hcq/MaxScale
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch '2.3' into develop

Browse Source
This commit is contained in:
Markus Mäkelä
2018-11-16 16:06:31 +02:00
parent 7d57e132a3 21be08f84a
commit adb06615f0
5 changed files with 84 additions and 58 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
BUILD/install_build_deps.sh
View File

@ -16,7 +16,7 @@ then
build-essential libssl-dev ncurses-dev bison flex \
perl libtool libpcre3-dev tcl tcl-dev uuid \
uuid-dev libsqlite3-dev liblzma-dev libpam0g-dev pkg-config \
libedit-dev
libedit-dev libsystemd-dev
## separatelibgnutls installation process for Ubuntu Trusty
cat /etc/*release | grep -E "Trusty|wheezy"
@ -71,6 +71,9 @@ else
# Attempt to install libasan, it'll only work on CentOS 7
sudo yum install -y --nogpgcheck libasan
# Attempt to install systemd-devel, doesn't work on CentOS 6
sudo yum install -y systemd-devel
grep "release 6" /etc/redhat-release
if [ $? == 0 ]
then

20
Documentation/Monitors/MariaDB-Monitor.md
View File

@ -141,6 +141,9 @@ switchover and rejoin-specific parameters are listed in their own
Deprecated and unused as of MaxScale 2.3. Can be defined but is ignored.
Is effectively always on. The monitor uses the "Seconds_Behind_Master"-field of
"SHOW SLAVE STATUS" to get the replication lag.
### `detect_stale_master`
Allow previous master to be available even in case of stopped or misconfigured
@ -365,12 +368,17 @@ operations.
### Manual activation
Cluster operations can be activated manually through the REST API, MaxCtrl or
MaxAdmin. The commands are only performed when MaxScale is in active mode. All
commands require the monitor instance name as the first parameter. Failover
MaxAdmin. The commands are only performed when MaxScale is in active mode. The
commands generally match their automatic versions. The exception is _rejoin_, in
which the manual command allows rejoining even when the joining server has empty
gtid:s. This rule allows the user to force a rejoin on a server without binary
logs.
All commands require the monitor instance name as the first parameter. Failover
selects the new master server automatically and does not require additional
parameters. Rejoin requires the name of the joining server as second parameter.
Replication reset accepts the name of the new master server as second
parameter. If not given, the current master is selected.
Replication reset accepts the name of the new master server as second parameter.
If not given, the current master is selected.
Switchover takes one to three parameters. If only the monitor name is given,
switchover will autoselect both the slave to promote and the current master as
@ -398,8 +406,8 @@ to demote (OldMasterServ). For rejoin, the server to join (OldMasterServ) is
required. Replication reset requires the server to promote (NewMasterServ).
It is safe to perform manual operations even with automatic failover, switchover
or rejoin enabled since the automatic operations cannot happen simultaneously
with the manual one.
or rejoin enabled since automatic operations cannot happen simultaneously
with manual ones.
If a switchover or failover fails, automatic failover is disabled to prevent
master changes to a possibly malfunctioning cluster. Automatic failover can be

