { coordinator = CPid,
+                                          known_senders = KS }) ->
+    case sets:is_element(ChPid, KS) of
+        true  -> State;
+        false -> ok = rabbit_mirror_queue_coordinator:ensure_monitoring(
+                        CPid, [ChPid]),
+                 State #state { known_senders = sets:add_element(ChPid, KS) }
+    end.
diff --git a/src/rabbit_mirror_queue_misc.erl b/src/rabbit_mirror_queue_misc.erl
new file mode 100644
index 00000000..046d3380
--- /dev/null
+++ b/src/rabbit_mirror_queue_misc.erl
@@ -0,0 +1,136 @@
+%% The contents of this file are subject to the Mozilla Public License
+%% Version 1.1 (the "License"); you may not use this file except in
+%% compliance with the License. You may obtain a copy of the License at
+%% http://www.mozilla.org/MPL/
+%%
+%% Software distributed under the License is distributed on an "AS IS"
+%% basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See the
+%% License for the specific language governing rights and limitations
+%% under the License.
+%%
+%% The Original Code is RabbitMQ.
+%%
+%% The Initial Developer of the Original Code is VMware, Inc.
+%% Copyright (c) 2007-2010 VMware, Inc.  All rights reserved.
+%%
+
+-module(rabbit_mirror_queue_misc).
+
+-export([remove_from_queue/2, on_node_up/0,
+         drop_slave/2, drop_slave/3, add_slave/2, add_slave/3]).
+
+-include("rabbit.hrl").
+
+%% If the dead pids include the queue pid (i.e. the master has died)
+%% then only remove that if we are about to be promoted. Otherwise we
+%% can have the situation where a slave updates the mnesia record for
+%% a queue, promoting another slave before that slave realises it has
+%% become the new master.
+remove_from_queue(QueueName, DeadPids) ->
+    DeadNodes = [node(DeadPid) || DeadPid <- DeadPids],
+    rabbit_misc:execute_mnesia_transaction(
+      fun () ->
+              %% Someone else could have deleted the queue before we
+              %% get here.
+              case mnesia:read({rabbit_queue, QueueName}) of
+                  [] -> {error, not_found};
+                  [Q = #amqqueue { pid         = QPid,
+                                   mirror_pids = MPids }] ->
+                      [QPid1 | MPids1] =
+                          [Pid || Pid <- [QPid | MPids],
+                                  not lists:member(node(Pid), DeadNodes)],
+                      case {{QPid, MPids}, {QPid1, MPids1}} of
+                          {Same, Same} ->
+                              ok;
+                          _ when QPid =:= QPid1 orelse node(QPid1) =:= node() ->
+                              %% Either master hasn't changed, so
+                              %% we're ok to update mnesia; or master
+                              %% has changed to become us!
+                              Q1 = Q #amqqueue { pid         = QPid1,
+                                                 mirror_pids = MPids1 },
+                              ok = rabbit_amqqueue:store_queue(Q1);
+                          _ ->
+                              %% Master has changed, and we're not it,
+                              %% so leave alone to allow the promoted
+                              %% slave to find it and make its
+                              %% promotion atomic.
+                              ok
+                      end,
+                      {ok, QPid1}
+              end
+      end).
+
+on_node_up() ->
+    Qs =
+        rabbit_misc:execute_mnesia_transaction(
+          fun () ->
+                  mnesia:foldl(
+                    fun (#amqqueue{ arguments = Args, name = QName }, QsN) ->
+                            case rabbit_misc:table_lookup(
+                                   Args, <<"x-mirror">>) of
+                                {_Type, []} ->
+                                    [QName | QsN];
+                                {_Type, Nodes} ->
+                                    Nodes1 = [list_to_atom(binary_to_list(Node))
+                                              || {longstr, Node} <- Nodes],
+                                    case lists:member(node(), Nodes1) of
+                                        true  -> [QName | QsN];
+                                        false -> QsN
+                                    end;
+                                _ ->
+                                    QsN
+                            end
+                    end, [], rabbit_queue)
+          end),
+    [add_slave(Q, node()) || Q <- Qs],
+    ok.
+
+drop_slave(VHostPath, QueueName, MirrorNode) ->
+    drop_slave(rabbit_misc:r(VHostPath, queue, QueueName), MirrorNode).
+
+drop_slave(Queue, MirrorNode) ->
+    if_mirrored_queue(
+      Queue,
+      fun (#amqqueue { name = Name, pid = QPid, mirror_pids = MPids }) ->
+              case [Pid || Pid <- [QPid | MPids], node(Pid) =:= MirrorNode] of
+                  [] ->
+                      {error, {queue_not_mirrored_on_node, MirrorNode}};
+                  [QPid | MPids] ->
+                      {error, cannot_drop_only_mirror};
+                  [Pid] ->
+                      rabbit_log:info("Dropping slave node on ~p for ~s~n",
+                                      [MirrorNode, rabbit_misc:rs(Name)]),
+                      exit(Pid, {shutdown, dropped}),
+                      ok
+              end
+      end).
+
+add_slave(VHostPath, QueueName, MirrorNode) ->
+    add_slave(rabbit_misc:r(VHostPath, queue, QueueName), MirrorNode).
+
+add_slave(Queue, MirrorNode) ->
+    if_mirrored_queue(
+      Queue,
+      fun (#amqqueue { name = Name, pid = QPid, mirror_pids = MPids } = Q) ->
+              case [Pid || Pid <- [QPid | MPids], node(Pid) =:= MirrorNode] of
+                  []  -> Result = rabbit_mirror_queue_slave_sup:start_child(
+                                    MirrorNode, [Q]),
+                         rabbit_log:info(
+                           "Adding slave node for ~s on node ~p: ~p~n",
+                           [rabbit_misc:rs(Name), MirrorNode, Result]),
+                         case Result of
+                             {ok, _Pid} -> ok;
+                             _          -> Result
+                         end;
+                  [_] -> {error, {queue_already_mirrored_on_node, MirrorNode}}
+              end
+      end).
+
+if_mirrored_queue(Queue, Fun) ->
+    rabbit_amqqueue:with(
+      Queue, fun (#amqqueue { arguments = Args } = Q) ->
+                     case rabbit_misc:table_lookup(Args, <<"x-mirror">>) of
+                         undefined -> ok;
+                         _         -> Fun(Q)
+                     end
+             end).
diff --git a/src/rabbit_mirror_queue_slave.erl b/src/rabbit_mirror_queue_slave.erl
new file mode 100644
index 00000000..c5f83c24
--- /dev/null
+++ b/src/rabbit_mirror_queue_slave.erl
@@ -0,0 +1,873 @@
+%% The contents of this file are subject to the Mozilla Public License
+%% Version 1.1 (the "License"); you may not use this file except in
+%% compliance with the License. You may obtain a copy of the License at
+%% http://www.mozilla.org/MPL/
+%%
+%% Software distributed under the License is distributed on an "AS IS"
+%% basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See the
+%% License for the specific language governing rights and limitations
+%% under the License.
