* [PATCH v4 2/3] Add arguments to replication test instances.
2018-08-14 10:02 [PATCH v4 0/3] replication: do not ignore replication_connect_quorum Serge Petrenko
2018-08-14 10:02 ` [PATCH v4 1/3] test: update test-run Serge Petrenko
@ 2018-08-14 10:02 ` Serge Petrenko
2018-08-14 10:02 ` [PATCH v4 3/3] replication: do not ignore replication_connect_quorum Serge Petrenko
2018-08-14 17:07 ` [PATCH v4 0/3] " Vladimir Davydov
3 siblings, 0 replies; 5+ messages in thread
From: Serge Petrenko @ 2018-08-14 10:02 UTC (permalink / raw)
To: tarantool-patches; +Cc: georgy, vdavydov.dev, Serge Petrenko
Add start arguments to replication test instances to control
replication_timeout and replication_connect_timeout settings between
restarts.
Prerequisite #3428
---
test/replication-py/init_storage.test.py | 2 +-
test/replication-py/master.lua | 1 +
test/replication-py/replica.lua | 1 +
test/replication/autobootstrap.lua | 6 +++++-
test/replication/autobootstrap.result | 4 ++--
test/replication/autobootstrap.test.lua | 4 ++--
test/replication/autobootstrap_guest.lua | 7 ++++++-
test/replication/autobootstrap_guest.result | 2 +-
test/replication/autobootstrap_guest.test.lua | 2 +-
test/replication/before_replace.result | 8 ++++----
test/replication/before_replace.test.lua | 8 ++++----
test/replication/catch.result | 2 +-
test/replication/catch.test.lua | 2 +-
test/replication/ddl.lua | 7 ++++++-
test/replication/ddl.result | 2 +-
test/replication/ddl.test.lua | 2 +-
test/replication/errinj.result | 6 +++---
test/replication/errinj.test.lua | 6 +++---
test/replication/master.lua | 1 +
test/replication/master_quorum.lua | 7 ++++++-
test/replication/misc.result | 2 +-
test/replication/misc.test.lua | 2 +-
test/replication/on_replace.lua | 14 +++++++++-----
test/replication/on_replace.result | 2 +-
test/replication/on_replace.test.lua | 2 +-
test/replication/quorum.lua | 8 ++++++--
test/replication/quorum.result | 16 ++++++++--------
test/replication/quorum.test.lua | 16 ++++++++--------
test/replication/rebootstrap.lua | 8 ++++++--
test/replication/rebootstrap.result | 6 +++---
test/replication/rebootstrap.test.lua | 6 +++---
test/replication/recover_missing_xlog.result | 4 ++--
test/replication/recover_missing_xlog.test.lua | 4 ++--
test/replication/replica_no_quorum.lua | 3 ++-
test/replication/replica_timeout.lua | 3 ++-
test/replication/replica_uuid_ro.lua | 7 ++++++-
test/replication/replicaset_ro_mostly.result | 8 +++++---
test/replication/replicaset_ro_mostly.test.lua | 8 +++++---
38 files changed, 122 insertions(+), 77 deletions(-)
diff --git a/test/replication-py/init_storage.test.py b/test/replication-py/init_storage.test.py
index 0911a02c0..32b4639f1 100644
--- a/test/replication-py/init_storage.test.py
+++ b/test/replication-py/init_storage.test.py
@@ -57,7 +57,7 @@ print '-------------------------------------------------------------'
server.stop()
replica = TarantoolServer(server.ini)
-replica.script = 'replication/replica.lua'
+replica.script = 'replication-py/replica.lua'
replica.vardir = server.vardir #os.path.join(server.vardir, 'replica')
replica.rpl_master = master
replica.deploy(wait=False)
diff --git a/test/replication-py/master.lua b/test/replication-py/master.lua
index 0f9f7a6f0..e924b5495 100644
--- a/test/replication-py/master.lua
+++ b/test/replication-py/master.lua
@@ -3,6 +3,7 @@ os = require('os')
box.cfg({
listen = os.getenv("LISTEN"),
memtx_memory = 107374182,
+ replication_timeout = 0.1
})
require('console').listen(os.getenv('ADMIN'))
diff --git a/test/replication-py/replica.lua b/test/replication-py/replica.lua
index 278291bba..32d888eff 100644
--- a/test/replication-py/replica.lua
+++ b/test/replication-py/replica.lua
@@ -7,6 +7,7 @@ box.cfg({
listen = os.getenv("LISTEN"),
replication = os.getenv("MASTER"),
memtx_memory = 107374182,
+ replication_timeout = 0.1
})
box_cfg_done = true
diff --git a/test/replication/autobootstrap.lua b/test/replication/autobootstrap.lua
index 4f55417ae..856b36e66 100644
--- a/test/replication/autobootstrap.lua
+++ b/test/replication/autobootstrap.lua
@@ -5,6 +5,9 @@ local INSTANCE_ID = string.match(arg[0], "%d")
local USER = 'cluster'
local PASSWORD = 'somepassword'
local SOCKET_DIR = require('fio').cwd()
+local TIMEOUT = tonumber(arg[1])
+local CON_TIMEOUT = arg[2] and tonumber(arg[2]) or 30.0
+
local function instance_uri(instance_id)
--return 'localhost:'..(3310 + instance_id)
return SOCKET_DIR..'/autobootstrap'..instance_id..'.sock';
@@ -21,7 +24,8 @@ box.cfg({
USER..':'..PASSWORD..'@'..instance_uri(2);
USER..':'..PASSWORD..'@'..instance_uri(3);
};
- replication_connect_timeout = 0.5,
+ replication_timeout = TIMEOUT;
+ replication_connect_timeout = CON_TIMEOUT;
})
box.once("bootstrap", function()
diff --git a/test/replication/autobootstrap.result b/test/replication/autobootstrap.result
index 04aeb4315..91badc1f1 100644
--- a/test/replication/autobootstrap.result
+++ b/test/replication/autobootstrap.result
@@ -13,7 +13,7 @@ SERVERS = { 'autobootstrap1', 'autobootstrap2', 'autobootstrap3' }
--
-- Start servers
--
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
---
...
--
@@ -161,7 +161,7 @@ box.space.test_u:select()
_ = test_run:cmd("switch autobootstrap1")
---
...
