diff options
Diffstat (limited to 'lib/diameter/src/base/diameter_service.erl')
-rw-r--r-- | lib/diameter/src/base/diameter_service.erl | 381 |
1 files changed, 294 insertions, 87 deletions
diff --git a/lib/diameter/src/base/diameter_service.erl b/lib/diameter/src/base/diameter_service.erl index fa9a730f4f..725cccda1e 100644 --- a/lib/diameter/src/base/diameter_service.erl +++ b/lib/diameter/src/base/diameter_service.erl @@ -43,8 +43,7 @@ subscriptions/0, services/0, services/1, - whois/1, - flush_stats/1]). + whois/1]). %% test/debug -export([call_module/3, @@ -65,9 +64,27 @@ -include_lib("diameter/include/diameter.hrl"). -include("diameter_internal.hrl"). +%% The "old" states maintained in this module historically. -define(STATE_UP, up). -define(STATE_DOWN, down). +-type op_state() :: ?STATE_UP + | ?STATE_DOWN. + +%% The RFC 3539 watchdog states that are now maintained, albeit +%% along with the old up/down. okay = up, else down. +-define(WD_INITIAL, initial). +-define(WD_OKAY, okay). +-define(WD_SUSPECT, suspect). +-define(WD_DOWN, down). +-define(WD_REOPEN, reopen). + +-type wd_state() :: ?WD_INITIAL + | ?WD_OKAY + | ?WD_SUSPECT + | ?WD_DOWN + | ?WD_REOPEN. + -define(DEFAULT_TC, 30000). %% RFC 3588 ch 2.1 -define(DEFAULT_TIMEOUT, 5000). %% for outgoing requests -define(RESTART_TC, 1000). %% if restart was this recent @@ -119,7 +136,8 @@ type :: match(connect | accept), ref :: match(reference()), %% key into diameter_config options :: match([diameter:transport_opt()]),%% from start_transport - op_state = ?STATE_DOWN :: match(?STATE_DOWN | ?STATE_UP), + op_state = {?STATE_DOWN, ?WD_INITIAL} + :: match(op_state() | {op_state(), wd_state()}), started = now(), %% at process start conn = false :: match(boolean() | pid())}). %% true at accept, pid() at connection_up (connT key) @@ -390,15 +408,6 @@ whois(SvcName) -> undefined end. -%%% --------------------------------------------------------------------------- -%%% # flush_stats/1 -%%% -%%% Output: list of {{SvcName, Alias, Counter}, Value} -%%% --------------------------------------------------------------------------- - -flush_stats(TPid) -> - diameter_stats:flush(TPid). - %% =========================================================================== %% =========================================================================== @@ -518,6 +527,34 @@ transition({reconnect, Pid}, S) -> reconnect(Pid, S), ok; +%% Watchdog is sending notification of a state transition. Note that +%% the connection_up/down messages are pre-date this message and are +%% still used. A 'watchdog' message will follow these and communicate +%% the same state as was set in handling connection_up/down. +transition({watchdog, Pid, {TPid, From, To}}, #state{service_name = SvcName, + peerT = PeerT}) -> + #peer{ref = Ref, type = T, options = Opts, op_state = {OS,_}} + = P + = fetch(PeerT, Pid), + insert(PeerT, P#peer{op_state = {OS, To}}), + send_event(SvcName, {watchdog, Ref, TPid, {From, To}, {T, Opts}}), + ok; +%% Death of a peer process results in the removal of it's peer and any +%% associated conn record when 'DOWN' is received (after this) but the +%% states will be {?STATE_UP, ?WD_DOWN} for a short time. (No real +%% problem since ?WD_* is only used in