+%%
+%% The Original Code is RabbitMQ.
+%%
+%% The Initial Developer of the Original Code is VMware, Inc.
+%% Copyright (c) 2007-2010 VMware, Inc.  All rights reserved.
+%%
+
+-module(rabbit_mirror_queue_slave).
+
+%% For general documentation of HA design, see
+%% rabbit_mirror_queue_coordinator
+%%
+%% We join the GM group before we add ourselves to the amqqueue
+%% record. As a result:
+%% 1. We can receive msgs from GM that correspond to messages we will
+%% never receive from publishers.
+%% 2. When we receive a message from publishers, we must receive a
+%% message from the GM group for it.
+%% 3. However, that instruction from the GM group can arrive either
+%% before or after the actual message. We need to be able to
+%% distinguish between GM instructions arriving early, and case (1)
+%% above.
+%%
+%% All instructions from the GM group must be processed in the order
+%% in which they're received.
+
+-export([start_link/1, set_maximum_since_use/2]).
+
+-export([init/1, handle_call/3, handle_cast/2, handle_info/2, terminate/2,
+         code_change/3, handle_pre_hibernate/1, prioritise_call/3,
+         prioritise_cast/2]).
+
+-export([joined/2, members_changed/3, handle_msg/3]).
+
+-behaviour(gen_server2).
+-behaviour(gm).
+
+-include("rabbit.hrl").
+-include("gm_specs.hrl").
+
+-define(SYNC_INTERVAL,                 25). %% milliseconds
+-define(RAM_DURATION_UPDATE_INTERVAL,  5000).
+-define(DEATH_TIMEOUT,                 20000). %% 20 seconds
+
+-record(state, { q,
+                 gm,
+                 master_pid,
+                 backing_queue,
+                 backing_queue_state,
+                 sync_timer_ref,
+                 rate_timer_ref,
+
+                 sender_queues, %% :: Pid -> {Q {Msg, Bool}, Set MsgId}
+                 msg_id_ack,    %% :: MsgId -> AckTag
+                 ack_num,
+
+                 msg_id_status,
+                 known_senders
+               }).
+
+start_link(Q) ->
+    gen_server2:start_link(?MODULE, [Q], []).
+
+set_maximum_since_use(QPid, Age) ->
+    gen_server2:cast(QPid, {set_maximum_since_use, Age}).
+
+init([#amqqueue { name = QueueName } = Q]) ->
+    process_flag(trap_exit, true), %% amqqueue_process traps exits too.
+    {ok, GM} = gm:start_link(QueueName, ?MODULE, [self()]),
+    receive {joined, GM} ->
+            ok
+    end,
+    Self = self(),
+    Node = node(),
+    {ok, MPid} =
+        rabbit_misc:execute_mnesia_transaction(
+          fun () ->
+                  [Q1 = #amqqueue { pid = QPid, mirror_pids = MPids }] =
+                      mnesia:read({rabbit_queue, QueueName}),
+                  %% ASSERTION
+                  [] = [Pid || Pid <- [QPid | MPids], node(Pid) =:= Node],
+                  MPids1 = MPids ++ [Self],
+                  mnesia:write(rabbit_queue,
+                               Q1 #amqqueue { mirror_pids = MPids1 },
+                               write),
+                  {ok, QPid}
+          end),
+    erlang:monitor(process, MPid),
+    ok = file_handle_cache:register_callback(
+           rabbit_amqqueue, set_maximum_since_use, [self()]),
+    ok = rabbit_memory_monitor:register(
+           self(), {rabbit_amqqueue, set_ram_duration_target, [self()]}),
+    {ok, BQ} = application:get_env(backing_queue_module),
+    BQS = bq_init(BQ, Q, false),
+    {ok, #state { q                   = Q,
+                  gm                  = GM,
+                  master_pid          = MPid,
+                  backing_queue       = BQ,
+                  backing_queue_state = BQS,
+                  rate_timer_ref      = undefined,
+                  sync_timer_ref      = undefined,
+
+                  sender_queues       = dict:new(),
+                  msg_id_ack          = dict:new(),
+                  ack_num             = 0,
+
+                  msg_id_status       = dict:new(),
+                  known_senders       = dict:new()
+                }, hibernate,
+     {backoff, ?HIBERNATE_AFTER_MIN, ?HIBERNATE_AFTER_MIN, ?DESIRED_HIBERNATE}}.
+
+handle_call({deliver_immediately, Delivery = #delivery {}}, From, State) ->
+    %% Synchronous, "immediate" delivery mode
+
+    %% It is safe to reply 'false' here even if a) we've not seen the
+    %% msg via gm, or b) the master dies before we receive the msg via
+    %% gm. In the case of (a), we will eventually receive the msg via
+    %% gm, and it's only the master's result to the channel that is
+    %% important. In the case of (b), if the master does die and we do
+    %% get promoted then at that point we have no consumers, thus
+    %% 'false' is precisely the correct answer. However, we must be
+    %% careful to _not_ enqueue the message in this case.
+
+    %% Note this is distinct from the case where we receive the msg
+    %% via gm first, then we're promoted to master, and only then do
+    %% we receive the msg from the channel.
+    gen_server2:reply(From, false), %% master may deliver it, not us
+    noreply(maybe_enqueue_message(Delivery, false, State));
+
+handle_call({deliver, Delivery = #delivery {}}, From, State) ->
+    %% Synchronous, "mandatory" delivery mode
+    gen_server2:reply(From, true), %% amqqueue throws away the result anyway
+    noreply(maybe_enqueue_message(Delivery, true, State));
+
+handle_call({gm_deaths, Deaths}, From,
+            State = #state { q          = #amqqueue { name = QueueName },
+                             gm         = GM,
+                             master_pid = MPid }) ->
+    rabbit_log:info("Mirrored-queue (~s): Slave ~s saw deaths of mirrors ~s~n",
+                    [rabbit_misc:rs(QueueName),
+                     rabbit_misc:pid_to_string(self()),
+                     [[rabbit_misc:pid_to_string(Pid), $ ] || Pid <- Deaths]]),
+    %% The GM has told us about deaths, which means we're not going to
+    %% receive any more messages from GM
+    case rabbit_mirror_queue_misc:remove_from_queue(QueueName, Deaths) of
+        {ok, Pid} when node(Pid) =:= node(MPid) ->
+            %% master hasn't changed
+            reply(ok, State);
+        {ok, Pid} when node(Pid) =:= node() ->
+            %% we've become master
+            promote_me(From, State);
+        {ok, Pid} ->
+            %% master has changed to not us.
