The candidate selection code used default values that would cause reads
past buffers. The code could also dereference the end iterator which
causes undefined behavior.
Given the fact that there exist only three possible categories, the map
can be replaced with a static array that needs no memory
allocations. Making this array thread-local allows it to be reused which
places an upper limit on the number of memory allocations.
The servers with a zero weight would be always used over ones that have a
weight. This means that the behavior was inverted and caused the
mxs2054_hybrid_cluster test to fail in 2.3.
Also fixed a typo in the deprecation message.
By biasing the values of all counter type scores to positive integers, the
server weights are always taken into use.
This fixes the case when weights were ignored until all score base values
were larger than zero (the mxs922_server test).
The additions into the server.h header used C++ language which caused C
programs to fail to compile. Moved the implementation of the EMAverage
class into the private Server class in the server.hh header and exposed it
via functions in the server.h header. Also temporarily moved
almost_equal_server_scores into the public server.hh as there is no
service.hh header.
See script directory for method. The script to run in the top level
MaxScale directory is called maxscale-uncrustify.sh, which uses
another script, list-src, from the same directory (so you need to set
your PATH). The uncrustify version was 0.66.
The math becomes simpler when the weight is inverted, i.e. a simple multiplication
to get the (inverse) score. Inverse weights are normalized to the range [0..1] where a lower
number is a higher weight,
The enum select_criteria_t is used to provide a std::function that takes the backends
as vector (rather than the prior pairwise compares) and returns the best backend.
This commit refactors slave selection. The compare is still pair-wise but isolated into a small run_comparison() function. The function get_slave_candidate() is used when new connections are created, which I both moved and modified (had to move due to scoping), so diff is off.
The slave selection for routing: get_slave_backend() now has the filtering logic from old get_slave_backend() and compare_backends(), the latter of which is removed.
Backend functions mostly take shared_ptr<SRWBackend> in various forms (as is, const ref, in a container). Ideally the shared_ptr would be used only to where it is really needed, and either naked ptrs or references to RWBackend would be used. This refactor does not address that issue, but compounds it by using even deeper shared_ptr structures. Fixing that in a future commit.
The main piece of code, slave selection (backend_cmp_response_time), uses the available
method of pair-wise comparison of slaves. This will be changed to selection using all
available slaves, along with removal of hard coded values.
The configuration doesn't need to be contained in shared pointer as each
session holds its own version of it. This removes most of the overhead in
configuration reloading. The only thing that's left is any overhead added
by the use of thread-local storage.
By using a shared pointer instead of a plain object, we can replace the
router configuration without it affecting existing sessions. This is a
change that is required to enable runtime reconfiguration of
readwritesplit.
Allowing calls to select_connect_backend_servers even when all slaves are
connected solves the debug assertion in select_connect_backend_servers
that happens when the execution of a queued query causes a new connection
to be created.
The logging in the RWSplit::select_connect_backend_servers was quite
cumbersome and doing the logging when the connection is created makes for
a more information rich output (we know what servers were taken into use
and when).
Changed the log messages to use the names of the servers instead of the
address and port.
Most of the funtionality is now a member function of either the RWSplit or
RWSplitSession class. This removes the need to pass the router and session
parameters to all functions.
Moved the RWBackend class implementation into its own file. Made some of
the command type functions a part of the <maxscale/protocol/mysql.h>
header to make it reusable.
The `master_reconnection` parameter now controls both the reconnection of
the master server as well as the migration of the master server to another
server. Although these two cases appear to be different, the end result
from readwritesplit's point of view is the same and are thus controlled
with the same parameter.
The RWBackend class now resets its internal state when it is closed. This
allows readwritesplit to handle the case when a result was expected from
the master but the master died before the result was returned. The same
code should also handle slave connection failures mid-result, allowing
Backend reuse.
Added a test case that verifies the new functionality when combined with
`master_failure_mode=error_on_write`.
Moved session command execution into the Backend class itself as the
session commands are defined as a related part of it. This allows all
connections to execute session commands if some are available.
Removed explicit SERVER_REF usage in the readwritesplit connection
creation code and replaced it with SRWBackend. This allows the removal of
the get_root_master_backend function which duplicated the functionality in
get_root_master.
When LEAST_BEHIND_MASTER routing criteria was used, the info level logging
function would fall through to the default case. In debug builds, this
would trigger a debug assertion.
The connections for a router session can now be done without a constructed
router session. This simplifies the creation of new router session by
removing the need to handle memory allocations.
Readwritesplit router sessions are now created in the static `create`
function which handles the actual creation of the connections and
allocation of the session itself.
Providing helper functions for the commonly used parts of the server makes
code easier to read. It also removes any possibility for formatting
problems by moving the URI and name string handling inside the Backend
class.
Renamed variables to better represent the types of variables they
represent. Reordered some of the functions so that the functions don't
need to be declared before they are used.
Return values instead of objects. This removes the need to handle cases
where a reference to a "debug value" is returned.
Return SRWBackend values instead of passing output references. This
doubles as a false boolan return value when an empty reference is
returned.
As the session commands are always appended to the end of the list, the
name should reflect that action. For this reason, the function was renamed
to append_session_command.
Readwritesplit supports replacement of slave servers by storing all
executed session commands in a list. To make the copying of this list a
bit cleaner, an overload for a list of session commands was added. This
will allow relatively smooth addition of server replacement to all router
modules that use the Backend class.
This is the first step to taking the Backend class into use. It is now
used in rwsplit_select_backends.cc and readwritesplit.hh. The module is
not yet functional and doesn't even compile.
Added some helper functions to the Backend class to get easier access to
the server referenced by the SERVER_REF and to check the state of the
backend.
Creating duplicate connections using the Backend class allows the
connections and their handling to be tested at the same time that the old
system is in place. This should make it somewhat easier to grasp what
changes and where when the new implementation is taken into use.