service_info.) We set ?WD_OKAY +%% as a consequence of connection_up since we know a watchdog is +%% coming. We can't set anything at connection_down since we don't +%% know if the subsequent watchdog message will be ?WD_DOWN or +%% ?WD_SUSPECT. We don't (yet) set ?STATE_* as a consequence of a +%% watchdog message since this requires changing some of the matching +%% on ?STATE_*. +%% +%% Death of a conn process results in connection_down followed by +%% watchdog ?WD_DOWN. The latter doesn't result in the conn record +%% being deleted since 'DOWN' from death of its peer doesn't (yet) +%% deal with the record having been removed. + %% Monitor process has died. Just die with a reason that tells %% diameter_config about the happening. If a cleaner shutdown is %% required then someone should stop us. @@ -881,7 +918,14 @@ accepted(Pid, _TPid, #state{peerT = PeerT} = S) -> fetch(Tid, Key) -> [T] = ets:lookup(Tid, Key), - T. + case T of + #peer{op_state = ?STATE_UP} = P -> + P#peer{op_state = {?STATE_UP, ?WD_OKAY}}; + #peer{op_state = ?STATE_DOWN} = P -> + P#peer{op_state = {?STATE_DOWN, ?WD_DOWN}}; + _ -> + T + end. %%% --------------------------------------------------------------------------- %%% # connection_up/3 @@ -927,12 +971,12 @@ connection_up(T, P, C, #state{peerT = PeerT, service = #diameter_service{applications = Apps}} = S) -> - #peer{conn = TPid, op_state = ?STATE_DOWN} + #peer{conn = TPid, op_state = {?STATE_DOWN, _}} = P, #conn{apps = SApps, caps = Caps} = C, - insert(PeerT, P#peer{op_state = ?STATE_UP}), + insert(PeerT, P#peer{op_state = {?STATE_UP, ?WD_OKAY}}), request_peer_up(TPid), report_status(up, P, C, S, T), @@ -947,27 +991,35 @@ ilp({Id, Alias}, {TC, SA}, LDict) -> init_conn(Id, Alias, TC, SA), ?Dict:append(Alias, TC, LDict). -init_conn(Id, Alias, TC, {SvcName, Apps}) -> +init_conn(Id, Alias, {TPid, _} = TC, {SvcName, Apps}) -> #diameter_app{module = ModX, id = Id} %% assert = find_app(Alias, Apps), - peer_cb({ModX, peer_up, [SvcName, TC]}, Alias). + peer_cb({ModX, peer_up, [SvcName, TC]}, Alias) + orelse exit(TPid, kill). %% fake transport failure + +%% find_app/2 find_app(Alias, Apps) -> - lists:keyfind(Alias, #diameter_app.alias, Apps). + case lists:keyfind(Alias, #diameter_app.alias, Apps) of + #diameter_app{options = E} = A when is_atom(E) -> %% upgrade + A#diameter_app{options = [{answer_errors, E}]}; + A -> + A + end. -%% A failing peer callback brings down the service. In the case of -%% peer_up we could just kill the transport and emit an error but for -%% peer_down we have no way to cleanup any state change that peer_up -%% may have introduced. +%% Don't bring down the service (and all associated connections) +%% regardless of what happens. peer_cb(MFA, Alias) -> try state_cb(MFA, Alias) of ModS -> - mod_state(Alias, ModS) + mod_state(Alias, ModS), + true catch - E: Reason -> - ?ERROR({E, Reason, MFA, ?STACK}) + E:R -> + diameter_lib:error_report({failure, {E, R, Alias, ?STACK}}, MFA), + false end. %%% --------------------------------------------------------------------------- @@ -981,22 +1033,22 @@ peer_cb(MFA, Alias) -> connection_down(Pid, #state{peerT = PeerT, connT = ConnT} = S) -> - #peer{op_state = ?STATE_UP, %% assert + #peer{op_state = {?STATE_UP, WS}, %% assert