Shulou(Shulou.com)06/01 Report--

It began on April 1, 2017-April Fool's Day.

1.1 MySQL 5.7 New Featur

This section introduces the new, obsolete, and deleted features in the new version of MySQL 5.7. More information can be found in Section 1.5 Section 1.5, "Server and Status Variables and Options Added, Deprecated, or Removed in MySQL 5.7". .

What's New in MySQL 5.7

MySQL 5.7 obsolete function

MySQL 5.7Delete function

What's New in MySQL 5.7

The following are the new feature options for MySQL 5.7:

Security is enhanced. :

This version requires that the plugin field in mysql.user is not empty and that accounts is not allowed to be null. To view server upgrades, see section 2.11.1.1 Section 2.11.1.1, "Changes Affecting Upgrades to MySQL 5.7". Mysql suggested that DBA use mysql_native_password instead of mysql_old_password. Because mysql_old_password has been deleted and is no longer supported. For account upgrade, please see Section 7.5.1.3 Section 7.5.1.3, "Migrating Away from Pre-4.1 Password Hashing and the mysql_old_password Plugin".

MySQL now allows database administrators to create an account that expires automatically using policies. Any expired account connected to the database server must change its password. In section 7.3.6 Section 7.3.6, "Password Expiration Policy". There is a detailed introduction.

Administrators can lock or unlock to better control who can log in, see section 7.3.10 Section 7.3.10, "User Account Locking". There is a detailed introduction.

In order to support secure connections more simply, MySQL server can use Openssl to automatically generate missing RSA authentication files at startup. In chapter 7.4.6.1, Section 7.4.6.1, "Creating SSL and RSA Certificates and Keys using MySQL". There is a detailed introduction.

All server, whether Openssl or yaSSL, do not explicitly configure SSL. When the server starts, it is found that the SSL file in the data directory will automatically enable SSL in the cabin. In chapter 7.4.4, Section 7.4.4, "Configuring MySQL to Use Secure Connections". There is a detailed introduction.

In addition, MySQL also configures a feature pack that contains mysql_ssl_rsa_setup, which allows you to manually create SSL and RSA key authentication files in Section 5.4.5 Section 5.4.5, "mysql_ssl_rsa_setup-Create SSL/RSA Files". There is a detailed introduction.

MySQL deployment initialization uses mysqld-initialize as the security default. The items listed below are mysql as the default project when deployed:

When Mysql initializes, it only creates a root account 'root'@'localhost', and automatically generates a random password for this account, and sets the password to expire. The MySQL administrator must log in with this password and set the new password to be given (the server writes the random password to the error log file)

Anonymous accounts are not created during initialization.

The test library is not created during initialization.

See section 2.10.1.1 Section 2.10.1.1, "Initializing the Data Directory Manually Using mysqld". More details are available.

SQL mode has made changes. STRICT_TRANS_TABLES is now changed to the default.

ONLY_FULL_GROUP_BY SQL mode is more accurate, and queries rejected in previous versions are no longer rejected. Therefore, the mode is turned on by default and now rejects only indeterminate queries that are not guaranteed to be unique in the group. ERROR_FOR_DIVISION_BY_ZERO, NO_ZERO_DATE, and NO_ZERO_IN_DATE SQL modes are now deprecated and enabled by default. In future releases, it is planned to add them to strict SQL mode and delete their display. It is described in detail in this section of SQL Mode Changes in MySQL 5.7.

The default SQL mode of the system that can be changed are: ONLY_FULL_GROUP_BY, STRICT_TRANS_TABLES, NO_ZERO_IN_DATE, NO_ZERO_DATE, ERROR_FOR_DIVISION_BY_ZERO, NO_AUTO_CREATE_USER, and NO_ENGINE_SUBSTITUTION.

Online ALTER TABLE modifies the table online. The alter table command can now rename the index's name ALTER TABLE now supports a RENAME INDEX clause that renames an index. Exe via rename index. This change is the source modification mode (in-place) that does not require a table copy operation and is valid for all engines. This change is an internal modification (in-place) that does not require a copy table operation and is valid for all engines. In chapter 14.1.8, Section 14.1.8, "ALTER TABLE Syntax". There is a detailed introduction.