-_ = test_run:cmd("restart server autobootstrap1 with cleanup=1")
+_ = test_run:cmd("restart server autobootstrap1 with cleanup=1, args ='0.1 0.5'")
_ = box.space.test_u:replace({5, 6, 7, 8})
---
...
diff --git a/test/replication/autobootstrap.test.lua b/test/replication/autobootstrap.test.lua
index f1e2a9991..752d5f317 100644
--- a/test/replication/autobootstrap.test.lua
+++ b/test/replication/autobootstrap.test.lua
@@ -8,7 +8,7 @@ SERVERS = { 'autobootstrap1', 'autobootstrap2', 'autobootstrap3' }
--
-- Start servers
--
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
--
-- Wait for full mesh
@@ -76,7 +76,7 @@ box.space.test_u:select()
-- Rebootstrap one node and check that others follow.
--
_ = test_run:cmd("switch autobootstrap1")
-_ = test_run:cmd("restart server autobootstrap1 with cleanup=1")
+_ = test_run:cmd("restart server autobootstrap1 with cleanup=1, args ='0.1 0.5'")
_ = box.space.test_u:replace({5, 6, 7, 8})
box.space.test_u:select()
diff --git a/test/replication/autobootstrap_guest.lua b/test/replication/autobootstrap_guest.lua
index 40fef2c7a..d7176ae5b 100644
--- a/test/replication/autobootstrap_guest.lua
+++ b/test/replication/autobootstrap_guest.lua
@@ -4,6 +4,10 @@
local INSTANCE_ID = string.match(arg[0], "%d")
local SOCKET_DIR = require('fio').cwd()
+
+local TIMEOUT = tonumber(arg[1])
+local CON_TIMEOUT = arg[2] and tonumber(arg[2]) or 30.0
+
local function instance_uri(instance_id)
--return 'localhost:'..(3310 + instance_id)
return SOCKET_DIR..'/autobootstrap_guest'..instance_id..'.sock';
@@ -20,7 +24,8 @@ box.cfg({
instance_uri(2);
instance_uri(3);
};
- replication_connect_timeout = 0.5,
+ replication_timeout = TIMEOUT;
+ replication_connect_timeout = CON_TIMEOUT;
})
box.once("bootstrap", function()
diff --git a/test/replication/autobootstrap_guest.result b/test/replication/autobootstrap_guest.result
index 49f9bee01..1efef310c 100644
--- a/test/replication/autobootstrap_guest.result
+++ b/test/replication/autobootstrap_guest.result
@@ -13,7 +13,7 @@ SERVERS = { 'autobootstrap_guest1', 'autobootstrap_guest2', 'autobootstrap_guest
--
-- Start servers
--
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
---
...
--
diff --git a/test/replication/autobootstrap_guest.test.lua b/test/replication/autobootstrap_guest.test.lua
index 70250cff9..3aad8a4da 100644
--- a/test/replication/autobootstrap_guest.test.lua
+++ b/test/replication/autobootstrap_guest.test.lua
@@ -7,7 +7,7 @@ SERVERS = { 'autobootstrap_guest1', 'autobootstrap_guest2', 'autobootstrap_guest
--
-- Start servers
--
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
--
-- Wait for full mesh
diff --git a/test/replication/before_replace.result b/test/replication/before_replace.result
index d561b4813..00f9bcb8b 100644
--- a/test/replication/before_replace.result
+++ b/test/replication/before_replace.result
@@ -11,7 +11,7 @@ SERVERS = { 'autobootstrap1', 'autobootstrap2', 'autobootstrap3' }
---
...
-- Deploy a cluster.
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
---
...
test_run:wait_fullmesh(SERVERS)
@@ -125,7 +125,7 @@ box.space.test:select()
- [9, 90]
- [10, 100]
...
-test_run:cmd('restart server autobootstrap1')
+test_run:cmd('restart server autobootstrap1 with args="0.1 0.5"')
box.space.test:select()
---
- - [1, 10]
@@ -156,7 +156,7 @@ box.space.test:select()
- [9, 90]
- [10, 100]
...
-test_run:cmd('restart server autobootstrap2')
+test_run:cmd('restart server autobootstrap2 with args="0.1 0.5"')
box.space.test:select()
---
- - [1, 10]
@@ -187,7 +187,7 @@ box.space.test:select()
- [9, 90]
- [10, 100]
...
-test_run:cmd('restart server autobootstrap3')
+test_run:cmd('restart server autobootstrap3 with args="0.1 0.5"')
box.space.test:select()
---
- - [1, 10]
diff --git a/test/replication/before_replace.test.lua b/test/replication/before_replace.test.lua
index 2c6912d06..b49d438b4 100644
--- a/test/replication/before_replace.test.lua
+++ b/test/replication/before_replace.test.lua
@@ -7,7 +7,7 @@ test_run = env.new()
SERVERS = { 'autobootstrap1', 'autobootstrap2', 'autobootstrap3' }
-- Deploy a cluster.
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
test_run:wait_fullmesh(SERVERS)
-- Setup space:before_replace trigger on all replicas.
@@ -54,15 +54,15 @@ vclock2 = test_run:wait_cluster_vclock(SERVERS, vclock)
-- and the state persists after restart.
test_run:cmd("switch autobootstrap1")
box.space.test:select()
-test_run:cmd('restart server autobootstrap1')
+test_run:cmd('restart server autobootstrap1 with args="0.1 0.5"')
box.space.test:select()
test_run:cmd("switch autobootstrap2")
box.space.test:select()
-test_run:cmd('restart server autobootstrap2')
+test_run:cmd('restart server autobootstrap2 with args="0.1 0.5"')
box.space.test:select()
test_run:cmd("switch autobootstrap3")
box.space.test:select()
-test_run:cmd('restart server autobootstrap3')
+test_run:cmd('restart server autobootstrap3 with args="0.1 0.5"')
box.space.test:select()
diff --git a/test/replication/catch.result b/test/replication/catch.result
index 91be32725..0f72e89e2 100644
--- a/test/replication/catch.result
+++ b/test/replication/catch.result
@@ -23,7 +23,7 @@ test_run:cmd("create server replica with rpl_master=default, script='replication
---
- true
...
