[PATCH] replication: fix bug with zero replication_connect_quorum

Konstantin Belyavskiy k.belyavskiy at tarantool.org
Fri Apr 6 11:39:45 MSK 2018


If 'box.cfg.read_only' is false, 'replication' defines at least one
replica (other than itself), but they are not available at the time
of box.cfg execution and replication_connect_quorum is set to zero,
master displays 'orphan' status instead of 'running' since logic
which cnange this state is executed only after successfull connection.

Closes #3278
---
ticket: https://github.com/tarantool/tarantool/issues/3278
branch: https://github.com/tarantool/tarantool/compare/gh-3278-fix-bug-with-zero-replication-connect-quorum

 src/box/replication.cc                 |  8 +++-
 test/replication/check_quorum.result   | 86 ++++++++++++++++++++++++++++++++++
 test/replication/check_quorum.test.lua | 30 ++++++++++++
 test/replication/replica_params.lua    | 21 +++++++++
 4 files changed, 143 insertions(+), 2 deletions(-)
 create mode 100644 test/replication/check_quorum.result
 create mode 100644 test/replication/check_quorum.test.lua
 create mode 100644 test/replication/replica_params.lua

diff --git a/src/box/replication.cc b/src/box/replication.cc
index b1c84d36c..760f83751 100644
--- a/src/box/replication.cc
+++ b/src/box/replication.cc
@@ -600,8 +600,12 @@ error:
 void
 replicaset_follow(void)
 {
-	if (replicaset.applier.total == 0) {
-		/* Replication is not configured. */
+	if (replicaset.applier.total == 0 || replicaset_quorum() == 0) {
+		/*
+		 * Replication is not configured or quorum is set to
+		 * zero so in the latter case we have no need to wait
+		 * for others.
+		 */
 		box_clear_orphan();
 		return;
 	}
diff --git a/test/replication/check_quorum.result b/test/replication/check_quorum.result
new file mode 100644
index 000000000..017d8380f
--- /dev/null
+++ b/test/replication/check_quorum.result
@@ -0,0 +1,86 @@
+--
+-- gh-3278: test different replication and replication_connect_quorum configs.
+--
+env = require('test_run')
+---
+...
+test_run = env.new()
+---
+...
+socket = require('socket')
+---
+...
+s = socket.tcp_server('localhost', 3371, function() end)
+---
+...
+test_run:cmd('switch default')
+---
+- true
+...
+listen = os.getenv("LISTEN")
+---
+...
+box.cfg{listen = listen, replication_timeout = 1, read_only = false}
+---
+...
+box.info.status -- running
+---
+- running
+...
+test_run:cmd("restart server default")
+listen = os.getenv("LISTEN")
+---
+...
+box.cfg{listen = listen, replication = { listen }, replication_timeout = 1, read_only = false}
+---
+...
+box.info.status -- running
+---
+- running
+...
+test_run:cmd("restart server default")
+listen = os.getenv("LISTEN")
+---
+...
+box.cfg{listen = listen, replication = { listen }, replication_timeout = 1, read_only = false, replication_connect_quorum = 1}
+---
+...
+box.info.status -- running
+---
+- running
+...
+test_run:cmd("create server replica with rpl_master=default, script='replication/replica_params.lua'")
+---
+- true
+...
+test_run:cmd("start server replica with args='2 1'")
+---
+- true
+...
+test_run:cmd('switch replica')
+---
+- true
+...
+box.info.status -- running
+---
+- running
+...
+box.cfg{replication_connect_quorum = 1}
+---
+...
+test_run:cmd('switch default')
+---
+- true
+...
+test_run:cmd("restart server replica with args='2 2'")
+---
+- true
+...
+test_run:cmd('switch replica')
+---
+- true
+...
+box.info.status -- orphan
+---
+- orphan
+...
diff --git a/test/replication/check_quorum.test.lua b/test/replication/check_quorum.test.lua
new file mode 100644
index 000000000..497d2af6c
--- /dev/null
+++ b/test/replication/check_quorum.test.lua
@@ -0,0 +1,30 @@
+--
+-- gh-3278: test different replication and replication_connect_quorum configs.
+--
+
+env = require('test_run')
+test_run = env.new()
+socket = require('socket')
+s = socket.tcp_server('localhost', 3371, function() end)
+test_run:cmd('switch default')
+listen = os.getenv("LISTEN")
+box.cfg{listen = listen, replication_timeout = 1, read_only = false}
+box.info.status -- running
+test_run:cmd("restart server default")
+listen = os.getenv("LISTEN")
+box.cfg{listen = listen, replication = { listen }, replication_timeout = 1, read_only = false}
+box.info.status -- running
+test_run:cmd("restart server default")
+listen = os.getenv("LISTEN")
+box.cfg{listen = listen, replication = { listen }, replication_timeout = 1, read_only = false, replication_connect_quorum = 1}
+box.info.status -- running
+
+test_run:cmd("create server replica with rpl_master=default, script='replication/replica_params.lua'")
+test_run:cmd("start server replica with args='2 1'")
+test_run:cmd('switch replica')
+box.info.status -- running
+box.cfg{replication_connect_quorum = 1}
+test_run:cmd('switch default')
+test_run:cmd("restart server replica with args='2 2'")
+test_run:cmd('switch replica')
+box.info.status -- orphan
diff --git a/test/replication/replica_params.lua b/test/replication/replica_params.lua
new file mode 100644
index 000000000..73a15e6c3
--- /dev/null
+++ b/test/replication/replica_params.lua
@@ -0,0 +1,21 @@
+#!/usr/bin/env tarantool
+
+local quorum = tonumber(arg[1])
+local n_replics = tonumber(arg[2])
+listen = os.getenv("LISTEN")
+-- Test different replicaset configurations:
+-- First, when the only address in the replicaset is itself.
+repl = {listen}
+-- To test situation with second master unavailable, add
+-- second address (should be empty).
+if n_replics == 2 then repl = { listen, '127.0.0.1:3371' } end
+
+box.cfg({
+    listen              = listen,
+    replication         = repl,
+    memtx_memory        = 107374182,
+    replication_connect_quorum = quorum,
+    replication_connect_timeout = 0.1,
+})
+
+require('console').listen(os.getenv('ADMIN'))
-- 
2.14.3 (Apple Git-98)




More information about the Tarantool-patches mailing list