Ngram and MeCab full-text analysis plug-in. MySQL provides a built-in full-text ngram analysis plug-in to support Chinese, Japanese, Korean and an installable MeCab full-text analysis Japanese plug-in. In chapters 13.9.8 Section 13.9.8, "ngram Full-Text Parser", and 13.9.9 chapters Section 13.9.9, "MeCab Full-Text Parser Plugin". Described in detail by.

InnoDB enhancement. :

The size of the VARCHAR field can be modified by ALTER TABLE, command, and in-place, for example:

ALTER TABLE T1 ALGORITHM=INPLACE, CHANGE COLUMN C1 C1 VARCHAR (255)

This is true as long as the number of length bytes required by a VARCHAR column remains the same. This can be achieved as long as the number of bytes occupied by the field varchar is the same as the previous one after the field is modified. For example, for VARCHAR values between 0 and 255, only one bytes. The value of VARCHAR is 256 bytes or greater than 256 requires two bytes. In this way, only changes between 0 and 255are supported through in-place ALTER TABLE, or between 256 and greater than 256. in-place alter table does not support changing varchar values from less than 256 to values greater than 256. Because in this case, the stored bytes will change from 1 byte to two bytes. Can only be modified by algorithm=copy. For example, changing the value of varchar to 256in-place alter would will return an error.

ALTER TABLE T1 ALGORITHM=INPLACE, CHANGE COLUMN C1 C1 VARCHAR

ERROR 0A000: ALGORITHM=INPLACE is not supported. Reason: Cannot change

Column type INPLACE. Try ALGORITHM=COPY.

Using IN-PLACE alter table to reduce the size of varchar is not supported. Reducing field size requires a copy table operation (ALGORITHM=COPY).

Enhancements have been made to DLL CREATE TABLE, DROP TABLE, TRUNCATE TABLE, and and ALTER TABLE of INNODB engine temporary tables.

The meta information of the InnoDB temporary table is not currently stored in the InnoDB system table. Instead of a new table INNODB_TEMP_TABLE_INFO, you can query the temporary table snapshot. This table records the temporary table information in the innodb instance created by all users and the system itself. When used for the first time. This table is created when the select statement is run for the first time

InnoDB now supports spatial data types. In previous versions, InnoDB could only store spatial data types as binary BLOB. BLOB is the basic storage type, but the spatial data type is now used as a new InnoDB engine internal data type, DATA_GEOMETRY.

All uncompressed INNODB engine temporary tables now provide their own tablespaces. Whenever a new tablespace in server startup is recreated, the directory is specified by default in DATADIR. A new parameter innodb_temp_data_file_path allows the user to customize the storage address of the temporary table.

The innochecksum feature has now been enhanced and several new parameters have been provided to extend its capabilities. Section 5.6.1 Section 5.6.1, "innochecksum-Offline InnoDB File Checksum Utility". More details are available.

Ordinary or compressed temporary tables and related objects are now stored in temporary tablespaces with a new non-redo undo log. Section 15.4.12.1 Section 15.4.12.1, "InnoDB Temporary Table Undo Logs" has more information.

InnoDB buffer pool dump and load have been enhanced. With innodb_buffer_pool_dump_pct, the parameter allows you to specify a percentage of the data in buffer _ pool, while there are other active IO operations in the background of innodb. INNODB attempts to use the innodb_io_capacity parameter to limit the amount of data loaded from buffer pool per second.

InnoDB supports full-text indexing. Check the chapters Full-Text Parser Plugins and Section 27.2.4.4, "Writing Full-Text Parser Plugins".

The use of InnoDB to clean up dirty data in buffer pool can use innodb_page_cleaner to specify multiple cleanup threads. This parameter defaults to 1. 0.

MySQL can use online DDL (ALGORITHM=INPLACE) periodically and step by step for the options listed below:

OPTIMIZE TABLE

ALTER TABLE... FORCE

ALTER TABLE... ENGINE=INNODB (when run on an InnoDB table)

Online DDL can reduce time and support concurrency, which can effectively reduce downtime section 15.13.1 Section 15.13.1, "Overview of Online DDL" is described in detail.