-test_run:cmd("start server replica with args='1'")
+test_run:cmd("start server replica with args='0.1'")
---
- true
...
diff --git a/test/replication/catch.test.lua b/test/replication/catch.test.lua
index 2e2e97bc4..457f910e9 100644
--- a/test/replication/catch.test.lua
+++ b/test/replication/catch.test.lua
@@ -9,7 +9,7 @@ errinj = box.error.injection
box.schema.user.grant('guest', 'replication')
test_run:cmd("create server replica with rpl_master=default, script='replication/replica_timeout.lua'")
-test_run:cmd("start server replica with args='1'")
+test_run:cmd("start server replica with args='0.1'")
test_run:cmd("switch replica")
test_run:cmd("switch default")
diff --git a/test/replication/ddl.lua b/test/replication/ddl.lua
index 694f40eac..72cf1db69 100644
--- a/test/replication/ddl.lua
+++ b/test/replication/ddl.lua
@@ -5,6 +5,10 @@ local INSTANCE_ID = string.match(arg[0], "%d")
local USER = 'cluster'
local PASSWORD = 'somepassword'
local SOCKET_DIR = require('fio').cwd()
+
+local TIMEOUT = tonumber(arg[1])
+local CON_TIMEOUT = arg[2] and tonumber(arg[2]) or 30.0
+
local function instance_uri(instance_id)
--return 'localhost:'..(3310 + instance_id)
return SOCKET_DIR..'/autobootstrap'..instance_id..'.sock';
@@ -22,7 +26,8 @@ box.cfg({
USER..':'..PASSWORD..'@'..instance_uri(3);
USER..':'..PASSWORD..'@'..instance_uri(4);
};
- replication_connect_timeout = 0.5,
+ replication_timeout = TIMEOUT,
+ replication_connect_timeout = CON_TIMEOUT,
})
box.once("bootstrap", function()
diff --git a/test/replication/ddl.result b/test/replication/ddl.result
index cc61fd4ce..8cd54cdfb 100644
--- a/test/replication/ddl.result
+++ b/test/replication/ddl.result
@@ -5,7 +5,7 @@ SERVERS = { 'ddl1', 'ddl2', 'ddl3', 'ddl4' }
---
...
-- Deploy a cluster.
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
---
...
test_run:wait_fullmesh(SERVERS)
diff --git a/test/replication/ddl.test.lua b/test/replication/ddl.test.lua
index 4cf4ffa38..f56071adc 100644
--- a/test/replication/ddl.test.lua
+++ b/test/replication/ddl.test.lua
@@ -3,7 +3,7 @@ test_run = require('test_run').new()
SERVERS = { 'ddl1', 'ddl2', 'ddl3', 'ddl4' }
-- Deploy a cluster.
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
test_run:wait_fullmesh(SERVERS)
test_run:cmd("switch ddl1")
test_run = require('test_run').new()
diff --git a/test/replication/errinj.result b/test/replication/errinj.result
index ca8af2988..3fc432010 100644
--- a/test/replication/errinj.result
+++ b/test/replication/errinj.result
@@ -418,7 +418,7 @@ test_run:cmd("create server replica_timeout with rpl_master=default, script='rep
---
- true
...
-test_run:cmd("start server replica_timeout with args='0.01'")
+test_run:cmd("start server replica_timeout with args='0.01 0.5'")
---
- true
...
@@ -474,7 +474,7 @@ errinj.set("ERRINJ_RELAY_REPORT_INTERVAL", 0)
...
-- Check replica's ACKs don't prevent the master from sending
-- heartbeat messages (gh-3160).
-test_run:cmd("start server replica_timeout with args='0.009'")
+test_run:cmd("start server replica_timeout with args='0.009 0.5'")
---
- true
...
@@ -522,7 +522,7 @@ for i = 0, 9999 do box.space.test:replace({i, 4, 5, 'test'}) end
-- during the join stage, i.e. a replica with a minuscule
-- timeout successfully bootstraps and breaks connection only
-- after subscribe.
-test_run:cmd("start server replica_timeout with args='0.00001'")
+test_run:cmd("start server replica_timeout with args='0.00001 0.5'")
---
- true
...
diff --git a/test/replication/errinj.test.lua b/test/replication/errinj.test.lua
index 463d89a8f..37375f45e 100644
--- a/test/replication/errinj.test.lua
+++ b/test/replication/errinj.test.lua
@@ -173,7 +173,7 @@ errinj.set("ERRINJ_RELAY_EXIT_DELAY", 0)
box.cfg{replication_timeout = 0.01}
test_run:cmd("create server replica_timeout with rpl_master=default, script='replication/replica_timeout.lua'")
-test_run:cmd("start server replica_timeout with args='0.01'")
+test_run:cmd("start server replica_timeout with args='0.01 0.5'")
test_run:cmd("switch replica_timeout")
fiber = require('fiber')
@@ -199,7 +199,7 @@ errinj.set("ERRINJ_RELAY_REPORT_INTERVAL", 0)
-- Check replica's ACKs don't prevent the master from sending
-- heartbeat messages (gh-3160).
-test_run:cmd("start server replica_timeout with args='0.009'")
+test_run:cmd("start server replica_timeout with args='0.009 0.5'")
test_run:cmd("switch replica_timeout")
fiber = require('fiber')
@@ -219,7 +219,7 @@ for i = 0, 9999 do box.space.test:replace({i, 4, 5, 'test'}) end
-- during the join stage, i.e. a replica with a minuscule
-- timeout successfully bootstraps and breaks connection only
-- after subscribe.
-test_run:cmd("start server replica_timeout with args='0.00001'")
+test_run:cmd("start server replica_timeout with args='0.00001 0.5'")
test_run:cmd("switch replica_timeout")
fiber = require('fiber')
while box.info.replication[1].upstream.message ~= 'timed out' do fiber.sleep(0.0001) end
diff --git a/test/replication/master.lua b/test/replication/master.lua
index 6d431aaeb..9b96b7891 100644
--- a/test/replication/master.lua
+++ b/test/replication/master.lua
@@ -4,6 +4,7 @@ box.cfg({
listen = os.getenv("LISTEN"),
memtx_memory = 107374182,
replication_connect_timeout = 0.5,
+ replication_timeout = 0.1
})
require('console').listen(os.getenv('ADMIN'))
diff --git a/test/replication/master_quorum.lua b/test/replication/master_quorum.lua
index fb5f7ec2b..05272ac5e 100644
--- a/test/replication/master_quorum.lua
+++ b/test/replication/master_quorum.lua
@@ -4,6 +4,10 @@
local INSTANCE_ID = string.match(arg[0], "%d")
local SOCKET_DIR = require('fio').cwd()
+
+local TIMEOUT = tonumber(arg[1])
+local CON_TIMEOUT = arg[2] and tonumber(arg[2]) or 30.0
+
local function instance_uri(instance_id)
--return 'localhost:'..(3310 + instance_id)
return SOCKET_DIR..'/master_quorum'..instance_id..'.sock';
@@ -20,7 +24,8 @@ box.cfg({
instance_uri(2);
};
replication_connect_quorum = 0;
- replication_connect_timeout = 0.1;
+ replication_timeout = TIMEOUT;
+ replication_connect_timeout = CON_TIMEOUT;
})
test_run = require('test_run').new()
diff --git a/test/replication/misc.result b/test/replication/misc.result
index 9d9d010c0..9df2a2c4b 100644
--- a/test/replication/misc.result
+++ b/test/replication/misc.result
@@ -93,7 +93,7 @@ SERVERS = { 'autobootstrap1', 'autobootstrap2', 'autobootstrap3' }
---
...