+            gen_server2:reply(From, ok),
+            erlang:monitor(process, Pid),
+            ok = gm:broadcast(GM, heartbeat),
+            noreply(State #state { master_pid = Pid });
+        {error, not_found} ->
+            gen_server2:reply(From, ok),
+            {stop, normal, State}
+    end;
+
+handle_call({run_backing_queue, Mod, Fun}, _From, State) ->
+    reply(ok, run_backing_queue(Mod, Fun, State));
+
+handle_call({commit, _Txn, _ChPid}, _From, State) ->
+    %% We don't support transactions in mirror queues
+    reply(ok, State).
+
+handle_cast({run_backing_queue, Mod, Fun}, State) ->
+    noreply(run_backing_queue(Mod, Fun, State));
+
+handle_cast({gm, Instruction}, State) ->
+    handle_process_result(process_instruction(Instruction, State));
+
+handle_cast({deliver, Delivery = #delivery {}}, State) ->
+    %% Asynchronous, non-"mandatory", non-"immediate" deliver mode.
+    noreply(maybe_enqueue_message(Delivery, true, State));
+
+handle_cast({set_maximum_since_use, Age}, State) ->
+    ok = file_handle_cache:set_maximum_since_use(Age),
+    noreply(State);
+
+handle_cast({set_ram_duration_target, Duration},
+            State = #state { backing_queue       = BQ,
+                             backing_queue_state = BQS }) ->
+    BQS1 = BQ:set_ram_duration_target(Duration, BQS),
+    noreply(State #state { backing_queue_state = BQS1 });
+
+handle_cast(update_ram_duration,
+            State = #state { backing_queue = BQ,
+                             backing_queue_state = BQS }) ->
+    {RamDuration, BQS1} = BQ:ram_duration(BQS),
+    DesiredDuration =
+        rabbit_memory_monitor:report_ram_duration(self(), RamDuration),
+    BQS2 = BQ:set_ram_duration_target(DesiredDuration, BQS1),
+    noreply(State #state { rate_timer_ref = just_measured,
+                           backing_queue_state = BQS2 });
+
+handle_cast(sync_timeout, State) ->
+    noreply(backing_queue_timeout(
+              State #state { sync_timer_ref = undefined }));
+
+handle_cast({rollback, _Txn, _ChPid}, State) ->
+    %% We don't support transactions in mirror queues
+    noreply(State).
+
+handle_info(timeout, State) ->
+    noreply(backing_queue_timeout(State));
+
+handle_info({'DOWN', _MonitorRef, process, MPid, _Reason},
+           State = #state { gm = GM, master_pid = MPid }) ->
+    ok = gm:broadcast(GM, {process_death, MPid}),
+    noreply(State);
+
+handle_info({'DOWN', _MonitorRef, process, ChPid, _Reason}, State) ->
+    noreply(local_sender_death(ChPid, State));
+
+handle_info({'EXIT', _Pid, Reason}, State) ->
+    {stop, Reason, State};
+
+handle_info(Msg, State) ->
+    {stop, {unexpected_info, Msg}, State}.
+
+%% If the Reason is shutdown, or {shutdown, _}, it is not the queue
+%% being deleted: it's just the node going down. Even though we're a
+%% slave, we have no idea whether or not we'll be the only copy coming
+%% back up. Thus we must assume we will be, and preserve anything we
+%% have on disk.
+terminate(_Reason, #state { backing_queue_state = undefined }) ->
+    %% We've received a delete_and_terminate from gm, thus nothing to
+    %% do here.
+    ok;
+terminate({shutdown, dropped} = R, #state { backing_queue       = BQ,
+                                            backing_queue_state = BQS }) ->
+    %% See rabbit_mirror_queue_master:terminate/2
+    BQ:delete_and_terminate(R, BQS);
+terminate(Reason, #state { q                   = Q,
+                           gm                  = GM,
+                           backing_queue       = BQ,
+                           backing_queue_state = BQS,
+                           rate_timer_ref      = RateTRef }) ->
+    ok = gm:leave(GM),
+    QueueState = rabbit_amqqueue_process:init_with_backing_queue_state(
+                   Q, BQ, BQS, RateTRef, [], [], dict:new()),
+    rabbit_amqqueue_process:terminate(Reason, QueueState);
+terminate([_SPid], _Reason) ->
+    %% gm case
+    ok.
+
+code_change(_OldVsn, State, _Extra) ->
+    {ok, State}.
+
+handle_pre_hibernate(State = #state { backing_queue       = BQ,
+                                      backing_queue_state = BQS }) ->
+    {RamDuration, BQS1} = BQ:ram_duration(BQS),
+    DesiredDuration =
+        rabbit_memory_monitor:report_ram_duration(self(), RamDuration),
+    BQS2 = BQ:set_ram_duration_target(DesiredDuration, BQS1),
+    BQS3 = BQ:handle_pre_hibernate(BQS2),
+    {hibernate, stop_rate_timer(State #state { backing_queue_state = BQS3 })}.
+
+prioritise_call(Msg, _From, _State) ->
+    case Msg of
+        {run_backing_queue, _Mod, _Fun}      -> 6;
+        {gm_deaths, _Deaths}                 -> 5;
+        _                                    -> 0
+    end.
+
+prioritise_cast(Msg, _State) ->
+    case Msg of
+        update_ram_duration                  -> 8;
+        {set_ram_duration_target, _Duration} -> 8;
+        {set_maximum_since_use, _Age}        -> 8;
+        {run_backing_queue, _Mod, _Fun}      -> 6;
+        sync_timeout                         -> 6;
+        {gm, _Msg}                           -> 5;
+        {post_commit, _Txn, _AckTags}        -> 4;
+        _                                    -> 0
+    end.
+
+%% ---------------------------------------------------------------------------
+%% GM
+%% ---------------------------------------------------------------------------
+
+joined([SPid], _Members) ->
+    SPid ! {joined, self()},
+    ok.
+
+members_changed([_SPid], _Births, []) ->
+    ok;
+members_changed([SPid], _Births, Deaths) ->
+    inform_deaths(SPid, Deaths).
+
+handle_msg([_SPid], _From, heartbeat) ->
+    ok;
+handle_msg([_SPid], _From, {ensure_monitoring, _Pid}) ->
+    %% This is only of value to the master
+    ok;
+handle_msg([SPid], _From, {process_death, Pid}) ->
+    inform_deaths(SPid, [Pid]);
+handle_msg([SPid], _From, Msg) ->
+    ok = gen_server2:cast(SPid, {gm, Msg}).