Linux's Fusion-io Non-Volatile Memory (NVM) file system provides atomic (atomic) write extensibility, making InnoDB doublewrite buffer redundant. Automatically disable InnoDB doublewrite buffer for system tablespaces (ibdata files) on fusion-io devices that support atomic writing.

InnoDB supports t Transportable Tablespace for partitioned tables and individual partitions. This enhancement simplifies section 15.7.6 Section 15.7.6, which is detailed in "Copying File-Per-Table Tablespaces to Another Server", to back up migration partition tables between different server.

The parameter innodb_buffer_pool_size can now be dynamically modified without downtime. This change is done by chunks copying the page into the new memory. . The size of the Chunk is specified by the parameter innodb_buffer_pool_chunk_size. You can monitor the progress by monitoring the view Innodb_buffer_pool_resize_status. Check out the chapter Configuring InnoDB Buffer Pool Size Online for more information.

Multi-threaded page cleaner support (innodb_page_cleaners) is lengthens shutdown and recovery phases time.

InnoDB supports spatial data indexing including use of ALTER TABLE. ALGORITHM=INPLACE for online operations (ADD SPATIAL INDEX).

InnoDB performs a chunk of loading when creating or creating an indexes. This way of creating an index is often called "sorted index build". This enhancement improves the efficiency of index creation and is also effective for full-text indexing. An idle global variable, innodb_fill_factor, defines the ratio of data stored on each page during sorted index build. The space reserved for future index growth section 15.8.2 Section 15.8.12, "Sorted Index Builds" is described in detail.

A new logging type (MLOG_FILE_NAME) is used to record tablespace changes since the last checkpoint. This enhancement makes it faster to calculate the redolog scale at the time of recovery. Section 15.18.2 Section 15.18.2, "Tablespace Discovery During Crash Recovery" is described in detail.

This enhancement changes the format of redo log and needs to be turned off.

You can empty the undo logs that you keep in undo tablespace. It is controlled by parameter innodb_undo_log_truncate. Section 15.7.8 Section 15.7.8, "Truncating Undo Logs That Reside in Undo Tablespaces" is described in detail.

InnoDB supports native partitioning. Previously, InnoDB relied on ha_partition maintenance to create a hander for each partition. For native partitions, a partitioned InnoDB table uses a separate partition-aware handler object. This enhancement reduces the amount of memory used by partitioned tables.

For MySQL 5. 7. 9 mysql_upgrade to find and try to upgrade the previous use of ha_partition handler. And in MySQL 5.7.9 and later, you can issue commands directly through client to upgrade ALTER TABLE. UPGRADE PARTITIONING.

InnoDB supports the creation of a common tablespace syntax as follows: CREATE TABLESPACE.

CREATE TABLESPACE tablespace_name

ADD DATAFILE 'file_name.ibd'

[FILE_BLOCK_SIZE = n]

General-purpose tablespaces can be created outside the datadir, can store multiple tables and support the all row format of tables.

Use CREATE TABLE tbl_name to add tables to common tablespaces. Syntax for TABLESPACE [=] tablespace_name or ALTER TABLE tbl_name TABLESPACE [=] tablespace_name.

Section 15.7.9 Section 15.7.9, "InnoDB General Tablespaces" has more details.

DYNAMIC replaces COMPACT as the row format of the implicit INNODB table. Use the parameter innodb_default_row_format to specify the default InnoDB row format. Section 15.11.2 Section 15.11.2, "Specifying the Row Format for a Table" has more information.

In MySQL version 5.7.11, InnoDB supports data-at-rest encryption for each file-per-table tablespace. Specifying the ENCRYPTION parameter when creating or modifying the innodb table will enable encryption. This function depends on InnoDB tablespace encryption and on keyring plug-ins. Section 5.7.4 Section 7.5.4, "The MySQL Keyring", and 15.7.10 Section 15.7.10, "InnoDB Tablespace Encryption" has more detailed information.

JSON support. The native JSON format is supported since version 5.7.8. Josn data is no longer stored as a string, but in an internal binary format that allows document elements to be read quickly. Json documents stored in json columns are automatically checked when they are updated or inserted. If the format does not match, an exception is thrown. Json is no different from ordinary columns in creation and can do =,!,!, and;, and most of the formal operation chapters, such as Comparison and Ordering of JSON Values, are described more.

