Export - CSV (All fields)
Export - CSV (Current fields)
Search
- MXS-5556Trailing parts of large session command are not routed correctly
- MXS-5555Improve smoothness of switchovers when used with readwritesplit
- MXS-5529Session commands with max_slave_connections=0 after switchover do not discard stale connections
- MXS-5527The "INSERT INTO...RETURNING" syntax breaks causal_reads
- MXS-5526Maxscale breaks connection complaining the 'retry_failed_reads' is disabled while it is enabled
- MXS-5507readwritesplit enables multi-statements regardless of the state of causal_reads
- MXS-5489Profile sync_transaction overhead
- MXS-5485Reads are not retried if connection creation fails due to sub-service failure
- MXS-5480disable_sescmd_history=true causes a use-after-free
- MXS-5470Add tests for sync_transaction
- MXS-5444Prototype implementation of synchronous writes
- MXS-5443Log message: Unknown prepared statement handler given to MaxScale
- MXS-5440causal_reads=local sends redundant MASTER_GTID_WAIT queries
- MXS-5415retry_failed_reads is not affected by delayed_retry_timeout
- MXS-5412SET tx_isolation cancels out strict_multi_stmt and strict_sp_calls
- MXS-5405Don't set session_track_system_variables if it already contains last_gtid
- MXS-5387Crash in MariaDBParser::Helper::get_query_info()
- MXS-5384maxscale version 24 auth_all_servers not picking up users on replicas
- MXS-5382Errors due to max_connections being exceeded are always fatal errors
- MXS-5377Debug assert if backend fails during multi-packet query
- MXS-5371Use @@gtid_binlog_pos in causal_reads=universal
- MXS-5359Transaction replay may deadlock with switchover
- MXS-5356Readwritesplit master_reconnection is incompatible with GET_LOCK
- MXS-5355Switchover may cause two transaction replays to be started
- MXS-5339Slow servers may cause OOM situations if prepared statements are used
- MXS-5316Debug assertion in sts_combinations in readwritesplit
- MXS-5269Unexpected response after failed session command
- MXS-5268Read-only error during read-write transaction should trigger transaction replay
- MXS-5259Retrying of reads on the current primary unnecessarily requires delayed_retry
- MXS-5258delayed_retry should not retry interrupted writes
- MXS-5231Connections to servers in maintenance are sometimes not discarded
- MXS-5223Log transaction contents on replay checksum mismatch
- MXS-5209Reads with max_slave_connections=0 after a switchover do not discard stale connections
- MXS-5206Readwritesplit does not drop connections to severely lagging servers
- MXS-5190dotnet EntityFrameworkCore generates insert queries that are getting routed to all nodes as session write
- MXS-5176Simple failover takes long time, new connections get "Internal error: Session creation failed" for over 10 secs
- MXS-5150client gets disconnected when using Maxscale
- MXS-5142Workaround Maxscale Not Distributing the load on all Galera Cluster nodes equally
- MXS-5127DEALLOCATE PREPARE is not routed to all nodes
- MXS-5112Modularize readwritesplit features
- MXS-5103query_classifier_cache_size=0 with short connections appears to leak memory
- MXS-5086Read with max_slave_connections=0 fails with master_failure_mode=fail_on_write
- MXS-5085max_slave_connections=0 may create slave connections after a switchover
- MXS-5041Don't replay autocommit statements with transaction_replay
- MXS-5023kill user and transaction_replay don't play well together in Galera cluster
- MXS-5007Top-level service reconnection may cause a use-after-free
- MXS-4978Read-only transactions are incorrectly tracked
- MXS-4961Result of KILL CONNECTION_ID() does not propagate all the way to the client
- MXS-4956Session commands ignore delayed_retry_timeout
- MXS-4943delayed_retry timeout errors do not have enough information