+
+inform_deaths(SPid, Deaths) ->
+    rabbit_misc:with_exit_handler(
+      fun () -> {stop, normal} end,
+      fun () ->
+              case gen_server2:call(SPid, {gm_deaths, Deaths}, infinity) of
+                  ok ->
+                      ok;
+                  {promote, CPid} ->
+                      {become, rabbit_mirror_queue_coordinator, [CPid]}
+              end
+      end).
+
+%% ---------------------------------------------------------------------------
+%% Others
+%% ---------------------------------------------------------------------------
+
+bq_init(BQ, Q, Recover) ->
+    Self = self(),
+    BQ:init(Q, Recover,
+            fun (Mod, Fun) ->
+                    rabbit_amqqueue:run_backing_queue_async(Self, Mod, Fun)
+            end,
+            fun (Mod, Fun) ->
+                    rabbit_misc:with_exit_handler(
+                      fun () -> error end,
+                      fun () ->
+                              rabbit_amqqueue:run_backing_queue(Self, Mod, Fun)
+                      end)
+            end).
+
+run_backing_queue(rabbit_mirror_queue_master, Fun, State) ->
+    %% Yes, this might look a little crazy, but see comments in
+    %% confirm_sender_death/1
+    Fun(?MODULE, State);
+run_backing_queue(Mod, Fun, State = #state { backing_queue       = BQ,
+                                             backing_queue_state = BQS }) ->
+    State #state { backing_queue_state = BQ:invoke(Mod, Fun, BQS) }.
+
+needs_confirming(#delivery{ msg_seq_no = undefined }, _State) ->
+    never;
+needs_confirming(#delivery { message = #basic_message {
+                               is_persistent = true } },
+                 #state { q = #amqqueue { durable = true } }) ->
+    eventually;
+needs_confirming(_Delivery, _State) ->
+    immediately.
+
+confirm_messages(MsgIds, State = #state { msg_id_status = MS }) ->
+    {MS1, CMs} =
+        lists:foldl(
+          fun (MsgId, {MSN, CMsN} = Acc) ->
+                  %% We will never see 'discarded' here
+                  case dict:find(MsgId, MSN) of
+                      error ->
+                          %% If it needed confirming, it'll have
+                          %% already been done.
+                          Acc;
+                      {ok, {published, ChPid}} ->
+                          %% Still not seen it from the channel, just
+                          %% record that it's been confirmed.
+                          {dict:store(MsgId, {confirmed, ChPid}, MSN), CMsN};
+                      {ok, {published, ChPid, MsgSeqNo}} ->
+                          %% Seen from both GM and Channel. Can now
+                          %% confirm.
+                          {dict:erase(MsgId, MSN),
+                           gb_trees_cons(ChPid, MsgSeqNo, CMsN)};
+                      {ok, {confirmed, _ChPid}} ->
+                          %% It's already been confirmed. This is
+                          %% probably it's been both sync'd to disk
+                          %% and then delivered and ack'd before we've
+                          %% seen the publish from the
+                          %% channel. Nothing to do here.
+                          Acc
+                  end
+          end, {MS, gb_trees:empty()}, MsgIds),
+    [ok = rabbit_channel:confirm(ChPid, MsgSeqNos)
+     || {ChPid, MsgSeqNos} <- gb_trees:to_list(CMs)],
+    State #state { msg_id_status = MS1 }.
+
+gb_trees_cons(Key, Value, Tree) ->
+    case gb_trees:lookup(Key, Tree) of
+        {value, Values} -> gb_trees:update(Key, [Value | Values], Tree);
+        none            -> gb_trees:insert(Key, [Value], Tree)
+    end.
+
+handle_process_result({ok,   State}) -> noreply(State);
+handle_process_result({stop, State}) -> {stop, normal, State}.
+
+promote_me(From, #state { q                   = Q,
+                          gm                  = GM,
+                          backing_queue       = BQ,
+                          backing_queue_state = BQS,
+                          rate_timer_ref      = RateTRef,
+                          sender_queues       = SQ,
+                          msg_id_ack          = MA,
+                          msg_id_status       = MS,
+                          known_senders       = KS }) ->
+    rabbit_log:info("Mirrored-queue (~s): Promoting slave ~s to master~n",
+                    [rabbit_misc:rs(Q #amqqueue.name),
+                     rabbit_misc:pid_to_string(self())]),
+    Q1 = Q #amqqueue { pid = self() },
+    {ok, CPid} = rabbit_mirror_queue_coordinator:start_link(
+                   Q1, GM, rabbit_mirror_queue_master:sender_death_fun()),
+    true = unlink(GM),
+    gen_server2:reply(From, {promote, CPid}),
+    ok = gm:confirmed_broadcast(GM, heartbeat),
+
+    %% Everything that we're monitoring, we need to ensure our new
+    %% coordinator is monitoring.
+
+    MonitoringPids = [begin true = erlang:demonitor(MRef),
+                            Pid
+                      end || {Pid, MRef} <- dict:to_list(KS)],
+    ok = rabbit_mirror_queue_coordinator:ensure_monitoring(
+           CPid, MonitoringPids),
+
+    %% We find all the messages that we've received from channels but
+    %% not from gm, and if they're due to be enqueued on promotion
+    %% then we pass them to the
+    %% queue_process:init_with_backing_queue_state to be enqueued.
+    %%
+    %% We also have to requeue messages which are pending acks: the
+    %% consumers from the master queue have been lost and so these
+    %% messages need requeuing. They might also be pending
+    %% confirmation, and indeed they might also be pending arrival of
+    %% the publication from the channel itself, if we received both
+    %% the publication and the fetch via gm first! Requeuing doesn't
+    %% affect confirmations: if the message was previously pending a
+    %% confirmation then it still will be, under the same msg_id. So
+    %% as a master, we need to be prepared to filter out the
+    %% publication of said messages from the channel (is_duplicate
+    %% (thus such requeued messages must remain in the msg_id_status
+    %% (MS) which becomes seen_status (SS) in the master)).
+    %%
+    %% Then there are messages we already have in the queue, which are
+    %% not currently pending acknowledgement:
+    %% 1. Messages we've only received via gm:
+    %%    Filter out subsequent publication from channel through
+    %%    validate_message. Might have to issue confirms then or
+    %%    later, thus queue_process state will have to know that
+    %%    there's a pending confirm.
+    %% 2. Messages received via both gm and channel:
+    %%    Queue will have to deal with issuing confirms if necessary.
+    %%
+    %% MS contains the following three entry types:
+    %%
+    %% a) {published, ChPid}:
+    %%   published via gm only; pending arrival of publication from
+    %%   channel, maybe pending confirm.