MySQL 5.7.8 also introduces a number of functions to support the manipulation of json data:

Create JSON data: JSON_ARRAY (), JSON_MERGE (), JSON_OBJECT (). See section Section 13.16.2, "Functions That Create JSON Values" for more.

Search for JSON data: JSON_CONTAINS (), JSON_CONTAINS_PATH (), JSON_EXTRACT (), JSON_KEYS (), and JSON_SEARCH (). See section Section 13.16.3, "Functions That Search JSON Values" for more.

Modify JSON data: JSON_APPEND (), JSON_ARRAY_APPEND (), JSON_ARRAY_INSERT (), JSON_INSERT (), JSON_QUOTE (), JSON_REMOVE (), JSON_REPLACE (), JSON_SET (), and JSON_UNQUOTE (). See section Section 13.16.4, "Functions That Modify JSON Values" for more.

View JSON data information: JSON_DEPTH (), JSON_LENGTH (), JSON_TYPE (), and JSON_VALID (). See section Section 13.16.5, "Functions That Return JSON Value Attributes" for more.

In version 5.7.9 and later, you can use column- > path as the fast JSON_EXTRACT (column, path). In this way, it is used in sql as a column similar to aliases, and can also be used in WHERE, ORDER BY, and GROUP BY conditions. These include SELECT, UPDATE, DELETE, CREATE TABLE, and other SQL statements. The one on the left must be the json column, and the one on the right is the expression that references josn path, which is the value returned by the json document.

Check out section 13.16.3 Section 13.16.3, "Functions That Search JSON Values" for more information about-> and JSON_EXTRACT (). Information, chapter Searching and Modifying JSON Values. There is more information about MySQL 5.7support for json. For the json index, please see this section Indexing a Generated Column to Provide a JSON Column Index.

System and status variables. System and state variable information can now be displayed in the performance schmema table. The table INFORMATION_SCHEMA contains those variables. This also involves SHOW VARIABLES and SHOW STATUS statements. The variable show_compatibility_56 system variable controls the output of the query status. Section 6.1.5 Section 6.1.5, "Server System Variables". More details are available.

Note

The variable show_compatibility_56 defaults to OFF. The program needs to set the value to ON when the environment is version 5.6. Section 24.18 Section 24.18, "Migrating to Performance Schema System and Status Variable Tables" has more details.

Sys schema. The schema MySQL branch of the system now contains sys schema, which provides DBA and developers with access to database performance information. Sys schema contains objects that can be used to analyze diagnostic databases. Chapter 25 Chapter 25, MySQL sys Schema. More details are available

Condition handling condition processing. MySQL now supports stacked diagnostics areas. Section 14.6.7.3 Section 14.6.7.3, "GET DIAGNOSTICS Syntax", and Section 14.6.7.7, "The MySQL Diagnostics Area". More details are available.

Optimizer. The optimizer has been enhanced and the following has been added:

EXPLAIN can get other execution plans from you:

EXPLAIN [options] FOR CONNECTION connection_id

Section 9.8.4 Section 9.8.4, "Obtaining Execution Plan Information for a Named Connection". There is more information.

Now you can add hinst to a single sql as an optimization way to control the execution plan of the statement, which can be achieved by setting optimizer_switch. It can also be used when obtaining the execution plan, chapter 9.9.3Section 9.9.3, "Optimizer Hints". More details are available.

Triggers. Trigger, a previous table can only have at most one similar trigger (INSERT, UPDATE, DELETE) action time (BEFORE, AFTER). This limitation has been broken and multiple triggers are allowed to exist in section 22.3 Section 22.3, "Using Triggers". More details are available.

Logging log. :

In previous versions, on unix and unix-like systems, MySQL sent server error logs to syslog through mysqld_safe. Now server supports native syslog and extends to windows systems. Section 6.4.2 Section 6.4.2, "The Error Log". More details are available.

The mysql client now has a-- syslog option that can interact with the system to write to syslog. Similar to ("IDENTIFIED:PASSWORD"), other things that will not be written to syslog can also be controlled by the parameter-histignore. 5.5.1.3 Chapter Section 5.5.1.3, "mysql Logging". More details are available.

