本节主要介绍对 secondary_check_script 调用 masterha_secondary_check 脚本的说明。
[root@yang-03 ~]# cat /usr/bin/masterha_secondary_check
#!/usr/bin/env perl
···
$| = 1;
GetOptions(
'help' => \$help,
'version' => \$version,
'secondary_host=s' => \@monitoring_servers,
'user=s' => \$ssh_user,
'port=s' => \$ssh_port,
'options=s' => \$ssh_options,
'master_host=s' => \$master_host,
'master_ip=s' => \$master_ip,
'master_port=i' => \$master_port,
'master_user=s' => \$master_user,
'master_password=s' => \$master_password,
'ping_type=s' => \$ping_type,
'timeout=i' => \$timeout,
);
if ($version) {
print "masterha_secondary_check version $MHA::ManagerConst::VERSION.\n";
exit 0;
}
if ($help) {
pod2usage(0);
}
unless ($master_host) {
pod2usage(1);
}
sub exit_by_signal {
exit 1;
}
local $SIG{INT} = $SIG{HUP} = $SIG{QUIT} = $SIG{TERM} = \&exit_by_signal;
$ssh_user = "root" unless ($ssh_user);
$ssh_port = 22 unless ($ssh_port);
$master_port = 3306 unless ($master_port);
if ($ssh_options) {
$MHA::ManagerConst::SSH_OPT_CHECK = $ssh_options;
}
$MHA::ManagerConst::SSH_OPT_CHECK =~ s/VAR_CONNECT_TIMEOUT/$timeout/;
# 0: master is not reachable from all monotoring servers
# 1: unknown errors
# 2: at least one of monitoring servers is not reachable from this script
# 3: master is reachable from at least one of monitoring servers
my $exit_code = 0;
foreach my $monitoring_server (@monitoring_servers) {
my $ssh_user_host = $ssh_user . '@' . $monitoring_server;
my $command =
"ssh $MHA::ManagerConst::SSH_OPT_CHECK -p $ssh_port $ssh_user_host \"perl -e "
. "\\\"use IO::Socket::INET; my \\\\\\\$sock = IO::Socket::INET->new"
. "(PeerAddr => \\\\\\\"$master_host\\\\\\\", PeerPort=> $master_port, "
. "Proto =>'tcp', Timeout => $timeout); if(\\\\\\\$sock) { close(\\\\\\\$sock); "
. "exit 3; } exit 0;\\\" \"";
my $ret = system($command);
$ret = $ret >> 8;
if ( $ret == 0 ) {
print
"Monitoring server $monitoring_server is reachable, Master is not reachable from $monitoring_server. OK.\n";
next;
}
if ( $ret == 3 ) {
if ( defined $ping_type
&& $ping_type eq $MHA::ManagerConst::PING_TYPE_INSERT )
{
my $ret_insert;
my $command_insert =
"ssh $MHA::ManagerConst::SSH_OPT_CHECK -p $ssh_port $ssh_user_host \'"
. "/usr/bin/mysql -u$master_user -p$master_password -h$master_host "
. "-e \"CREATE DATABASE IF NOT EXISTS infra; "
. "CREATE TABLE IF NOT EXISTS infra.chk_masterha (\\`key\\` tinyint NOT NULL primary key,\\`val\\` int(10) unsigned NOT NULL DEFAULT '0'\) engine=InnoDB; "
. "INSERT INTO infra.chk_masterha values (1,unix_timestamp()) ON DUPLICATE KEY UPDATE val=unix_timestamp()\"\'";
my $sigalrm_timeout = 3;
eval {
local $SIG{ALRM} = sub {
die "timeout.\n";
};
alarm $sigalrm_timeout;
$ret_insert = system($command_insert);
$ret_insert = $ret_insert >> 8;
alarm 0;
};
if ( $@ || $ret_insert != 0 ) {
print
"Monitoring server $monitoring_server is reachable, Master is not writable from $monitoring_server. OK.\n";
next;
}
}
print "Master is reachable from $monitoring_server!\n";
$exit_code = 3;
last;
}
else {
print "Monitoring server $monitoring_server is NOT reachable!\n";
$exit_code = 2;
last;
}
}
exit $exit_code;
···
看到上面的脚本有点头大,作为一个专一的 DBA,一直追求着 shell,Perl 语言不熟啊,细看之后发现与 shell 脚本有异曲同工之妙,从相应的注释中,我们找到了该检查脚本的返回值,即 0,1,2,3,分别代表的含义如下:
# 0: master is not reachable from all monotoring servers
# 1: unknown errors
# 2: at least one of monitoring servers is not reachable from this script
# 3: master is reachable from at least one of monitoring servers
下面就让我们做几个测试体会体会吧。