+    %%
+    %% b) {published, ChPid, MsgSeqNo}:
+    %%   published via gm and channel; pending confirm.
+    %%
+    %% c) {confirmed, ChPid}:
+    %%   published via gm only, and confirmed; pending publication
+    %%   from channel.
+    %%
+    %% d) discarded
+    %%   seen via gm only as discarded. Pending publication from
+    %%   channel
+    %%
+    %% The forms a, c and d only, need to go to the master state
+    %% seen_status (SS).
+    %%
+    %% The form b only, needs to go through to the queue_process
+    %% state to form the msg_id_to_channel mapping (MTC).
+    %%
+    %% No messages that are enqueued from SQ at this point will have
+    %% entries in MS.
+    %%
+    %% Messages that are extracted from MA may have entries in MS, and
+    %% those messages are then requeued. However, as discussed above,
+    %% this does not affect MS, nor which bits go through to SS in
+    %% Master, or MTC in queue_process.
+    %%
+    %% Everything that's in MA gets requeued. Consequently the new
+    %% master should start with a fresh AM as there are no messages
+    %% pending acks (txns will have been rolled back).
+
+    MSList = dict:to_list(MS),
+    SS = dict:from_list(
+           [E || E = {_MsgId, discarded} <- MSList] ++
+               [{MsgId, Status}
+                || {MsgId, {Status, _ChPid}} <- MSList,
+                   Status =:= published orelse Status =:= confirmed]),
+
+    MasterState = rabbit_mirror_queue_master:promote_backing_queue_state(
+                    CPid, BQ, BQS, GM, SS, MonitoringPids),
+
+    MTC = dict:from_list(
+            [{MsgId, {ChPid, MsgSeqNo}} ||
+                {MsgId, {published, ChPid, MsgSeqNo}} <- dict:to_list(MS)]),
+    NumAckTags = [NumAckTag || {_MsgId, NumAckTag} <- dict:to_list(MA)],
+    AckTags = [AckTag || {_Num, AckTag} <- lists:sort(NumAckTags)],
+    Deliveries = [Delivery || {_ChPid, {PubQ, _PendCh}} <- dict:to_list(SQ),
+                              {Delivery, true} <- queue:to_list(PubQ)],
+    QueueState = rabbit_amqqueue_process:init_with_backing_queue_state(
+                   Q1, rabbit_mirror_queue_master, MasterState, RateTRef,
+                   AckTags, Deliveries, MTC),
+    {become, rabbit_amqqueue_process, QueueState, hibernate}.
+
+noreply(State) ->
+    {NewState, Timeout} = next_state(State),
+    {noreply, NewState, Timeout}.
+
+reply(Reply, State) ->
+    {NewState, Timeout} = next_state(State),
+    {reply, Reply, NewState, Timeout}.
+
+next_state(State = #state{backing_queue = BQ, backing_queue_state = BQS}) ->
+    {MsgIds, BQS1} = BQ:drain_confirmed(BQS),
+    State1 = ensure_rate_timer(
+               confirm_messages(MsgIds, State #state {
+                                          backing_queue_state = BQS1 })),
+    case BQ:needs_timeout(BQS1) of
+        false -> {stop_sync_timer(State1),   hibernate};
+        idle  -> {stop_sync_timer(State1),   0        };
+        timed -> {ensure_sync_timer(State1), 0        }
+    end.
+
+backing_queue_timeout(State = #state { backing_queue = BQ }) ->
+    run_backing_queue(BQ, fun (M, BQS) -> M:timeout(BQS) end, State).
+
+ensure_sync_timer(State = #state { sync_timer_ref = undefined }) ->
+    {ok, TRef} = timer:apply_after(
+                   ?SYNC_INTERVAL, rabbit_amqqueue, sync_timeout, [self()]),
+    State #state { sync_timer_ref = TRef };
+ensure_sync_timer(State) ->
+    State.
+
+stop_sync_timer(State = #state { sync_timer_ref = undefined }) ->
+    State;
+stop_sync_timer(State = #state { sync_timer_ref = TRef }) ->
+    {ok, cancel} = timer:cancel(TRef),
+    State #state { sync_timer_ref = undefined }.
+
+ensure_rate_timer(State = #state { rate_timer_ref = undefined }) ->
+    {ok, TRef} = timer:apply_after(
+                   ?RAM_DURATION_UPDATE_INTERVAL,
+                   rabbit_amqqueue, update_ram_duration,
+                   [self()]),
+    State #state { rate_timer_ref = TRef };
+ensure_rate_timer(State = #state { rate_timer_ref = just_measured }) ->
+    State #state { rate_timer_ref = undefined };
+ensure_rate_timer(State) ->
+    State.
+
+stop_rate_timer(State = #state { rate_timer_ref = undefined }) ->
+    State;
+stop_rate_timer(State = #state { rate_timer_ref = just_measured }) ->
+    State #state { rate_timer_ref = undefined };
+stop_rate_timer(State = #state { rate_timer_ref = TRef }) ->
+    {ok, cancel} = timer:cancel(TRef),
+    State #state { rate_timer_ref = undefined }.
+
+ensure_monitoring(ChPid, State = #state { known_senders = KS }) ->
+    case dict:is_key(ChPid, KS) of
+        true  -> State;
+        false -> MRef = erlang:monitor(process, ChPid),
+                 State #state { known_senders = dict:store(ChPid, MRef, KS) }
+    end.
+
+local_sender_death(ChPid, State = #state { known_senders = KS }) ->
+    ok = case dict:is_key(ChPid, KS) of
+             false -> ok;
+             true  -> confirm_sender_death(ChPid)
+         end,
+    State.
+
+confirm_sender_death(Pid) ->
+    %% We have to deal with the possibility that we'll be promoted to
+    %% master before this thing gets run. Consequently we set the
+    %% module to rabbit_mirror_queue_master so that if we do become a
+    %% rabbit_amqqueue_process before then, sane things will happen.
+    Fun =
+        fun (?MODULE, State = #state { known_senders = KS,
+                                       gm            = GM }) ->
+                %% We're running still as a slave
+                ok = case dict:is_key(Pid, KS) of
+                         false -> ok;
+                         true  -> gm:broadcast(GM, {ensure_monitoring, [Pid]}),
+                                  confirm_sender_death(Pid)
+                     end,
+                State;
+            (rabbit_mirror_queue_master, State) ->
+                %% We've become a master. State is now opaque to
+                %% us. When we became master, if Pid was still known
+                %% to us then we'd have set up monitoring of it then,
+                %% so this is now a noop.