Generated Columns generates columns. MySQL supports column generation in CREATE TABLE and ALTER TABLE. The value of the generated column can be generated through an expression. The generated column can be a virtual column or a stred section 14.1.18.7 Section 14.1.18.7, "CREATE TABLE and Generated Columns". More details are available.

Mysql client mysql client. Earlier, Control+C could terminate the current statement or exit the client. Now Control+C terminates the current statement or other input but does not exit the client program.

Database name rewriting with mysqlbinlog. Starting from 5.7.1, for binlog in row format, when using the tool mysqlbinlog, you can modify the dbname with the parameter-rewrite-db, using the format:-- rewrite-db='dboldname- > dbnewname'

The format of the parameter is-- rewrite-db='dboldname- > dbnewname'. You can write multiple rules to specify multiple times

HANDLER with partitioned tables partition table and handler. The user's partition table can also use the HANDLER statement. For more information, see Section 21.2 (see Section 21.2, "Partitioning Types").

Index condition pushdown support for partitioned tables partitioned tables support index push-down ICP. Partitioned table queries using the INNODB and MyISAM engines can use the index condition pushdown (ICP) optimization that has been introduced in 5. 6. Section 9.2.1.6 Section 9.2.1.6, "Index Condition Pushdown Optimization", for more details.

WITHOUT VALIDATION support for ALTER TABLE... EXCHANGE PARTITION partition exchange does not require verification. The MySQL5.7.5 version begins, ALTER TABLE. The EXCHANGE PARTITION command contains a {WITH | WITHOUT} VALIDATION option. When WITHOUT VALIDATION is specified, ALTER TABLE... EXCHANGE PARTITION partition swapping allows the database administrator to assume the boundaries of the partition. If an error occurs when the boundary is exceeded, it will work normally if it is not exceeded. WITH VALIDATION. Is the default value and does not need to be explicitly set. Section 21.3.3 Section 21.3.3, "Exchanging Partitions and Subpartitions with Tables". There are more details.

Master dump thread improvements master dump thread enhancement. The refactoring of the master dump site reduces locks and improves the throughput of maste. In the previous version 5.7.2, a lock was acquired when the dump site read the binary log again. In later versions, the lock was acquired only when the last successfully written event was read. This means that multithreaded dumps are already supported and that the dump thread can also read when the client writes binaries. .

Globalization improvements globalization support has been enhanced. MySQL 5.7.4 added character set gb18030. Section 11.1 Section 11.1, "Character Set Support". There are more introductions.

Changing the replication master without STOP SLAVE replication schema modification master does not need to stop slave. MySQL 5.7.4 and later, stop slave commands are no longer required before CHANGE MASTER TO. Now it is slave sql thread and slave I thread O thread that decide whether or not slave needs to be stopped. I:

If SQL thread stops, you can execute CHANGE MASTER TO using any combination of RELAY_LOG_FILE, RELAY_LOG_POS, and MASTER_DELAY options, even though slave I _ thread is still running.

If thread O thread stops, you can execute CHANGE MASTER TO and use any option except RELAY_LOG_FILE, RELAY_LOG_POS, and MASTER_DELAY. Even though SQL thread is still running. These three options should not be used when Imax O thread is running.

SQL thread and the I SQL thread and the O thread must stop when parameter CHANGE MASTER TO... MASTER_AUTO_POSITION = 1. Hour

You can use SHOW SLAVE STATUS. Check the status of sql thread and Imax O thread.

If you use statement-based replication and temporary tables, executing CHANGE MASTER TO after executing stop slave will leave the temporary tables in slave. As an enhancement to this part, a warning warning is now thrown when stop slave is executed. When this replication uses Slave_open_temp_tables and has a value of 0.

Section 14.4.2.1 Section 14.4.2.1, "CHANGE MASTER TO Syntax", and section 18.3.7 Section 18.3.7, "Switching Masters During Failover". I have more information.

Test suite. MySQL test suite uses InnoDB as the default storage engine.

Multi-source replication is now possible. Now supports multi-master replication MySQL multi-source replication supports multiple master to one slave replication. The topology of MySQL multi-source replication can back up multiple server to a single server. To back up multiple servers to a single server, from multi-server sharding and data collection to a single server chapter 18.1.4Section 18.1.4, "MySQL Multi-Source Replication". There is a detailed introduction.