-- Deploy a cluster.
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
---
...
test_run:wait_fullmesh(SERVERS)
diff --git a/test/replication/misc.test.lua b/test/replication/misc.test.lua
index da5a90239..979c5d58c 100644
--- a/test/replication/misc.test.lua
+++ b/test/replication/misc.test.lua
@@ -37,7 +37,7 @@ box.cfg{read_only = false}
SERVERS = { 'autobootstrap1', 'autobootstrap2', 'autobootstrap3' }
-- Deploy a cluster.
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
test_run:wait_fullmesh(SERVERS)
test_run:cmd("switch autobootstrap1")
test_run = require('test_run').new()
diff --git a/test/replication/on_replace.lua b/test/replication/on_replace.lua
index 03f15d94c..40c12a9ea 100644
--- a/test/replication/on_replace.lua
+++ b/test/replication/on_replace.lua
@@ -5,6 +5,10 @@ local INSTANCE_ID = string.match(arg[0], "%d")
local USER = 'cluster'
local PASSWORD = 'somepassword'
local SOCKET_DIR = require('fio').cwd()
+
+local TIMEOUT = tonumber(arg[1])
+local CON_TIMEOUT = arg[2] and tonumber(arg[2]) or 30.0
+
local function instance_uri(instance_id)
--return 'localhost:'..(3310 + instance_id)
return SOCKET_DIR..'/on_replace'..instance_id..'.sock';
@@ -12,6 +16,9 @@ end
-- start console first
require('console').listen(os.getenv('ADMIN'))
+env = require('test_run')
+test_run = env.new()
+engine = test_run:get_cfg('engine')
box.cfg({
listen = instance_uri(INSTANCE_ID);
@@ -20,13 +27,10 @@ box.cfg({
USER..':'..PASSWORD..'@'..instance_uri(1);
USER..':'..PASSWORD..'@'..instance_uri(2);
};
- replication_connect_timeout = 0.5,
+ replication_timeout = TIMEOUT,
+ replication_connect_timeout = CON_TIMEOUT,
})
-env = require('test_run')
-test_run = env.new()
-engine = test_run:get_cfg('engine')
-
box.once("bootstrap", function()
box.schema.user.create(USER, { password = PASSWORD })
box.schema.user.grant(USER, 'replication')
diff --git a/test/replication/on_replace.result b/test/replication/on_replace.result
index 1736c53b7..4ffa3b25a 100644
--- a/test/replication/on_replace.result
+++ b/test/replication/on_replace.result
@@ -98,7 +98,7 @@ box.schema.user.revoke('guest', 'replication')
SERVERS = { 'on_replace1', 'on_replace2' }
---
...
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.2"})
---
...
test_run:wait_fullmesh(SERVERS)
diff --git a/test/replication/on_replace.test.lua b/test/replication/on_replace.test.lua
index 10bba2ddf..371b71cbd 100644
--- a/test/replication/on_replace.test.lua
+++ b/test/replication/on_replace.test.lua
@@ -44,7 +44,7 @@ box.schema.user.revoke('guest', 'replication')
-- gh-2682 on_replace on slave server with data change
SERVERS = { 'on_replace1', 'on_replace2' }
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.2"})
test_run:wait_fullmesh(SERVERS)
test_run:cmd('switch on_replace1')
diff --git a/test/replication/quorum.lua b/test/replication/quorum.lua
index 9c7bf5c93..f61c8748f 100644
--- a/test/replication/quorum.lua
+++ b/test/replication/quorum.lua
@@ -4,6 +4,10 @@
local INSTANCE_ID = string.match(arg[0], "%d")
local SOCKET_DIR = require('fio').cwd()
+
+local TIMEOUT = tonumber(arg[1])
+local CON_TIMEOUT = arg[2] and tonumber(arg[2]) or 30.0
+
local function instance_uri(instance_id)
--return 'localhost:'..(3310 + instance_id)
return SOCKET_DIR..'/quorum'..instance_id..'.sock';
@@ -14,9 +18,9 @@ require('console').listen(os.getenv('ADMIN'))
box.cfg({
listen = instance_uri(INSTANCE_ID);
- replication_timeout = 0.05;
+ replication_timeout = TIMEOUT;
replication_sync_lag = 0.01;
- replication_connect_timeout = 0.1;
+ replication_connect_timeout = CON_TIMEOUT;
replication_connect_quorum = 3;
replication = {
instance_uri(1);
diff --git a/test/replication/quorum.result b/test/replication/quorum.result
index 8f6e7a070..a55b0d087 100644
--- a/test/replication/quorum.result
+++ b/test/replication/quorum.result
@@ -5,7 +5,7 @@ SERVERS = {'quorum1', 'quorum2', 'quorum3'}
---
...
-- Deploy a cluster.
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
---
...
test_run:wait_fullmesh(SERVERS)
@@ -27,7 +27,7 @@ test_run:cmd('switch quorum2')
---
- true
...
-test_run:cmd('restart server quorum2')
+test_run:cmd('restart server quorum2 with args="0.1 0.5"')
box.info.status -- orphan
---
- orphan
@@ -51,7 +51,7 @@ box.info.status -- running
---
- running
...
-test_run:cmd('restart server quorum2')
+test_run:cmd('restart server quorum2 with args="0.1 0.5"')
box.info.status -- orphan
---
- orphan
@@ -82,7 +82,7 @@ box.info.status -- running
---
- running
...
-test_run:cmd('restart server quorum2')
+test_run:cmd('restart server quorum2 with args="0.1 0.5"')
box.info.status -- orphan
---
- orphan
@@ -99,7 +99,7 @@ box.space.test:replace{100} -- error
---
- error: Can't modify data because this instance is in read-only mode.
...
-test_run:cmd('start server quorum1')
+test_run:cmd('start server quorum1 with args="0.1 0.5"')
---
- true
...
@@ -158,7 +158,7 @@ fiber = require('fiber')
fiber.sleep(0.1)
---
...
-test_run:cmd('start server quorum1')
+test_run:cmd('start server quorum1 with args="0.1 0.5"')
---
- true
...
