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

MXS-359: Rename parameter to master_reconnection

Browse Source 
Changing the parameter name to `master_reconnection` allows similar
reconnection functionality to be combined under one parameter.
This commit is contained in:
Markus Mäkelä
2018-03-28 09:49:30 +03:00
parent 3f803ab3e8
commit 481db8acbc
7 changed files with 11 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
Documentation/Routers/ReadWriteSplit.md
View File

@ -105,7 +105,7 @@ connections alive even if they are not used. This is a common problem if the
backend servers have a low _wait_timeout_ value and the client connections live
for a long time.
### `allow_master_change`
### `master_reconnection`
Allow the master server to change mid-session. This feature was introduced in
MaxScale 2.3.0 and is disabled by default.
@ -114,7 +114,7 @@ When a readwritesplit session starts, it will pick a master server as the
current master server of that session. By default, when this master server
changes mid-session, the connection will be closed.
If the `allow_master_change` parameter is enabled, the master server is allowed
If the `master_reconnection` parameter is enabled, the master server is allowed
to change as long as the session meets the following criteria:
* The session is already connected to the slave that was chosen to be the new master
@ -123,7 +123,7 @@ to change as long as the session meets the following criteria:
* No `LOAD DATA LOCAL INFILE` is in progress
* There are no queries being actively routed to the old master
When `allow_master_change` is enabled in conjunction with either
When `master_reconnection` is enabled in conjunction with either
`master_failure_mode=fail_on_write` or `master_failure_mode=error_on_write`, the
session can recover from the loss of a master server. This means that when a
session starts without a master server and later a slave server that it is
@ -295,7 +295,7 @@ long as slave servers are available.
**Note:** If _master_failure_mode_ is set to _error_on_write_ and the connection
to the master is lost, by default, clients will not be able to execute write
queries without reconnecting to MariaDB MaxScale once a new master is
available. If [`allow_master_change`](#allow_master_change) is enabled, the
available. If [`master_reconnection`](#master_reconnection) is enabled, the
session can recover if one of the slaves is promoted as the master.
### `retry_failed_reads`

2
maxscale-system-test/cnf/maxscale.cnf.template.mxs359_error_on_write
View File

@ -16,7 +16,7 @@ router=readwritesplit
servers=server1,server2,server3,server4
user=maxskysql
passwd=skysql
allow_master_change=true
master_reconnection=true
master_failure_mode=error_on_write
[RW-Split-Listener]

2
maxscale-system-test/cnf/maxscale.cnf.template.mxs359_master_switch
View File

@ -16,7 +16,7 @@ router=readwritesplit
servers=server1,server2,server3,server4
user=maxskysql
passwd=skysql
allow_master_change=true
master_reconnection=true
[RW-Split-Listener]
type=listener

2
maxscale-system-test/cnf/maxscale.cnf.template.mxs359_read_only
View File

@ -16,7 +16,7 @@ router=readwritesplit
servers=server1,server2,server3,server4
user=maxskysql
passwd=skysql
allow_master_change=true
master_reconnection=true
master_failure_mode=fail_on_write
[RW-Split-Listener]

2
server/modules/routing/readwritesplit/readwritesplit.cc
View File

@ -1495,7 +1495,7 @@ MXS_MODULE *MXS_CREATE_MODULE()
{"connection_keepalive", MXS_MODULE_PARAM_COUNT, "0"},
{"enable_causal_read", MXS_MODULE_PARAM_BOOL, "false"},
{"causal_read_timeout", MXS_MODULE_PARAM_STRING, "0"},
{"allow_master_change", MXS_MODULE_PARAM_BOOL, "false"},
{"master_reconnection", MXS_MODULE_PARAM_BOOL, "false"},
{MXS_END_MODULE_PARAMS}
}
};

4
server/modules/routing/readwritesplit/readwritesplit.hh
View File

@ -175,7 +175,7 @@ struct Config
max_slave_connections(0),
enable_causal_read(config_get_bool(params, "enable_causal_read")),
causal_read_timeout(config_get_string(params, "causal_read_timeout")),
allow_master_change(config_get_bool(params, "allow_master_change"))
master_reconnection(config_get_bool(params, "master_reconnection"))
{
if (enable_causal_read)
{
@ -202,7 +202,7 @@ struct Config
int max_slave_connections; /**< Maximum number of slaves for each connection*/
bool enable_causal_read; /**< Enable causual read */
std::string causal_read_timeout; /**< Timetout, second parameter of function master_wait_gtid */
bool allow_master_change; /**< Allow changes in master server */
bool master_reconnection; /**< Allow changes in master server */
};
/**

2
server/modules/routing/readwritesplit/rwsplit_route_stmt.cc
View File

@ -690,7 +690,7 @@ static void log_master_routing_failure(RWSplitSession *rses, bool found,
bool should_replace_master(RWSplitSession *rses, SRWBackend& target)
{
return rses->rses_config.allow_master_change &&
return rses->rses_config.master_reconnection &&
// We have a target server and it's not the current master
target && target != rses->current_master &&
// We are not inside a transaction (also checks for autocommit=1)