+                State
+        end,
+    %% Note that we do not remove our knowledge of this ChPid until we
+    %% get the sender_death from GM.
+    {ok, _TRef} = timer:apply_after(
+                    ?DEATH_TIMEOUT, rabbit_amqqueue, run_backing_queue_async,
+                    [self(), rabbit_mirror_queue_master, Fun]),
+    ok.
+
+maybe_enqueue_message(
+  Delivery = #delivery { message    = #basic_message { id = MsgId },
+                         msg_seq_no = MsgSeqNo,
+                         sender     = ChPid,
+                         txn        = none },
+  EnqueueOnPromotion,
+  State = #state { sender_queues = SQ, msg_id_status = MS }) ->
+    State1 = ensure_monitoring(ChPid, State),
+    %% We will never see {published, ChPid, MsgSeqNo} here.
+    case dict:find(MsgId, MS) of
+        error ->
+            {MQ, PendingCh} = get_sender_queue(ChPid, SQ),
+            MQ1 = queue:in({Delivery, EnqueueOnPromotion}, MQ),
+            SQ1 = dict:store(ChPid, {MQ1, PendingCh}, SQ),
+            State1 #state { sender_queues = SQ1 };
+        {ok, {confirmed, ChPid}} ->
+            %% BQ has confirmed it but we didn't know what the
+            %% msg_seq_no was at the time. We do now!
+            ok = rabbit_channel:confirm(ChPid, [MsgSeqNo]),
+            SQ1 = remove_from_pending_ch(MsgId, ChPid, SQ),
+            State1 #state { sender_queues = SQ1,
+                            msg_id_status = dict:erase(MsgId, MS) };
+        {ok, {published, ChPid}} ->
+            %% It was published to the BQ and we didn't know the
+            %% msg_seq_no so couldn't confirm it at the time.
+            case needs_confirming(Delivery, State1) of
+                never ->
+                    SQ1 = remove_from_pending_ch(MsgId, ChPid, SQ),
+                    State1 #state { msg_id_status = dict:erase(MsgId, MS),
+                                    sender_queues = SQ1 };
+                eventually ->
+                    State1 #state {
+                      msg_id_status =
+                          dict:store(MsgId, {published, ChPid, MsgSeqNo}, MS) };
+                immediately ->
+                    ok = rabbit_channel:confirm(ChPid, [MsgSeqNo]),
+                    SQ1 = remove_from_pending_ch(MsgId, ChPid, SQ),
+                    State1 #state { msg_id_status = dict:erase(MsgId, MS),
+                                    sender_queues = SQ1 }
+            end;
+        {ok, discarded} ->
+            %% We've already heard from GM that the msg is to be
+            %% discarded. We won't see this again.
+            SQ1 = remove_from_pending_ch(MsgId, ChPid, SQ),
+            State1 #state { msg_id_status = dict:erase(MsgId, MS),
+                            sender_queues = SQ1 }
+    end;
+maybe_enqueue_message(_Delivery, _EnqueueOnPromotion, State) ->
+    %% We don't support txns in mirror queues.
+    State.
+
+get_sender_queue(ChPid, SQ) ->
+    case dict:find(ChPid, SQ) of
+        error     -> {queue:new(), sets:new()};
+        {ok, Val} -> Val
+    end.
+
+remove_from_pending_ch(MsgId, ChPid, SQ) ->
+    case dict:find(ChPid, SQ) of
+        error ->
+            SQ;
+        {ok, {MQ, PendingCh}} ->
+            dict:store(ChPid, {MQ, sets:del_element(MsgId, PendingCh)}, SQ)
+    end.
+
+process_instruction(
+  {publish, Deliver, ChPid, MsgProps, Msg = #basic_message { id = MsgId }},
+  State = #state { sender_queues       = SQ,
+                   backing_queue       = BQ,
+                   backing_queue_state = BQS,
+                   msg_id_status       = MS }) ->
+
+    %% We really are going to do the publish right now, even though we
+    %% may not have seen it directly from the channel. As a result, we
+    %% may know that it needs confirming without knowing its
+    %% msg_seq_no, which means that we can see the confirmation come
+    %% back from the backing queue without knowing the msg_seq_no,
+    %% which means that we're going to have to hang on to the fact
+    %% that we've seen the msg_id confirmed until we can associate it
+    %% with a msg_seq_no.
+    State1 = ensure_monitoring(ChPid, State),
+    {MQ, PendingCh} = get_sender_queue(ChPid, SQ),
+    {MQ1, PendingCh1, MS1} =
+        case queue:out(MQ) of
+            {empty, _MQ2} ->
+                {MQ, sets:add_element(MsgId, PendingCh),
+                 dict:store(MsgId, {published, ChPid}, MS)};
+            {{value, {Delivery = #delivery {
+                        msg_seq_no = MsgSeqNo,
+                        message    = #basic_message { id = MsgId } },
+                      _EnqueueOnPromotion}}, MQ2} ->
+                %% We received the msg from the channel first. Thus we
+                %% need to deal with confirms here.
+                case needs_confirming(Delivery, State1) of
+                    never ->
+                        {MQ2, PendingCh, MS};
+                    eventually ->
+                        {MQ2, sets:add_element(MsgId, PendingCh),
+                         dict:store(MsgId, {published, ChPid, MsgSeqNo}, MS)};
+                    immediately ->
+                        ok = rabbit_channel:confirm(ChPid, [MsgSeqNo]),
+                        {MQ2, PendingCh, MS}
+                end;
+            {{value, {#delivery {}, _EnqueueOnPromotion}}, _MQ2} ->
+                %% The instruction was sent to us before we were
+                %% within the mirror_pids within the #amqqueue{}
+                %% record. We'll never receive the message directly
+                %% from the channel. And the channel will not be
+                %% expecting any confirms from us.
+                {MQ, PendingCh, MS}
+        end,
+
+    SQ1 = dict:store(ChPid, {MQ1, PendingCh1}, SQ),
+    State2 = State1 #state { sender_queues = SQ1, msg_id_status = MS1 },
+
+    {ok,
+     case Deliver of
+         false ->
+             BQS1 = BQ:publish(Msg, MsgProps, ChPid, BQS),
+             State2 #state { backing_queue_state = BQS1 };
+         {true, AckRequired} ->
+             {AckTag, BQS1} = BQ:publish_delivered(AckRequired, Msg, MsgProps,
+                                                   ChPid, BQS),
+             maybe_store_ack(AckRequired, MsgId, AckTag,
+                             State2 #state { backing_queue_state = BQS1 })
+     end};
+process_instruction({discard, ChPid, Msg = #basic_message { id = MsgId }},
+                    State = #state { sender_queues       = SQ,
+                                     backing_queue       = BQ,
+                                     backing_queue_state = BQS,
+                                     msg_id_status       = MS }) ->
+    %% Many of the comments around the publish head above apply here
+    %% too.