As part of multi-source replication, replication pipes have been added. Replication channels runs slave to open multiple connections to multiple connection sources at the same time. Section 18.2.3 Section 18.2.3, "Replication Channels". There is a detailed introduction.

Group Replication Performance Schema tables Group Replication performance depends on the chart. MySQL 5.7 adds a large number of performance charts for replication groups and channels in Performance Schema:

Replication_applier_configuration

Replication_applier_status

Replication_applier_status_by_coordinator

Replication_applier_status_by_worker

Replication_connection_configuration

Replication_connection_status

Replication_group_members

Replication_group_member_stats

MySQL 5.7.2 began to introduce the above tables, and it is also important to note that replication_group_members and replication_group_member_stats, two tables are the details provided by MySQL 5.7.6.t see Section 24.10.11, "Performance Schema Replication Tables".

Group Replication SQL. Group Replication SQL. The following parameters were introduced in MySQL 5.7.6 to control group replication:

START GROUP_REPLICATION

STOP GROUP_REPLICATION

For more information, see Section 14.4.3, "SQL Statements for Controlling Group Replication".

Obsolete features in MySQL 5.7

The following features of The following features are deprecated in MySQL 5.7 and may be or will be removed in a future series are obsolete in Mysql 5.7 and may be removed in future releases. Where alternatives are shown, applications should be updated to use them.

ERROR_FOR_DIVISION_BY_ZERO, NO_ZERO_DATE, and NO_ZERO_IN_DATE SQL modes have been deprecated and enabled by default. The future plan is to integrate them into strict SQL mode and remove them completely.

To prevent statements from having error patterns ERROR_FOR_DIVISION_BY_ZERO, NO_ZERO_DATE, and NO_ZERO_IN_DATE are still recommended, but will be deleted in the new version.

User management has changed:

You can no longer use grant to create a user-replaced CREATE USER. This makes the associated NO_AUTO_CREATE_USER SQL mode invalid so it is discarded.

Using GRANT to modify account properties other than privilege assignments. This includes authentication, SSL, and resource-limit properties. Instead, establish such properties at account-creation time with CREATE USER or modify them afterward with ALTER USER.

Use GRANT to modify user account properties instead of permission assignments, including authentication, SSL, and resource properties. Instead of dealing with attributes when creating users and alter user.

IDENTIFIED BY PASSWORD 'hash_string' syntax for CREATE USER and GRANT. Instead, use IDENTIFIED WITH auth_plugin AS 'hash_string' for CREATE USER and ALTER USER, where the' hash_string' value is in a format compatible with the named plugin.

The IDENTIFIED BY PASSWORD 'hash_string' syntax used by CREATE USER and GRANT is replaced by the syntax IDENTIFIED WITH auth_plugin AS' hash_string' in CREATE USER and ALTER USER

The PASSWORD () function is deprecated and should be avoided in any context. Thus, SET PASSWORD... = PASSWORD ('auth_string') syntax is also deprecated. SET PASSWORD... = 'auth_string' syntax is not deprecated; nevertheless, ALTER USER is now the preferred statement for assigning passwords.

The PASSWORD () function is obsolete, as is set password...= password ('string_word'), while set password='string_word' is somewhat obsolete. Alter user is now used to set the user's password.

The old_passwords system variable. Account authentication plugins can no longer be left unspecified in the mysql.user table, so any statement that assigns a password from a cleartext string can unambiguously determine the hashing method to use on the string before storing it in the mysql.user table. This renders old_passwords superflous.

?? The system variable old_password, the account authentication plug-in must now be specified in the mysql table

Relying on implicit GROUP BY sorting in MySQL 5.7 is deprecated. To achieve a specific sort order of grouped results, it is preferable to use an explicit ORDER BY clause. GROUP BY sorting is a MySQL extension that may change in a future release; for example, to make it possible for the optimizer to order groupings in whatever manner it deems most efficient and to avoid the sorting overhead.

Implicit group by sorting has been deprecated in mysql 5.7, and the specified ORDER BY must be displayed in order to specify the sort order. Group by sorting will be available in future versions of mysql.