@@ -196,7 +196,7 @@ test_run:cmd('switch quorum1')
---
- true
...
-test_run:cmd('restart server quorum1 with cleanup=1')
+test_run:cmd('restart server quorum1 with cleanup=1, args="0.1 0.5"')
box.space.test:count() -- 100
---
- 100
@@ -356,7 +356,7 @@ SERVERS = {'master_quorum1', 'master_quorum2'}
---
...
-- Deploy a cluster.
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
---
...
test_run:wait_fullmesh(SERVERS)
diff --git a/test/replication/quorum.test.lua b/test/replication/quorum.test.lua
index 1df0ae1e7..8290f8ea5 100644
--- a/test/replication/quorum.test.lua
+++ b/test/replication/quorum.test.lua
@@ -3,7 +3,7 @@ test_run = require('test_run').new()
SERVERS = {'quorum1', 'quorum2', 'quorum3'}
-- Deploy a cluster.
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
test_run:wait_fullmesh(SERVERS)
-- Stop one replica and try to restart another one.
@@ -18,7 +18,7 @@ test_run:cmd('stop server quorum1')
test_run:cmd('switch quorum2')
-test_run:cmd('restart server quorum2')
+test_run:cmd('restart server quorum2 with args="0.1 0.5"')
box.info.status -- orphan
box.ctl.wait_rw(0.001) -- timeout
box.info.ro -- true
@@ -26,7 +26,7 @@ box.space.test:replace{100} -- error
box.cfg{replication={}}
box.info.status -- running
-test_run:cmd('restart server quorum2')
+test_run:cmd('restart server quorum2 with args="0.1 0.5"')
box.info.status -- orphan
box.ctl.wait_rw(0.001) -- timeout
box.info.ro -- true
@@ -36,12 +36,12 @@ box.ctl.wait_rw()
box.info.ro -- false
box.info.status -- running
-test_run:cmd('restart server quorum2')
+test_run:cmd('restart server quorum2 with args="0.1 0.5"')
box.info.status -- orphan
box.ctl.wait_rw(0.001) -- timeout
box.info.ro -- true
box.space.test:replace{100} -- error
-test_run:cmd('start server quorum1')
+test_run:cmd('start server quorum1 with args="0.1 0.5"')
box.ctl.wait_rw()
box.info.ro -- false
box.info.status -- running
@@ -63,7 +63,7 @@ for i = 1, 100 do box.space.test:insert{i} end
fiber = require('fiber')
fiber.sleep(0.1)
-test_run:cmd('start server quorum1')
+test_run:cmd('start server quorum1 with args="0.1 0.5"')
test_run:cmd('switch quorum1')
box.space.test:count() -- 100
@@ -79,7 +79,7 @@ test_run:cmd('switch quorum2')
box.snapshot()
test_run:cmd('switch quorum1')
-test_run:cmd('restart server quorum1 with cleanup=1')
+test_run:cmd('restart server quorum1 with cleanup=1, args="0.1 0.5"')
box.space.test:count() -- 100
@@ -136,7 +136,7 @@ box.schema.user.revoke('guest', 'replication')
-- Second case, check that master-master works.
SERVERS = {'master_quorum1', 'master_quorum2'}
-- Deploy a cluster.
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
test_run:wait_fullmesh(SERVERS)
test_run:cmd("switch master_quorum1")
repl = box.cfg.replication
diff --git a/test/replication/rebootstrap.lua b/test/replication/rebootstrap.lua
index e743577e4..3e7d8f062 100644
--- a/test/replication/rebootstrap.lua
+++ b/test/replication/rebootstrap.lua
@@ -4,6 +4,10 @@
local INSTANCE_ID = string.match(arg[0], "%d")
local SOCKET_DIR = require('fio').cwd()
+
+local TIMEOUT = tonumber(arg[1])
+local CON_TIMEOUT = arg[2] and tonumber(arg[2]) or 30.0
+
local function instance_uri(instance_id)
return SOCKET_DIR..'/rebootstrap'..instance_id..'.sock';
end
@@ -14,8 +18,8 @@ require('console').listen(os.getenv('ADMIN'))
box.cfg({
listen = instance_uri(INSTANCE_ID),
instance_uuid = '12345678-abcd-1234-abcd-123456789ef' .. INSTANCE_ID,
- replication_timeout = 0.1,
- replication_connect_timeout = 0.5,
+ replication_timeout = TIMEOUT,
+ replication_connect_timeout = CON_TIMEOUT,
replication = {
instance_uri(1);
instance_uri(2);
diff --git a/test/replication/rebootstrap.result b/test/replication/rebootstrap.result
index afbfc8e65..ea390c19f 100644
--- a/test/replication/rebootstrap.result
+++ b/test/replication/rebootstrap.result
@@ -4,7 +4,7 @@ test_run = require('test_run').new()
SERVERS = {'rebootstrap1', 'rebootstrap2'}
---
...
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
---
...
test_run:wait_fullmesh(SERVERS)
@@ -20,11 +20,11 @@ test_run:cmd('stop server rebootstrap1')
---
- true
...
-test_run:cmd('restart server rebootstrap2 with cleanup=True, wait=False, wait_load=False')
+test_run:cmd('restart server rebootstrap2 with cleanup=True, wait=False, wait_load=False, args="0.1 2.0"')
---
- true
...
-test_run:cmd('start server rebootstrap1')
+test_run:cmd('start server rebootstrap1 with args="0.1 0.5"')
---
- true
...
diff --git a/test/replication/rebootstrap.test.lua b/test/replication/rebootstrap.test.lua
index 954726ddb..8ddf77912 100644
--- a/test/replication/rebootstrap.test.lua
+++ b/test/replication/rebootstrap.test.lua
@@ -2,7 +2,7 @@ test_run = require('test_run').new()
SERVERS = {'rebootstrap1', 'rebootstrap2'}
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
test_run:wait_fullmesh(SERVERS)
--
@@ -12,8 +12,8 @@ test_run:wait_fullmesh(SERVERS)
-- in 'orphan' mode.
--
test_run:cmd('stop server rebootstrap1')
-test_run:cmd('restart server rebootstrap2 with cleanup=True, wait=False, wait_load=False')
-test_run:cmd('start server rebootstrap1')
+test_run:cmd('restart server rebootstrap2 with cleanup=True, wait=False, wait_load=False, args="0.1 2.0"')
+test_run:cmd('start server rebootstrap1 with args="0.1 0.5"')
test_run:cmd('switch rebootstrap1')
box.info.status -- running
diff --git a/test/replication/recover_missing_xlog.result b/test/replication/recover_missing_xlog.result
index 027f8761e..5ed05c635 100644
--- a/test/replication/recover_missing_xlog.result
+++ b/test/replication/recover_missing_xlog.result
@@ -8,7 +8,7 @@ SERVERS = { 'autobootstrap1', 'autobootstrap2', 'autobootstrap3' }
---
...