+    State1 = ensure_monitoring(ChPid, State),
+    {MQ, PendingCh} = get_sender_queue(ChPid, SQ),
+    {MQ1, PendingCh1, MS1} =
+        case queue:out(MQ) of
+            {empty, _MQ} ->
+                {MQ, sets:add_element(MsgId, PendingCh),
+                 dict:store(MsgId, discarded, MS)};
+            {{value, {#delivery { message = #basic_message { id = MsgId } },
+                      _EnqueueOnPromotion}}, MQ2} ->
+                %% We've already seen it from the channel, we're not
+                %% going to see this again, so don't add it to MS
+                {MQ2, PendingCh, MS};
+            {{value, {#delivery {}, _EnqueueOnPromotion}}, _MQ2} ->
+                %% The instruction was sent to us before we were
+                %% within the mirror_pids within the #amqqueue{}
+                %% record. We'll never receive the message directly
+                %% from the channel.
+                {MQ, PendingCh, MS}
+        end,
+    SQ1 = dict:store(ChPid, {MQ1, PendingCh1}, SQ),
+    BQS1 = BQ:discard(Msg, ChPid, BQS),
+    {ok, State1 #state { sender_queues       = SQ1,
+                         msg_id_status       = MS1,
+                         backing_queue_state = BQS1 }};
+process_instruction({set_length, Length},
+                    State = #state { backing_queue       = BQ,
+                                     backing_queue_state = BQS }) ->
+    QLen = BQ:len(BQS),
+    ToDrop = QLen - Length,
+    {ok, case ToDrop > 0 of
+             true  -> BQS1 =
+                          lists:foldl(
+                            fun (const, BQSN) ->
+                                    {{_Msg, _IsDelivered, _AckTag, _Remaining},
+                                     BQSN1} = BQ:fetch(false, BQSN),
+                                    BQSN1
+                            end, BQS, lists:duplicate(ToDrop, const)),
+                      State #state { backing_queue_state = BQS1 };
+             false -> State
+         end};
+process_instruction({fetch, AckRequired, MsgId, Remaining},
+                    State = #state { backing_queue       = BQ,
+                                     backing_queue_state = BQS }) ->
+    QLen = BQ:len(BQS),
+    {ok, case QLen - 1 of
+             Remaining ->
+                 {{#basic_message{id = MsgId}, _IsDelivered,
+                   AckTag, Remaining}, BQS1} = BQ:fetch(AckRequired, BQS),
+                 maybe_store_ack(AckRequired, MsgId, AckTag,
+                                 State #state { backing_queue_state = BQS1 });
+             Other when Other < Remaining ->
+                 %% we must be shorter than the master
+                 State
+         end};
+process_instruction({ack, MsgIds},
+                    State = #state { backing_queue       = BQ,
+                                     backing_queue_state = BQS,
+                                     msg_id_ack          = MA }) ->
+    {AckTags, MA1} = msg_ids_to_acktags(MsgIds, MA),
+    {MsgIds1, BQS1} = BQ:ack(AckTags, BQS),
+    [] = MsgIds1 -- MsgIds, %% ASSERTION
+    {ok, State #state { msg_id_ack          = MA1,
+                        backing_queue_state = BQS1 }};
+process_instruction({requeue, MsgPropsFun, MsgIds},
+                    State = #state { backing_queue       = BQ,
+                                     backing_queue_state = BQS,
+                                     msg_id_ack          = MA }) ->
+    {AckTags, MA1} = msg_ids_to_acktags(MsgIds, MA),
+    {ok, case length(AckTags) =:= length(MsgIds) of
+             true ->
+                 {MsgIds, BQS1} = BQ:requeue(AckTags, MsgPropsFun, BQS),
+                 State #state { msg_id_ack          = MA1,
+                                backing_queue_state = BQS1 };
+             false ->
+                 %% The only thing we can safely do is nuke out our BQ
+                 %% and MA. The interaction between this and confirms
+                 %% doesn't really bear thinking about...
+                 {_Count, BQS1} = BQ:purge(BQS),
+                 {_MsgIds, BQS2} = ack_all(BQ, MA, BQS1),
+                 State #state { msg_id_ack          = dict:new(),
+                                backing_queue_state = BQS2 }
+         end};
+process_instruction({sender_death, ChPid},
+                    State = #state { sender_queues = SQ,
+                                     msg_id_status = MS,
+                                     known_senders = KS }) ->
+    {ok, case dict:find(ChPid, KS) of
+             error ->
+                 State;
+             {ok, MRef} ->
+                 true = erlang:demonitor(MRef),
+                 MS1 = case dict:find(ChPid, SQ) of
+                           error ->
+                               MS;
+                           {ok, {_MQ, PendingCh}} ->
+                               lists:foldl(fun dict:erase/2, MS,
+                                           sets:to_list(PendingCh))
+                       end,
+                 State #state { sender_queues = dict:erase(ChPid, SQ),
+                                msg_id_status = MS1,
+                                known_senders = dict:erase(ChPid, KS) }
+         end};
+process_instruction({delete_and_terminate, Reason},
+                    State = #state { backing_queue       = BQ,
+                                     backing_queue_state = BQS }) ->
+    BQ:delete_and_terminate(Reason, BQS),
+    {stop, State #state { backing_queue_state = undefined }}.
+
+msg_ids_to_acktags(MsgIds, MA) ->
+    {AckTags, MA1} =
+        lists:foldl(
+          fun (MsgId, {Acc, MAN}) ->
+                  case dict:find(MsgId, MA) of
+                      error                -> {Acc, MAN};
+                      {ok, {_Num, AckTag}} -> {[AckTag | Acc],
+                                               dict:erase(MsgId, MAN)}
+                  end
+          end, {[], MA}, MsgIds),
+    {lists:reverse(AckTags), MA1}.
+
+ack_all(BQ, MA, BQS) ->
+    BQ:ack([AckTag || {_MsgId, {_Num, AckTag}} <- dict:to_list(MA)], BQS).
+
+maybe_store_ack(false, _MsgId, _AckTag, State) ->
+    State;
+maybe_store_ack(true, MsgId, AckTag, State = #state { msg_id_ack = MA,
+                                                      ack_num    = Num }) ->
+    State #state { msg_id_ack = dict:store(MsgId, {Num, AckTag}, MA),
+                   ack_num    = Num + 1 }.
diff --git a/src/rabbit_mirror_queue_slave_sup.erl b/src/rabbit_mirror_queue_slave_sup.erl
new file mode 100644
index 00000000..2ce5941e
--- /dev/null
+++ b/src/rabbit_mirror_queue_slave_sup.erl
@@ -0,0 +1,60 @@
+%% The contents of this file are subject to the Mozilla Public License
+%% Version 1.1 (the "License"); you may not use this file except in
+%% compliance with the License. You may obtain a copy of the License at
+%% http://www.mozilla.org/MPL/
+%%
+%% Software distributed under the License is distributed on an "AS IS"
+%% basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See the
+%% License for the specific language governing rights and limitations
+%% under the License.