The EXTENDED and PARTITIONS keywords for the EXPLAIN statement are deprecated. These keywords are still recognized but are now unnecessary because their effect is always enabled.

The keywords EXTEND and partiitons used in explan have been deprecated. These keywords are available, but they are now enabled all the time.

The option-skip-innodb and his synonym-innodb=OFF,--disable-innodb are also discarded. Because the INNODB engine cannot be disabled at the beginning of 5.7.

The client-ssl and-- ssl-verify-server-cert options are discarded, use-- ssl-mode=REQUIRED instead of-- ssl=1 and-- enable-ssl Use, use-ssl-mode=DISABLED instead of-ssl=0,--skip-ssl or-- disable-ssl, use-- ssl-mode=VERIFY_IDENTITY instead of-- ssl-verify-server-cert, note that these parameters are not discarded on the server side

The server-side options of the system variables log_warnings and-- log-warnings are discarded and the variable log_error_verbosity is used instead

The temp-pool service option, the system variable, is deprecated.

The system variable binlog_max_flush_queue_time does nothing in version MySQL 5.7 and is discarded in version MySQL 5.7.9.

The system variable innodb_support_xa was used to enable the InnoDB engine to support two-phase commit in XA transactions, in version MySQL 5.7.10. Obsolete and exist as always enabled objects in future versions

The system variables metadata_locks_cache_size and metadata_locks_hash_instances are obsolete and reversion MySQL 5.7.4. Nothing has been done in the

The system variable sync_frm is discarded.

The global system variables character_set_database and collation_database are discarded and will be removed in future releases.

The system parameter that sets character_set_database and collation_database for the session is deprecated if you try to generate a warning, this parameter will be set to read only in a later version and will throw an exception if allocated.

The encryption functions ENCRYPT (), ENCODE (), DECODE (), DES_ENCRYPT (), and DES_DECRYPT () are replaced by AES_ENCRYPT () and AES_DECRYPT ().

The space function MBREqual () is discarded. Use MBREquals () instead.

The INFORMATION_SCHEMA table PROFILING of the performance library is obsolete. Using Performance Schema instead of Performance Schema can be found in the section Chapter 24, MySQL Performance Schema.

The tables INNODB_LOCKS and INNODB_LOCK_WAITS in the performance library INFORMATION_SCHEMA are discarded, removed in version 8.0, and replaced by other views.

The pronoun\ N of NULL is discarded and will be removed in 8.0. replace it with NULL.

This change will not affect the use of LOAD DATA INFILE or SELECT. INTO OUTFILE, import and export the text. See Section 14.2.6, "LOAD DATA INFILE Syntax".

Mysqld_safe no longer supports syslog. Use local service syslog support instead. See Section 6.4.2, "The Error Log".

The database name of the MYSQL version before the conversion 5.1 contains the keyword format (using # mysql150#) is discarded. Here, the-- fix-db-names and-- fix-table-names option used by mysqlcheck and the UPGRADE DATA DIRECTORY NAME is also discarded.

Mysql_install_db function report has been integrated into mysqld. When you need to initialize the database, you used mysql_install_db, but mysqld now calls the-- initialize or-- initialize-insecure option. The difference is whether you want the server to randomly generate a password for the 'root'@'localhost' account.

Mysql_install_db is now obsolete, and mysqld needs to use the parameter-- bootstrap.

The The mysql_plugin function pack is discarded. Optionally, specify the-- plugin-load and-- plugin-load-add options at server startup time, or use the INSTALL PLUGIN statement at run time.

The API interfaces mysql_kill (), mysql_list_fields (), mysql_list_processes (), and mysql_refresh () provided to C language are discarded. Instead of mysql_query (), add KILL, SHOW COLUMNS, SHOW PROCESSLIST, or FLUSH statement, respectively.

The mysql_shutdown () C API interface is obsolete. Instead, use mysql_query () to execute the SHUTDOWN command

Replace feature pack is obsolete

DTrace supports obsolescence.

Features removed from MySQL version 5.7

The following items are features that were removed in version 5.7.

The hash format using the old version of pre-4.1 password has been removed.

