* Actually using the replica recovery settings for cluster
* A bit of doc on DBMS_CONNECTION_POOL_WITH_FAILOVER_MAX_ERROR_COUNT
* StorageDistributedDirectoryMonitor using settings for ConnectionPoolWithFailover
* Using SettingSeconds instead of SettingUInt64 for replica_error_decrease_period
* Added a limit on how many errors can replica accumulate
* Decreased default error halving time to 60 seconds
* Made both configurable via settings
* Showing errors count and estimated recovery time for each replica in system.clusters
* Fix build
* cmake: fix cpuinfo
* Fix includes after processors merge
Conflicts:
dbms/src/Processors/Formats/Impl/CapnProtoRowInputFormat.cpp
dbms/src/Processors/Formats/Impl/ParquetBlockOutputFormat.cpp
dbms/src/Processors/Formats/Impl/ProtobufRowInputFormat.cpp
dbms/src/Processors/Formats/Impl/ProtobufRowOutputFormat.cpp
* Fix build in gcc8
* fix test link
* fix test link
* Fix test link
* link fix
* Fix includes after processors merge 2
Conflicts:
dbms/src/Processors/Formats/Impl/ParquetBlockInputFormat.cpp
* Fix includes after processors merge 3
* link fix
* Fix likely/unlikely conflict with cython
* Fix conflict with protobuf/stubs/atomicops.h
* remove unlikely.h
* Fix macos build (do not use timer_t)
ThreadSanitizer reported that a data race occurs when we attach to a
database and load its tables concurrently. Remove the custom code that
waits for all tables to load, and replace it with a local thread pool
and its wait() method.
This changes observable behavior: before, we would load the tables
sequentially when loading the system database, or performing an ATTACH
DATABASE query. Now we always load tables of ordinary databases in
parallel.
and making the setting to be in seconds instead of usec
* Updated LIVE VIEW table to keep version consistent as long as the server is up
* Removed calculation of hash in writeIntoLiveView method as it is recalculated
in the LiveViewBlockOutputStream
* Changed to using values from the local context for live_view_heartbeat_interval and
temporary_live_view_timeout instead of the global context
* Updated LIVE VIEW tests to match the changes
* CLICKHOUSE-3771: compatible version transfer
* Fix
* Fixed compatibility
* fix
* Auto version update to [1.1.54401] [54398]
* fix auto revision up
* fix
* comments
* clean
* fix
* Send/recieve patch if we can
* fix
* Auto version update to [18.2.0] [54397]
* Client version in query_log
* Update TCPHandler.cpp
* Update Connection.cpp
* Update Defines.h