conn = TPid} = P = fetch(PeerT, Pid), C = fetch(ConnT, TPid), - insert(PeerT, P#peer{op_state = ?STATE_DOWN}), + insert(PeerT, P#peer{op_state = {?STATE_DOWN, WS}}), connection_down(P,C,S). %% connection_down/3 -connection_down(#peer{op_state = ?STATE_DOWN}, _, S) -> +connection_down(#peer{op_state = {?STATE_DOWN, _}}, _, S) -> S; connection_down(#peer{conn = TPid, - op_state = ?STATE_UP} + op_state = {?STATE_UP, _}} = P, #conn{caps = Caps, apps = SApps} @@ -1045,7 +1097,7 @@ peer_down(Pid, Reason, #state{peerT = PeerT} = S) -> %% Send an event at connection establishment failure. closed({shutdown, {close, _TPid, Reason}}, - #peer{op_state = ?STATE_DOWN, + #peer{op_state = {?STATE_DOWN, _}, ref = Ref, type = Type, options = Opts}, @@ -1354,7 +1406,7 @@ send_request(Pkt, TPid, Caps, App, Opts, Caller, SvcName) -> #diameter_app{alias = Alias, dictionary = Dict, module = ModX, - answer_errors = AE} + options = [{answer_errors, AE} | _]} = App, EPkt = encode(Dict, Pkt), @@ -2795,20 +2847,45 @@ transports(#state{peerT = PeerT}) -> 'Vendor-Specific-Application-Id', 'Firmware-Revision']). +%% The config returned by diameter:service_info(SvcName, all). -define(ALL_INFO, [capabilities, applications, transport, - pending, - statistics]). + pending]). + +%% The rest. +-define(OTHER_INFO, [connections, + name, + peers, + statistics]). -service_info(Items, S) - when is_list(Items) -> - [{complete(I), service_info(I,S)} || I <- Items]; service_info(Item, S) when is_atom(Item) -> - service_info(Item, S, true). + case tagged_info(Item, S) of + {_, T} -> T; + undefined = No -> No + end; + +service_info(Items, S) -> + tagged_info(Items, S). + +tagged_info(Item, S) + when is_atom(Item) -> + case complete(Item) of + {value, I} -> + {I, complete_info(I,S)}; + false -> + undefined + end; + +tagged_info(Items, S) + when is_list(Items) -> + [T || I <- Items, T <- [tagged_info(I,S)], T /= undefined, T /= []]; + +tagged_info(_, _) -> + undefined. -service_info(Item, #state{service = Svc} = S, Complete) -> +complete_info(Item, #state{service = Svc} = S) -> case Item of name -> S#state.service_name; @@ -2852,84 +2929,160 @@ service_info(Item, #state{service = Svc} = S, Complete) -> applications -> info_apps(S); transport -> info_transport(S); pending -> info_pending(S); - statistics -> info_stats(S); - keys -> ?ALL_INFO ++ ?CAP_INFO; %% mostly for test + keys -> ?ALL_INFO ++ ?CAP_INFO ++ ?OTHER_INFO; all -> service_info(?ALL_INFO, S); - _ when Complete -> service_info(complete(Item), S, false); - _ -> undefined + statistics -> info_stats(S); + connections -> info_connections(S); + peers -> info_peers(S) end. +complete(I) + when I == keys; + I == all -> + {value, I}; complete(Pre) -> P = atom_to_list(Pre), - case [I || I <- [name | ?ALL_INFO] ++ ?CAP_INFO, + case [I || I <- ?ALL_INFO ++ ?CAP_INFO ++ ?OTHER_INFO, lists:prefix(P, atom_to_list(I))] of - [I] -> I; - _ -> Pre + [I] -> {value, I}; + _ -> false end. +%% info_stats/1 + info_stats(#state{peerT = PeerT}) -> - Peers = ets:select(PeerT, [{#peer{ref = '$1', conn = '$2', _ = '_'}, - [{'is_pid', '$2'}], - [['$1', '$2']]}]), - diameter_stats:read(lists:append(Peers)). -%% TODO: include peer identities in return value - -info_transport(#state{peerT = PeerT, connT = ConnT}) -> - dict:fold(fun it/3, + MatchSpec = [{#peer{ref = '$1', conn = '$2', _ = '_'}, + [{'is_pid', '$2'}], + [['$1', '$2']]}], + diameter_stats:read(lists:append(ets:select(PeerT, MatchSpec))). + +%% info_transport/1 +%% +%% One entry per configured transport. Statistics for each entry are +%% the accumulated values for the ref and associated peer pids. + +info_transport(S) -> + PeerD = peer_dict(S, config_dict(S)), + RefsD = dict:map(fun(_, Ls) -> [P || L <- Ls, {peer, {P,_}} <- L] end, + PeerD), + Refs = lists:append(dict:fold(fun(R, Ps, A) -> [[R|Ps] | A] end, + [], + RefsD)), + Stats = diameter_stats:read(Refs), + dict:fold(fun(R, Ls, A) -> + Ps = dict:fetch(R, RefsD), + [[{ref, R} | transport(Ls)] ++ [stats([R|Ps], Stats)] + | A] + end, [], - ets:foldl(fun(T,A) -> it_acc(ConnT, A, T) end, - dict:new(), - PeerT)). - -it(Ref, [[{type, connect} | _] = L], Acc) -> - [[{ref, Ref} | L] | Acc]; -it(Ref, [[{type, accept}, {options, Opts} | _] | _] = L, Acc) -> - [[{ref, Ref}, - {type, listen}, - {options, Opts}, - {accept, [lists:nthtail(2,A) || A <- L]}] - | Acc]. -%% Each entry has the same Opts. (TODO) - -it_acc(ConnT, Acc, #peer{pid = Pid, - type = Type, - ref = Ref, - options = Opts, - op_state = OS, - started = T, - conn = TPid}) -> + PeerD). + +%% Only a config entry for a listening transport: use it. +transport([[{type, listen}, _] = L]) -> + L ++ [{accept, []}]; + +%% Only one config or peer entry for a connecting transport: use it. +transport([[{type, connect} | _] = L]) -> + L; + +%% Peer entries: discard config. Note that the peer entries have +%% length at least 3. +transport([[_,_] | L]) -> + transport(L); + +%% Possibly many peer entries for a listening transport. Note that all +%% have the same options by construction, which is not terribly space +%% efficient. (TODO: all entries for the same Ref should share options.) +transport([[{type, accept}, {options, Opts} | _] | _] = Ls) -> + [{type, listen}, + {options, Opts}, + {accept, [lists:nthtail(2,L) || L <- Ls]}]. + +peer_dict(#state{peerT = PeerT, connT = ConnT}, Dict0) -> + ets:foldl(fun(T,A) -> peer_acc(ConnT, A, T) end, Dict0, PeerT). + +peer_acc(ConnT, Acc, #peer{pid = Pid, + type = Type, + ref = Ref, + options = Opts, + op_state = OS, + started = T, + conn = TPid}) -> + WS = wd_state(OS), dict:append(Ref, [{type, Type}, {options, Opts}, - {watchdog, {Pid, T, OS}} - | info_conn(ConnT, TPid)], + {watchdog, {Pid, T, WS}} + | info_conn(ConnT, TPid, WS /= ?WD_DOWN)], Acc). -info_conn(ConnT, TPid) -> - info_conn(ets:lookup(ConnT, TPid)). +info_conn(ConnT, TPid, true) + when is_pid(TPid) -> + info_conn(ets:lookup(ConnT, TPid)); +info_conn(_, _, _) -> + []. + +%% The point of extracting the config here is so that 'transport' info +%% has one entry for each transport ref, the peer table only +%% containing entries that have a living watchdog. + +config_dict(#state{service_name = SvcName}) -> + lists:foldl(fun config_acc/2, + dict:new(), + diameter_config:lookup(SvcName)). + +config_acc({Ref, T, Opts}, Dict) + when T == listen; + T == connect -> + dict:store(Ref, [[{type, T}, {options, Opts}]], Dict); +config_acc(_, Dict) -> + Dict. + +wd_state({_,S}) -> + S; +wd_state(?STATE_UP) -> + ?WD_OKAY; +wd_state(?STATE_DOWN) -> + ?WD_DOWN. info_conn([#conn{pid = Pid, apps = SApps, caps = Caps, started = T}]) -> [{peer, {Pid, T}}, {apps, SApps}, - {caps, info_caps(Caps)}]; + {caps, info_caps(Caps)} + | try [{port, info_port(Pid)}] catch _:_ -> [] end]; info_conn([] = No) -> No. +%% Extract information that the processes involved are expected to +%% "publish" in their process dictionaries. Simple but backhanded. +info_port(Pid) -> + {_, PD} = process_info(Pid, dictionary), + {_, T} = lists:keyfind({diameter_peer_fsm, start}, 1, PD), + {TPid, {_Type, TMod, _Cfg}} = T, + {_, TD} = process_info(TPid, dictionary), + {_, Data} = lists:keyfind({TMod, info}, 1, TD), + [{owner, TPid}, + {module, TMod} + | try TMod:info(Data) catch _:_ -> [] end]. + +%% Use the fields names from diameter_caps instead of +%% diameter_base_CER to distinguish between the 2-tuple values +%% compared to the single capabilities values. Note also that the +%% returned list is tagged 'caps' rather than 'capabilities' to +%% emphasize the difference. info_caps(#diameter_caps{} = C) -> lists:zip(record_info(fields, diameter_caps), tl(tuple_to_list(C))). info_apps(#state{service = #diameter_service{applications = Apps}}) -> lists:map(fun mk_app/1, Apps). -mk_app(#diameter_app{alias = Alias, - dictionary = Dict, - module = ModX, - id = Id}) -> - [{alias, Alias}, - {dictionary, Dict}, - {module, ModX}, - {id, Id}]. +mk_app(#diameter_app{} = A) -> + lists:zip(record_info(fields, diameter_app), tl(tuple_to_list(A))). + +%% info_pending/1 +%% +%% One entry for each outgoing request whose answer is outstanding. info_pending(#state{} = S) -> MatchSpec = [{{'$1', @@ -2944,3 +3097,57 @@ info_pending(#state{} = S) -> {{from, '$3'}}]}}]}], ets:select(?REQUEST_TABLE, MatchSpec). + +%% info_connections/1 +%% +%% One entry per transport connection. Statistics for each entry are +%% for the peer pid only. + +info_connections(S) -> + ConnL = conn_list(S), + Stats = diameter_stats:read([P || L <- ConnL, {peer, {P,_}} <- L]), + [L ++ [stats([P], Stats)] || L <- ConnL, {peer, {P,_}} <- L]. + +conn_list(S) -> + lists:append(dict:fold(fun conn_acc/3, [], peer_dict(S, dict:new()))). + +conn_acc(Ref, Peers, Acc) -> + [[[{ref, Ref} | L] || L <- Peers, lists:keymember(peer, 1, L)] + | Acc]. + +stats(Refs, Stats) -> + {statistics, dict:to_list(lists:foldl(fun(R,D) -> + stats_acc(R, D, Stats) + end, + dict:new(), + Refs))}. + +stats_acc(Ref, Dict, Stats) -> + lists:foldl(fun({C,N}, D) -> dict:update_counter(C, N, D) end, + Dict, + proplists:get_value(Ref, Stats, [])). + +%% info_peers/1 +%% +%% One entry per peer Origin-Host. Statistics for each entry are +%% accumulated values for all peer pids. + +info_peers(S) -> + {PeerD, RefD} = lists:foldl(fun peer_acc/2, + {dict:new(), dict:new()}, + conn_list(S)), + Refs = lists:append(dict:fold(fun(_, Rs, A) -> [Rs|A] end, + [], + RefD)), + Stats = diameter_stats:read(Refs), + dict:fold(fun(OH, Cs, A) -> + Rs = dict:fetch(OH, RefD), + [{OH, [{connections, Cs}, stats(Rs, Stats)]} | A] + end, + [], + PeerD). + +peer_acc(Peer, {PeerD, RefD}) -> + [{TPid, _}, [{origin_host, {_, OH}} | _]] + = [proplists:get_value(K, Peer) || K <- [peer, caps]], + {dict:append(OH, Peer, PeerD), dict:append(OH, TPid, RefD)}. |