Export - CSV (All fields)
Export - CSV (Current fields)
MXS-5316
Debug assertion in sts_combinations in readwritesplit
Error msg reports write query failure for read queries
Unexpected response after failed session commandMXS-3438
A question about maxscale weights mechanism
Debug assert with COM_STMT_EXECUTE
Bias value for open connections is too large
Document that readwritesplit may replay about-to-rollback transactions
Crash on transaction replay if log_info is on and session starts with no master
Server weights don't work with LEAST_BEHIND_MASTER
COM_PING is sent to all servers
Top-level service reconnection may cause a use-after-free
Errors due to max_connections being exceeded are always fatal errors
Transaction replay may deadlock with switchover
max_slave_connections=0 may create slave connections after a switchover
SET tx_isolation cancels out strict_multi_stmt and strict_sp_callsMXS-4669
Make transaction_replay work better in read-only scenarios
causal_reads=local is serialized as causal_reads=true
Temporary tables are lost with master_reconnection
Read-only transaction sometimes loses statements with causal_reads=universal and transaction_replay=true
slave_selection_criteria should accept lowercase version of the values
Readwritesplit does not detect unrecoverable situations
Log message: Unknown prepared statement handler given to MaxScale
kill user and transaction_replay don't play well together in Galera cluster
Debug assertion with default config and transaction_replay=true
Execution of PS fails when strict_sp_calls is enabled
Active operation count is wrong after failed causal read
Transaction replay doesn't reset transaction on implicit commit
Connections to servers in maintenance are sometimes not discarded
Readwritesplit does not drop connections to severely lagging servers
MaxScale tries to query @@session_track_system_variables despite causal_reads=none
Deferred execution of session commands
Improve speed by not using kernel clock where possible
Load not balanced when pressure is low
Readwritesplit logs wrong warning when replication lag is not available
Adaptive blog
Monitor status when running a 2 node + arbiter setup will break.
mxs1507_trx_replay: debug assert in routeQuery
Rapid reconnection on server failure
SHOW DATABASES is routed to the master
New query statistics
New adaptive routing algorithm
Get_lock is not consistent.
Connections to servers in maintenance aren't closed
hint "maxscale route to slave" does not work
MYSQL_session never refresh with the changing of database in a RWSplitSession
Maxscale Workload Statistics By Server By readwrite available via REST API
Changing masters does not close readwritesplit session immediately
[readwritesplit] Unexpected internal state
debug assert at rwsplit_route_stmt.cc:469 failed: btype != BE_MASTER
Readwritesplit slave receiving alter requests