-- Start servers
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
---
...
-- Wait for full mesh
@@ -66,7 +66,7 @@ fio.unlink(fio.pathjoin(fio.abspath("."), string.format('autobootstrap1/%020d.xl
---
- true
...
-test_run:cmd("start server autobootstrap1")
+test_run:cmd('start server autobootstrap1 with args="0.1 0.5"')
---
- true
...
diff --git a/test/replication/recover_missing_xlog.test.lua b/test/replication/recover_missing_xlog.test.lua
index 57bc7d31f..d2d378837 100644
--- a/test/replication/recover_missing_xlog.test.lua
+++ b/test/replication/recover_missing_xlog.test.lua
@@ -3,7 +3,7 @@ test_run = env.new()
SERVERS = { 'autobootstrap1', 'autobootstrap2', 'autobootstrap3' }
-- Start servers
-test_run:create_cluster(SERVERS)
+test_run:create_cluster(SERVERS, "replication", {args="0.1"})
-- Wait for full mesh
test_run:wait_fullmesh(SERVERS)
@@ -28,7 +28,7 @@ fio = require('fio')
-- Also check that there is no concurrency, i.e. master is
-- in 'read-only' mode unless it receives all data.
fio.unlink(fio.pathjoin(fio.abspath("."), string.format('autobootstrap1/%020d.xlog', 8)))
-test_run:cmd("start server autobootstrap1")
+test_run:cmd('start server autobootstrap1 with args="0.1 0.5"')
test_run:cmd("switch autobootstrap1")
for i = 10, 19 do box.space.test:insert{i, 'test' .. i} end
diff --git a/test/replication/replica_no_quorum.lua b/test/replication/replica_no_quorum.lua
index b9edeea94..c30c043cc 100644
--- a/test/replication/replica_no_quorum.lua
+++ b/test/replication/replica_no_quorum.lua
@@ -5,7 +5,8 @@ box.cfg({
replication = os.getenv("MASTER"),
memtx_memory = 107374182,
replication_connect_quorum = 0,
- replication_connect_timeout = 0.1,
+ replication_timeout = 0.1,
+ replication_connect_timeout = 0.5,
})
require('console').listen(os.getenv('ADMIN'))
diff --git a/test/replication/replica_timeout.lua b/test/replication/replica_timeout.lua
index 64f119763..38922fa3d 100644
--- a/test/replication/replica_timeout.lua
+++ b/test/replication/replica_timeout.lua
@@ -1,13 +1,14 @@
#!/usr/bin/env tarantool
local TIMEOUT = tonumber(arg[1])
+local CON_TIMEOUT = arg[2] and tonumber(arg[2]) or 30.0
box.cfg({
listen = os.getenv("LISTEN"),
replication = os.getenv("MASTER"),
memtx_memory = 107374182,
replication_timeout = TIMEOUT,
- replication_connect_timeout = TIMEOUT * 3,
+ replication_connect_timeout = CON_TIMEOUT,
})
require('console').listen(os.getenv('ADMIN'))
diff --git a/test/replication/replica_uuid_ro.lua b/test/replication/replica_uuid_ro.lua
index 8e1c6cc47..d5ba55852 100644
--- a/test/replication/replica_uuid_ro.lua
+++ b/test/replication/replica_uuid_ro.lua
@@ -5,6 +5,10 @@ local INSTANCE_ID = string.match(arg[0], "%d")
local USER = 'cluster'
local PASSWORD = 'somepassword'
local SOCKET_DIR = require('fio').cwd()
+
+local TIMEOUT = tonumber(arg[2])
+local CON_TIMEOUT = arg[3] and tonumber(arg[3]) or 30.0
+
local function instance_uri(instance_id)
--return 'localhost:'..(3310 + instance_id)
return SOCKET_DIR..'/replica_uuid_ro'..instance_id..'.sock';
@@ -22,7 +26,8 @@ box.cfg({
USER..':'..PASSWORD..'@'..instance_uri(2);
};
read_only = (INSTANCE_ID ~= '1' and true or false);
- replication_connect_timeout = 0.5,
+ replication_timeout = TIMEOUT;
+ replication_connect_timeout = CON_TIMEOUT;
})
box.once("bootstrap", function()
diff --git a/test/replication/replicaset_ro_mostly.result b/test/replication/replicaset_ro_mostly.result
index b9e8f1fe8..1ce7d6f8e 100644
--- a/test/replication/replicaset_ro_mostly.result
+++ b/test/replication/replicaset_ro_mostly.result
@@ -27,7 +27,7 @@ UUID = sort({uuid1, uuid2}, sort_cmp)
create_cluster_cmd1 = 'create server %s with script="replication/%s.lua"'
---
...
-create_cluster_cmd2 = 'start server %s with args="%s", wait_load=False, wait=False'
+create_cluster_cmd2 = 'start server %s with args="%s %s", wait_load=False, wait=False'
---
...
test_run:cmd("setopt delimiter ';'")
@@ -37,7 +37,9 @@ test_run:cmd("setopt delimiter ';'")
function create_cluster_uuid(servers, uuids)
for i, name in ipairs(servers) do
test_run:cmd(create_cluster_cmd1:format(name, name))
- test_run:cmd(create_cluster_cmd2:format(name, uuids[i]))
+ end
+ for i, name in ipairs(servers) do
+ test_run:cmd(create_cluster_cmd2:format(name, uuids[i], "0.1"))
end
end;
---
@@ -61,7 +63,7 @@ test_run:cmd(create_cluster_cmd1:format(name, name))
---
- true
...
