The slave nodes par od the cluster are sorted by wsrep_local_index or
server priority.
The sorted list od nodes is then used in ‘wsrep_sst_donor’ variable.
Note: candidate master servers are always at the end of the list
The embedded parser is not capable of parsing preparable statements
containing ?:s as ordinary statements, so they have to be replaced
with something else. Replacing them with a dummy variable cannot be
done as that will affect the type of the statement. Replacing with
0 seems to be a good compromise.
Since the whole preparable statement will be available, it is
superfluous to provide a function using which the operation of
the preparable statement can be obtained.
This function will return the preparable statement of a PREPARE
statement as a COM_QUERY GWBUF. That is, once obtained, the normal
query classifier functions can be used for obtaining information
about the preparable statement itself.
- Only types of fixed size used in API interface
- The actual function return value specifies whether the parsing
process succeeded, while "logical" return values are returned
as out arguments.
The wrapper function currently ignores the function return value.
Galeramon queries all slave nodes for SHOW VARIABLES LIKE
'wsrep_node_name'
The server list, currently not sorted, is used in SET GLOBAL
wsrep_sst_donor = “list”. The statement is sent to all slave nodes
MXS-391. The user hosts in the SQL backends can now contain wildcard
addresses (e.g.basicuser@%.com or someuser@myhost_.org). Authenticating
these types of users is rather heavy since it requires looking
up the client host name.
All modules now declare a name for the module. This is name is added as a
prefix to all messages logged by a module. The prefix should help
determine which part of the system logs a message.
The external config.h header defined functions and structures that are
intended to be used by modules. The internal header contains functions
that are used only by the MaxScale core.
The header is divided into two parts, an external and an internal one. The
actual splitting is done in a later commit and this commit only prepares
the header for the split.
The filter converts bulk inserts inside transactions into LOAD DATA LOCAL
INFILE statements. This allows multiple INSERT statements to be streamed
in a more efficient manner.
The module should be in the experimental package as it is still at an
experimental stage.
The module commands now support an optional flag for arguments that when
enabled checks that the argument module name matches the registered domain
name. This can be used to enforce argument type validity for arguments
that are given to modules that expect objects of a certain type.
For example, this is used by the cache and dbfwfilter to prevent valid
filters but of the wrong type being given as arguments.
Tee no longer uses the dummy_filterdef FILTER_DEF member, so
it can be removed. That means that filter_free() can be removed
from the public filter.h include.
Better to use types whose size is unquestionably unambiguous
in plugin interfaces.
As on current platforms sizeof(int32_t) == sizeof(int) plugins
need not immediately be changed to return int32_t:s instead of
int:s.
Added documentation for dbfwfilter and mentioned the new directory in the
release notes. The configuration guide also gives an example of how the
path parameters are resolved.
The filters don't need to process relative paths as the core handles
that. All path parameters are interpreted as relative to the module
configuration directory, /etc/maxscale.modules.d/.