64
server/modules/monitor/mariadbmon/cluster_manipulation.cc
View File

@ -115,16 +115,42 @@ bool MariaDBMonitor::manual_rejoin(SERVER* rejoin_server, json_t** output)
if (mon_slave_cand)
{
MariaDBServer* slave_cand = get_server_info(mon_slave_cand);
if (server_is_rejoin_suspect(slave_cand, output))
{
if (m_master->update_gtids())
string gtid_update_error;
if (m_master->update_gtids(&gtid_update_error))
{
// The manual version of rejoin does not need to be as careful as the automatic one.
// The rules are mostly the same, the only difference is that a server with empty gtid:s
// can be rejoined manually.
// TODO: Add the warning to JSON output.
string no_rejoin_reason;
if (slave_cand->can_replicate_from(m_master, &no_rejoin_reason))
bool safe_rejoin = slave_cand->can_replicate_from(m_master, &no_rejoin_reason);
bool empty_gtid = slave_cand->m_gtid_current_pos.empty();
bool rejoin_allowed = false;
if (safe_rejoin)
{
ServerArray joinable_server;
joinable_server.push_back(slave_cand);
rejoin_allowed = true;
}
else
{
if (empty_gtid)
{
rejoin_allowed = true;
MXB_WARNING("gtid_curren_pos of %s is empty. Manual rejoin is unsafe "
"but allowed.", rejoin_serv_name);
}
else
{
PRINT_MXS_JSON_ERROR(output, "%s cannot replicate from master server %s: %s",
rejoin_serv_name, m_master->name(),
no_rejoin_reason.c_str());
}
}
if (rejoin_allowed)
{
ServerArray joinable_server = {slave_cand};
if (do_rejoin(joinable_server, output) == 1)
{
rval = true;
@ -135,28 +161,18 @@ bool MariaDBMonitor::manual_rejoin(SERVER* rejoin_server, json_t** output)
PRINT_MXS_JSON_ERROR(output, "Rejoin attempted but failed.");
}
}
else
{
PRINT_MXS_JSON_ERROR(output,
"Server '%s' cannot replicate from cluster master '%s': "
"%s.",
rejoin_serv_name,
m_master->name(),
no_rejoin_reason.c_str());
}
}
else
{
PRINT_MXS_JSON_ERROR(output,
"Cluster master '%s' gtid info could not be updated.",
m_master->name());
"The GTIDs of master server %s could not be updated: %s",
m_master->name(), gtid_update_error.c_str());
}
} // server_is_rejoin_suspect has added any error messages to the output, no need to print here
}
else
{
PRINT_MXS_JSON_ERROR(output,
"The given server '%s' is not monitored by this monitor.",
PRINT_MXS_JSON_ERROR(output, "The given server '%s' is not monitored by this monitor.",
rejoin_serv_name);
}
}
@ -687,7 +703,8 @@ bool MariaDBMonitor::get_joinable_servers(ServerArray* output)
bool comm_ok = true;
if (!suspects.empty())
{
if (m_master->update_gtids())
string gtid_update_error;
if (m_master->update_gtids(&gtid_update_error))
{
for (size_t i = 0; i < suspects.size(); i++)
{
@ -710,6 +727,8 @@ bool MariaDBMonitor::get_joinable_servers(ServerArray* output)
}
else
{
MXS_ERROR("The GTIDs of master server %s could not be updated while attempting an automatic "
"rejoin: %s", m_master->name(), gtid_update_error.c_str());
comm_ok = false;
}
}
@ -1760,10 +1779,7 @@ void MariaDBMonitor::handle_auto_rejoin()
MXS_NOTICE("%d server(s) redirected or rejoined the cluster.", joins);
}
}
else
{
MXS_ERROR("Query error to master '%s' prevented a possible rejoin operation.", m_master->name());
}
// get_joinable_servers prints an error if master is unresponsive
}
void MariaDBMonitor::report_and_disable(const string& operation, const string& setting_name,
@ -1864,4 +1880,4 @@ MariaDBMonitor::FailoverParams::FailoverParams(const ServerOperation& promotion,
, demotion_target(demotion_target)
, general(general)
{
}
}

43
server/modules/monitor/mariadbmon/mariadbserver.cc
View File

@ -693,35 +693,32 @@ json_t* MariaDBServer::to_json() const
return result;
}
bool MariaDBServer::can_replicate_from(MariaDBServer* master, string* error_out)
bool MariaDBServer::can_replicate_from(MariaDBServer* master, string* reason_out)
{
bool rval = false;
if (update_gtids())
mxb_assert(reason_out);
mxb_assert(is_usable()); // The server must be running.
bool can_replicate = false;
if (m_gtid_current_pos.empty())
{
if (m_gtid_current_pos.empty())
{
*error_out = string("'") + name() + "' does not have a valid 'gtid_current_pos'.";
}
else if (master->m_gtid_binlog_pos.empty())
{
*error_out = string("'") + master->name() + "' does not have a valid 'gtid_binlog_pos'.";
}
else
{
rval = m_gtid_current_pos.can_replicate_from(master->m_gtid_binlog_pos);
if (!rval)
{
*error_out = string("gtid_current_pos of '") + name() + "' ("
+ m_gtid_current_pos.to_string() + ") is incompatible with gtid_binlog_pos of '"
+ master->name() + "' (" + master->m_gtid_binlog_pos.to_string() + ").";
}
}
*reason_out = string_printf("%s does not have a valid gtid_current_pos.", name());
}
else if (master->m_gtid_binlog_pos.empty())
{
*reason_out = string_printf("%s does not have a valid gtid_binlog_pos.", master->name());
}
else
{
*error_out = string("Server '") + name() + "' could not be queried.";
can_replicate = m_gtid_current_pos.can_replicate_from(master->m_gtid_binlog_pos);
if (!can_replicate)
{
*reason_out = string_printf("gtid_current_pos of %s (%s) is incompatible with "
"gtid_binlog_pos of %s (%s).",
name(), m_gtid_current_pos.to_string().c_str(),
master->name(), master->m_gtid_binlog_pos.to_string().c_str());
}
}
return rval;
return can_replicate;
}
bool MariaDBServer::redirect_one_slave(const string& change_cmd)

10
server/modules/monitor/mariadbmon/mariadbserver.hh
View File

@ -261,14 +261,16 @@ public:
const SlaveStatus* slave_connection_status_host_port(const MariaDBServer* target) const;
/**
* Checks if this server can replicate from master. Only considers gtid:s and only detects obvious errors.
* The non-detected errors will mostly be detected once the slave tries to start replicating.
* Checks if this server can replicate from master. Only considers gtid:s and only detects obvious
* errors. The non-detected errors will mostly be detected once the slave tries to start replicating.
* Before calling this, update the gtid:s of the master so that the the gtid:s of the master are more
* recent than those of this server.
*
* @param master_info Master server
* @param error_out Details the reason for a negative result
* @param reason_out Details the reason for a negative result
* @return True if slave can replicate from master
*/
bool can_replicate_from(MariaDBServer* master, std::string* error_out);
bool can_replicate_from(MariaDBServer* master, std::string* reason_out);
/**
* Redirect one slave server to another master