The mysql_old_password authentication plug-in is removed. Use the modified plug-in to disable users and record "unknown plugin" information in error.log when the service is started. For upgrade information about certification, see see Section 7.5.1.3, "Migrating Away from Pre-4.1 Password Hashing and the mysql_old_password Plugin".

The-- secure-auth is now an empty operation that will be removed in future versions

The-- the skip-secure-auth option is no longer supported. Error will be reported if you use it.

The value of the The secure_auth system variable is limited to 1, and the 0 value is no longer allowed.

For the old_passwords system variable value is not allowed

The The OLD_PASSWORD () function is removed.

In version 5.6.6 of MySQL, the data type YEAR (2) is obsolete. The current version 5. 7 YEAR (2) has been abandoned. If you upgrade the service to version 5.7 or later, please remember to upgrade your data type from YEAR (2) to YEAR (4). Check the section, see Section 12.3.4, "YEAR (2) Limitations and Migrating to YEAR (4)". Looking for conversion strategy input, run mysql_upgrade after upgrading.

The The innodb_mirrored_log_groups system variable supports only 1. 0 values. So it is meaningless.

The The storage_engine system variable uses default_storage_engine

The thread_concurrency is discarded.

The timed_mutexes is discarded.

The IGNORE clause for ALTER TABLE.

INSERT DELAYED no longer supports. The service recognizes and ignores the DELAYED keyword, processes insert as a non-latency insert and generates an ER_WARN_LEGACY_SYNTAX_CONVERTED warning. ("INSERT DELAYED is no longer supported The statement was converted to INSERT.") Similarly, REPLACE DELAYED is treated as a non-delay replace. The keyword DELAYED will be removed in future versions.

In addition, the service-related DELAYED-related option is also removed:

The-- delayed-insert option for mysqldump.

The COUNT_WRITE_DELAYED, SUM_TIMER_WRITE_DELAYED, MIN_TIMER_WRITE_DELAYED, AVG_TIMER_WRITE_DELAYED, and MAX_TIMER_WRITE_DELAYED columns of the Performance Schema table_lock_waits_summary_by_table table.

Mysqlbinlog no longer writes comments mentioning INSERT DELAYED.

The database symbolic link of the .fym file on windows has been removed because it has been repeated with the local symbolic link mklink. All sym image links will be ignored and must be created using mklink instead. See Section 9.12.3.3, "Using Symbolic Links for Databases on Windows".

Unused-- basedir,-- datadir, and-- tmpdir options in mysql_upgrade have been removed

In the early days, program options could use the full name or an explicit prefix. For example, the option-- compress can be specified as-- compr when using mysqldump, not-- comp because the word cannot be specified. Now the prefix option is no longer supported and accepts only full-text nouns. This is because there will be a problem if a new variable comes in:

The option-- key-buffer must be specified as-- key-buffer-size.

The option-- skip-grant must be specified as-- skip-grant-tables.

The SHOW ENGINE INNODB MUTEX output is removed. The corresponding information can be found in the table provided by Performance Schema.

The InnoDB tablespace monitoring and InnoDB table monitoring have been removed. The monitoring of INNODB table can find the corresponding information in the table in InnoDB INFORMATION_SCHEMA.

Tables with special names are used to enable or disable standard INNODB and innodb lock monitoring (innodb_monitor and innodb_lock_monitor). It has been disabled to replace it: innodb_status_output and innodb_status_output_locks. For more information, see Section 15.17, "InnoDB Monitogers".

The The innodb_use_sys_malloc and innodb_additional_mem_pool_size system variables were removed in version 5.6.3

The msql2mysql, mysql_convert_table_format, mysql_find_rows, mysql_fix_extensions, mysql_setpermission, mysql_waitpid, mysql_zap, mysqlaccess, and mysqlbug utilities. Be removed

The The mysqlhotcopy function has been removed. It was replaced by mysqldump and MySQL Enterprise Backup.

The binary-configure.sh script. Be deleted

The The INNODB_PAGE_ATOMIC_REF_COUNT CMake option is removed.

The The innodb_create_intrinsic option is removed.

The The innodb_optimize_point_storage option and the associated data type (DATA_POINT and DATA_VAR_POINT) are removed.

The The innodb_log_checksum_algorithm option is removed.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.