riak_test/tests/repl_cancel_fullsync.erl
Engel A. Sanchez 556cb7210c Ensure riak_repl service is up across the board
Trying to use the repl features before newly started nodes have
riak_repl completely initialized leads to all sorts of nasty crashes and
noise. Frequently it makes fullsync stuck forever, which makes a lot of
the tests fail.

This also tweaks the AAE fullsync tests to remove assumptions about
failure stats when AAE transient errors occur. The behavior in the
handling of those errors has changed recently with the introduction of
soft exits.
2014-12-18 16:07:00 -05:00

148 lines
4.9 KiB
Erlang

-module(repl_cancel_fullsync).
-behavior(riak_test).
-export([confirm/0]).
-include_lib("eunit/include/eunit.hrl").
-define(TEST_BUCKET,
<<"repl-cancel-fullsync-failures-systest_a">>).
-define(NUM_KEYS, 1000).
-define(CONF(Retries), [
{riak_core,
[
{ring_creation_size, 8},
{default_bucket_props, [{n_val, 1}]}
]
},
{riak_kv,
[
%% Specify fast building of AAE trees
{anti_entropy, {on, []}},
{anti_entropy_build_limit, {100, 1000}},
{anti_entropy_concurrency, 100}
]
},
{riak_repl,
[
{fullsync_strategy, keylist},
{fullsync_on_connect, false},
{fullsync_interval, disabled},
{max_fssource_retries, Retries}
]}
]).
%% @doc Ensure we can cancel a fullsync and restart it.
confirm() ->
rt:set_advanced_conf(all, ?CONF(5)),
Nodes = [ANodes, BNodes] = rt:build_clusters([3, 3]),
rt:wait_for_cluster_service(ANodes, riak_repl),
rt:wait_for_cluster_service(BNodes, riak_repl),
lager:info("ANodes: ~p", [ANodes]),
lager:info("BNodes: ~p", [BNodes]),
AFirst = hd(ANodes),
BFirst = hd(BNodes),
lager:info("Naming clusters."),
repl_util:name_cluster(AFirst, "A"),
repl_util:name_cluster(BFirst, "B"),
lager:info("Waiting for convergence."),
rt:wait_until_ring_converged(ANodes),
rt:wait_until_ring_converged(BNodes),
lager:info("Waiting for transfers to complete."),
rt:wait_until_transfers_complete(ANodes),
rt:wait_until_transfers_complete(BNodes),
lager:info("Get leaders."),
LeaderA = repl_util:get_leader(AFirst),
LeaderB = repl_util:get_leader(BFirst),
lager:info("Finding connection manager ports."),
BPort = repl_util:get_port(LeaderB),
lager:info("Connecting cluster A to B"),
repl_util:connect_cluster_by_name(LeaderA, BPort, "B"),
repl_util:write_to_cluster(AFirst, 1, ?NUM_KEYS, ?TEST_BUCKET),
repl_util:read_from_cluster(BFirst, 1, ?NUM_KEYS, ?TEST_BUCKET,
?NUM_KEYS),
lager:info("Test fullsync from cluster A leader ~p to cluster B",
[LeaderA]),
repl_util:enable_fullsync(LeaderA, "B"),
rt:wait_until_ring_converged(ANodes),
lager:info("Starting fullsync."),
rt:log_to_nodes(Nodes, "Starting fullsync."),
R1 = rpc:call(LeaderA, riak_repl_console, fullsync, [["start"]]),
?assertEqual(ok, R1),
repl_util:wait_until_fullsync_started(LeaderA),
lager:info("Fullsync running."),
%% Get all active keylist server pids
Coordinators = [Pid || {"B", Pid} <-
riak_repl2_fscoordinator_sup:started(LeaderA)],
States = [sys:get_state(P) || P <- Coordinators],
KeylistPids = lists:flatten([element(14, State) || State <- States]),
KLStates = [sys:get_state(Pid) || {Pid, _} <- KeylistPids],
[?assertEqual(state, element(1, State)) || State <- KLStates],
lager:info("Stopping fullsync."),
rt:log_to_nodes(Nodes, "Stopping fullsync."),
R2 = rpc:call(LeaderA, riak_repl_console, fullsync, [["stop"]]),
?assertEqual(ok, R2),
repl_util:wait_until_fullsync_stopped(LeaderA),
lager:info("Fullsync stopped."),
%% Give keylist pids time to stop
timer:sleep(500),
%% Ensure keylist pids are actually gone
Exits = [catch sys:get_state(Pid) || {Pid, _} <- KeylistPids],
[?assertMatch({'EXIT', _}, Exit) || Exit <- Exits],
[{"B", S1}] = rpc:call(LeaderA, riak_repl2_fscoordinator, status, []),
?assertEqual(true, lists:member({fullsyncs_completed, 0}, S1)),
lager:info("Fullsync not completed."),
[{"B", S2}] = rpc:call(LeaderA, riak_repl2_fscoordinator, status, []),
?assertEqual(true, lists:member({in_progress, 0}, S2)),
lager:info("** ~p", [S2]),
lager:info("Starting fullsync."),
rt:log_to_nodes(Nodes, "Starting fullsync."),
R3 = rpc:call(LeaderA, riak_repl_console, fullsync, [["start"]]),
?assertEqual(ok, R3),
repl_util:wait_until_fullsync_started(LeaderA),
lager:info("Fullsync running again."),
Res = rt:wait_until(LeaderA,
fun(_) ->
Status = rpc:call(LeaderA,
riak_repl_console,
status,
[quiet]),
case proplists:get_value(server_fullsyncs, Status) of
1 ->
true;
_ ->
false
end
end),
?assertEqual(ok, Res),
repl_util:read_from_cluster(BFirst, 1, ?NUM_KEYS, ?TEST_BUCKET, 0),
[{"B", S3}] = rpc:call(LeaderA, riak_repl2_fscoordinator, status, []),
?assertEqual(true, lists:member({fullsyncs_completed, 1}, S3)),
lager:info("Fullsync Complete"),
rt:log_to_nodes(Nodes, "Test completed."),
rt:clean_cluster(ANodes),
rt:clean_cluster(BNodes),
pass.