-test_run:cmd(create_cluster_cmd2:format(name, uuid.new()))
+test_run:cmd(create_cluster_cmd2:format(name, uuid.new(), "0.1"))
---
- true
...
diff --git a/test/replication/replicaset_ro_mostly.test.lua b/test/replication/replicaset_ro_mostly.test.lua
index f2c2d0d11..c75af7218 100644
--- a/test/replication/replicaset_ro_mostly.test.lua
+++ b/test/replication/replicaset_ro_mostly.test.lua
@@ -12,13 +12,15 @@ function sort(t) table.sort(t, sort_cmp) return t end
UUID = sort({uuid1, uuid2}, sort_cmp)
create_cluster_cmd1 = 'create server %s with script="replication/%s.lua"'
-create_cluster_cmd2 = 'start server %s with args="%s", wait_load=False, wait=False'
+create_cluster_cmd2 = 'start server %s with args="%s %s", wait_load=False, wait=False'
test_run:cmd("setopt delimiter ';'")
function create_cluster_uuid(servers, uuids)
for i, name in ipairs(servers) do
test_run:cmd(create_cluster_cmd1:format(name, name))
- test_run:cmd(create_cluster_cmd2:format(name, uuids[i]))
+ end
+ for i, name in ipairs(servers) do
+ test_run:cmd(create_cluster_cmd2:format(name, uuids[i], "0.1"))
end
end;
test_run:cmd("setopt delimiter ''");
@@ -30,7 +32,7 @@ test_run:wait_fullmesh(SERVERS)
-- Add third replica
name = 'replica_uuid_ro3'
test_run:cmd(create_cluster_cmd1:format(name, name))
-test_run:cmd(create_cluster_cmd2:format(name, uuid.new()))
+test_run:cmd(create_cluster_cmd2:format(name, uuid.new(), "0.1"))
test_run:cmd('switch replica_uuid_ro3')
test_run:cmd('switch default')
--
2.15.2 (Apple Git-101.1)
^ permalink raw reply [flat|nested] 5+ messages in thread* [PATCH v4 3/3] replication: do not ignore replication_connect_quorum.
2018-08-14 10:02 [PATCH v4 0/3] replication: do not ignore replication_connect_quorum Serge Petrenko
2018-08-14 10:02 ` [PATCH v4 1/3] test: update test-run Serge Petrenko
2018-08-14 10:02 ` [PATCH v4 2/3] Add arguments to replication test instances Serge Petrenko
@ 2018-08-14 10:02 ` Serge Petrenko
2018-08-14 17:07 ` [PATCH v4 0/3] " Vladimir Davydov
3 siblings, 0 replies; 5+ messages in thread
From: Serge Petrenko @ 2018-08-14 10:02 UTC (permalink / raw)
To: tarantool-patches; +Cc: georgy, vdavydov.dev, Serge Petrenko
On bootstrap and after replication reconfiguration
replication_connect_quorum was ignored. The instance tried to connect to
every replica listed in replication parameter, and errored if it wasn't
possible.
The patch alters this behaviour. An instance still tries to connect to
every node listed in replication, but does not raise an error if it was
able to connect to at least replication_connect_quorum instances.
Closes #3428
@TarantoolBot document
Title: replication_connect_quorum is not ignored.
Now on replica set bootstrap and in case of replication reconfiguration
(e.g. calling box.cfg{replication=...} for the second time) tarantool
doesn't fail, if it couldn't connect to to every replica, but could
connect to replication_connect_quorum replicas. If after
replication_connect_timeout seconds the instance is not connected to at
least replication_connect_quorum other instances, we throw an error.
---
src/box/box.cc | 35 +++++++++++++++++---------
src/box/replication.cc | 11 ++++++---
src/box/replication.h | 7 +++---
test/replication/quorum.result | 49 +++++++++++++++++++++++++++++++++++++
test/replication/quorum.test.lua | 20 +++++++++++++++
test/replication/replica_quorum.lua | 24 ++++++++++++++++++
6 files changed, 128 insertions(+), 18 deletions(-)
create mode 100644 test/replication/replica_quorum.lua
diff --git a/src/box/box.cc b/src/box/box.cc
index e3eb2738f..8d7454d1f 100644
--- a/src/box/box.cc
+++ b/src/box/box.cc
@@ -595,7 +595,7 @@ cfg_get_replication(int *p_count)
* don't start appliers.
*/
static void
-box_sync_replication(double timeout, bool connect_all)
+box_sync_replication(bool connect_quorum)
{
int count = 0;
struct applier **appliers = cfg_get_replication(&count);
@@ -607,7 +607,7 @@ box_sync_replication(double timeout, bool connect_all)
applier_delete(appliers[i]); /* doesn't affect diag */
});
- replicaset_connect(appliers, count, timeout, connect_all);
+ replicaset_connect(appliers, count, connect_quorum);
guard.is_active = false;
}
@@ -625,8 +625,13 @@ box_set_replication(void)
}
box_check_replication();
- /* Try to connect to all replicas within the timeout period */
- box_sync_replication(replication_connect_timeout, true);
+ /*
+ * Try to connect to all replicas within the timeout period.
+ * The configuration will succeed as long as we've managed
+ * to connect to at least replication_connect_quorum
+ * masters.
+ */
+ box_sync_replication(true);
/* Follow replica */
replicaset_follow();
}
@@ -1865,8 +1870,13 @@ box_cfg_xc(void)
title("orphan");
- /* Wait for the cluster to start up */
- box_sync_replication(replication_connect_timeout, false);
+ /*
+ * In case of recovering from a checkpoint we
+ * don't need to wait for 'quorum' masters, since
+ * the recovered _cluster space will have all the
+ * information about cluster.
+ */
+ box_sync_replication(false);
} else {
if (!tt_uuid_is_nil(&instance_uuid))
INSTANCE_UUID = instance_uuid;
@@ -1883,12 +1893,15 @@ box_cfg_xc(void)
/*
* Wait for the cluster to start up.
*
- * Note, when bootstrapping a new instance, we have to
- * connect to all masters to make sure all replicas
- * receive the same replica set UUID when a new cluster
- * is deployed.
+ * Note, when bootstrapping a new instance, we try to
+ * connect to all masters during timeout to make sure
+ * all replicas recieve the same replica set UUID when
+ * a new cluster is deployed.
+ * If we fail to do so, settle with connecting to
+ * 'replication_connect_quorum' masters.
+ * If this also fails, throw an error.