+%%
+%% The Original Code is RabbitMQ.
+%%
+%% The Initial Developer of the Original Code is VMware, Inc.
+%% Copyright (c) 2007-2010 VMware, Inc.  All rights reserved.
+%%
+
+-module(rabbit_mirror_queue_slave_sup).
+
+-rabbit_boot_step({mirror_queue_slave_sup,
+                   [{description, "mirror queue slave sup"},
+                    {mfa,         {rabbit_mirror_queue_slave_sup, start, []}},
+                    {requires,    recovery},
+                    {enables,     routing_ready}]}).
+
+-rabbit_boot_step({mirrored_queues,
+                   [{description, "adding mirrors to queues"},
+                    {mfa,         {rabbit_mirror_queue_misc, on_node_up, []}},
+                    {requires,    mirror_queue_slave_sup},
+                    {enables,     routing_ready}]}).
+
+-behaviour(supervisor2).
+
+-export([start/0, start_link/0, start_child/2]).
+
+-export([init/1]).
+
+-include_lib("rabbit.hrl").
+
+-define(SERVER, ?MODULE).
+
+start() ->
+    {ok, _} =
+        supervisor2:start_child(
+          rabbit_sup,
+          {rabbit_mirror_queue_slave_sup,
+           {rabbit_mirror_queue_slave_sup, start_link, []},
+           transient, infinity, supervisor, [rabbit_mirror_queue_slave_sup]}),
+    ok.
+
+start_link() ->
+    supervisor2:start_link({local, ?SERVER}, ?MODULE, []).
+
+start_child(Node, Args) ->
+    supervisor2:start_child({?SERVER, Node}, Args).
+
+init([]) ->
+    {ok, {{simple_one_for_one_terminate, 10, 10},
+          [{rabbit_mirror_queue_slave,
+            {rabbit_mirror_queue_slave, start_link, []},
+            temporary, ?MAX_WAIT, worker, [rabbit_mirror_queue_slave]}]}}.
diff --git a/src/rabbit_mnesia.erl b/src/rabbit_mnesia.erl
index 568b9ce6..8d5c8646 100644
--- a/src/rabbit_mnesia.erl
+++ b/src/rabbit_mnesia.erl
@@ -241,7 +241,8 @@ table_definitions() ->
      {rabbit_queue,
       [{record_name, amqqueue},
        {attributes, record_info(fields, amqqueue)},
-       {match, #amqqueue{name = queue_name_match(), _='_'}}]}].
+       {match, #amqqueue{name = queue_name_match(), _='_'}}]}]
+        ++ gm:table_definitions().
 
 binding_match() ->
     #binding{source = exchange_name_match(),
diff --git a/src/rabbit_router.erl b/src/rabbit_router.erl
index 8f166672..b1d940d2 100644
--- a/src/rabbit_router.erl
+++ b/src/rabbit_router.erl
@@ -110,8 +110,10 @@ check_delivery(_   , _   , {_    , Qs}) -> {routed, Qs}.
 lookup_qpids(QNames) ->
     lists:foldl(fun (QName, QPids) ->
                         case mnesia:dirty_read({rabbit_queue, QName}) of
-                            [#amqqueue{pid = QPid}] -> [QPid | QPids];
-                            []                      -> QPids
+                            [#amqqueue{pid = QPid, mirror_pids = MPids}] ->
+                                MPids ++ [QPid | QPids];
+                            [] ->
+                                QPids
                         end
                 end, [], QNames).
 
diff --git a/src/rabbit_types.erl b/src/rabbit_types.erl
index a18118e3..22204100 100644
--- a/src/rabbit_types.erl
+++ b/src/rabbit_types.erl
@@ -124,7 +124,8 @@
                   auto_delete     :: boolean(),
                   exclusive_owner :: rabbit_types:maybe(pid()),
                   arguments       :: rabbit_framing:amqp_table(),
-                  pid             :: rabbit_types:maybe(pid())}).
+                  pid             :: rabbit_types:maybe(pid()),
+                  mirror_pids     :: [pid()]}).
 
 -type(exchange() ::
         #exchange{name        :: rabbit_exchange:name(),
diff --git a/src/rabbit_upgrade_functions.erl b/src/rabbit_upgrade_functions.erl
index c2dd860a..b4ac3328 100644
--- a/src/rabbit_upgrade_functions.erl
+++ b/src/rabbit_upgrade_functions.erl
@@ -30,6 +30,8 @@
 -rabbit_upgrade({exchange_event_serial, mnesia, []}).
 -rabbit_upgrade({trace_exchanges,       mnesia, []}).
 -rabbit_upgrade({user_admin_to_tags,    mnesia, [user_to_internal_user]}).
+-rabbit_upgrade({mirror_pids,           mnesia, []}).
+-rabbit_upgrade({gm,                    mnesia, []}).
 
 %% -------------------------------------------------------------------
 
@@ -45,6 +47,8 @@
 -spec(exchange_event_serial/0 :: () -> 'ok').
 -spec(trace_exchanges/0       :: () -> 'ok').
 -spec(user_admin_to_tags/0    :: () -> 'ok').
+-spec(mirror_pids/0           :: () -> 'ok').
+-spec(gm/0                    :: () -> 'ok').
 
 -endif.
 
@@ -133,6 +137,23 @@ user_admin_to_tags() ->
       end,
       [username, password_hash, tags], internal_user).
 
+mirror_pids() ->
+    Tables = [rabbit_queue, rabbit_durable_queue],
+    AddMirrorPidsFun =
+        fun ({amqqueue, Name, Durable, AutoDelete, Owner, Arguments, Pid}) ->
+                {amqqueue, Name, Durable, AutoDelete, Owner, Arguments, Pid, []}
+        end,
+    [ ok = transform(T,
+                     AddMirrorPidsFun,
+                     [name, durable, auto_delete, exclusive_owner, arguments,
+                      pid, mirror_pids])
+      || T <- Tables ],
+    ok.
+
+gm() ->
+    create(gm_group, [{record_name, gm_group},
+                      {attributes, [name, version, members]}]).
+
 %%--------------------------------------------------------------------
 
 transform(TableName, Fun, FieldList) ->