*/
- box_sync_replication(TIMEOUT_INFINITY, true);
+ box_sync_replication(true);
/* Bootstrap a new master */
bootstrap(&replicaset_uuid, &is_bootstrap_leader);
}
diff --git a/src/box/replication.cc b/src/box/replication.cc
index 528fe4459..861ce34ea 100644
--- a/src/box/replication.cc
+++ b/src/box/replication.cc
@@ -46,7 +46,7 @@ struct tt_uuid INSTANCE_UUID;
struct tt_uuid REPLICASET_UUID;
double replication_timeout = 1.0; /* seconds */
-double replication_connect_timeout = 4.0; /* seconds */
+double replication_connect_timeout = 30.0; /* seconds */
int replication_connect_quorum = REPLICATION_CONNECT_QUORUM_ALL;
double replication_sync_lag = 10.0; /* seconds */
@@ -540,7 +540,7 @@ applier_on_connect_f(struct trigger *trigger, void *event)
void
replicaset_connect(struct applier **appliers, int count,
- double timeout, bool connect_all)
+ bool connect_quorum)
{
if (count == 0) {
/* Cleanup the replica set. */
@@ -571,6 +571,9 @@ replicaset_connect(struct applier **appliers, int count,
state.connected = state.failed = 0;
fiber_cond_create(&state.wakeup);
+ double timeout = replication_connect_timeout;
+ int quorum = MIN(count, replication_connect_quorum);
+
/* Add triggers and start simulations connection to remote peers */
for (int i = 0; i < count; i++) {
struct applier *applier = appliers[i];
@@ -587,7 +590,7 @@ replicaset_connect(struct applier **appliers, int count,
double wait_start = ev_monotonic_now(loop());
if (fiber_cond_wait_timeout(&state.wakeup, timeout) != 0)
break;
- if (state.failed > 0 && connect_all)
+ if (count - state.failed < quorum)
break;
timeout -= ev_monotonic_now(loop()) - wait_start;
}
@@ -595,7 +598,7 @@ replicaset_connect(struct applier **appliers, int count,
say_crit("failed to connect to %d out of %d replicas",
count - state.connected, count);
/* Timeout or connection failure. */
- if (connect_all)
+ if (connect_quorum && state.connected < quorum)
goto error;
} else {
say_verbose("connected to %d replicas", state.connected);
diff --git a/src/box/replication.h b/src/box/replication.h
index 95122eb45..06a2867b6 100644
--- a/src/box/replication.h
+++ b/src/box/replication.h
@@ -357,12 +357,13 @@ replicaset_add(uint32_t replica_id, const struct tt_uuid *instance_uuid);
* \param appliers the array of appliers
* \param count size of appliers array
* \param timeout connection timeout
- * \param connect_all if this flag is set, fail unless all
- * appliers have successfully connected
+ * \param connect_quorum if this flag is set, fail unless at
+ * least replication_connect_quorum
+ * appliers have successfully connected.
*/
void
replicaset_connect(struct applier **appliers, int count,
- double timeout, bool connect_all);
+ bool connect_quorum);
/**
* Resume all appliers registered with the replica set.
diff --git a/test/replication/quorum.result b/test/replication/quorum.result
index a55b0d087..265b099b7 100644
--- a/test/replication/quorum.result
+++ b/test/replication/quorum.result
@@ -401,3 +401,52 @@ test_run:cmd("switch default")
test_run:drop_cluster(SERVERS)
---
...
+-- Test that quorum is not ignored neither during bootstrap, nor
+-- during reconfiguration.
+box.schema.user.grant('guest', 'replication')
+---
+...
+test_run:cmd('create server replica_quorum with script="replication/replica_quorum.lua"')
+---
+- true
+...
+-- Arguments are: replication_connect_quorum, replication_timeout
+-- replication_connect_timeout.
+-- If replication_connect_quorum was ignored here, the instance
+-- would exit with an error.
+test_run:cmd('start server replica_quorum with wait=True, wait_load=True, args="1 0.05 0.1"')
+---
+- true
+...
+test_run:cmd('switch replica_quorum')
+---
+- true
+...
+-- If replication_connect_quorum was ignored here, the instance
+-- would exit with an error.
+box.cfg{replication={INSTANCE_URI, nonexistent_uri(1)}}
+---
+...
+box.info.id
+---
+- 1
+...
+test_run:cmd('switch default')
+---
+- true
+...
+test_run:cmd('stop server replica_quorum')
+---
+- true
+...
+test_run:cmd('cleanup server replica_quorum')
+---
+- true
+...
+test_run:cmd('delete server replica_quorum')
+---
+- true
+...
+box.schema.user.revoke('guest', 'replication')
+---
+...
diff --git a/test/replication/quorum.test.lua b/test/replication/quorum.test.lua
index 8290f8ea5..5a43275c2 100644
--- a/test/replication/quorum.test.lua
+++ b/test/replication/quorum.test.lua
@@ -150,3 +150,23 @@ box.space.test:select()
test_run:cmd("switch default")
-- Cleanup.
test_run:drop_cluster(SERVERS)
+
+-- Test that quorum is not ignored neither during bootstrap, nor
+-- during reconfiguration.
+box.schema.user.grant('guest', 'replication')
+test_run:cmd('create server replica_quorum with script="replication/replica_quorum.lua"')
+-- Arguments are: replication_connect_quorum, replication_timeout
+-- replication_connect_timeout.
+-- If replication_connect_quorum was ignored here, the instance
+-- would exit with an error.
+test_run:cmd('start server replica_quorum with wait=True, wait_load=True, args="1 0.05 0.1"')
+test_run:cmd('switch replica_quorum')
+-- If replication_connect_quorum was ignored here, the instance
+-- would exit with an error.
+box.cfg{replication={INSTANCE_URI, nonexistent_uri(1)}}
+box.info.id
+test_run:cmd('switch default')
+test_run:cmd('stop server replica_quorum')
+test_run:cmd('cleanup server replica_quorum')
+test_run:cmd('delete server replica_quorum')
+box.schema.user.revoke('guest', 'replication')
diff --git a/test/replication/replica_quorum.lua b/test/replication/replica_quorum.lua
new file mode 100644
index 000000000..dd42b8214
--- /dev/null
+++ b/test/replication/replica_quorum.lua
@@ -0,0 +1,24 @@
+#!/usr/bin/env tarantool
+
+local SOCKET_DIR = require('fio').cwd()
+
+local QUORUM = tonumber(arg[1])
+local TIMEOUT = arg[2] and tonumber(arg[2]) or 0.1
+local CON_TIMEOUT = arg[3] and tonumber(arg[3]) or 30.0
+INSTANCE_URI = SOCKET_DIR .. '/replica_quorum.sock'
+
+function nonexistent_uri(id)
+ return SOCKET_DIR .. '/replica_quorum' .. (1000 + id) .. '.sock'
+end
+
+require('console').listen(os.getenv('ADMIN'))
+
+box.cfg{
+ listen = INSTANCE_URI,
+ replication_timeout = TIMEOUT,
+ replication_connect_timeout = CON_TIMEOUT,
+ replication_connect_quorum = QUORUM,
+ replication = {INSTANCE_URI,
+ nonexistent_uri(1),
+ nonexistent_uri(2)}
+}
--
2.15.2 (Apple Git-101.1)
^ permalink raw reply [flat|nested] 5+ messages in thread