Log message:
mariadb: Update to 10.6.20

                         MariaDB 10.6.20 Release Notes

Notable Items

  Storage Engines

    InnoDB

     * Fix attempt to read outside the bounds of the file: ./ibdata1
       (MDEV-34453)
     * Fix MariaDB crash with SIGILL because the OS does not support AVX512
       (MDEV-34565)
     * Fix deadlock caused by lock_rec_unlock_unmodified() (MDEV-34690)
     * Fix duplicate key error in table 'mysql.innodb_table_stats'
       (MDEV-34207)
     * Fix possible adaptive hash index corruption with ALTER TABLE...IMPORT
       TABLESPACE and FULLTEXT SEARCH (MDEV-35059)
     * XA prepare now correctly releases unmodified records in non-blocking
       mode (MDEV-34466)
     * Fix contention between secondary index UPDATE and purge due to large
       innodb_purge_batch_size (MDEV-34515)
          * New default of 127 for innodb_purge_batch_size (previously 1000).
     * Fix redundant page lookups that hurt performance (MDEV-34791)

    Spider

     * Fix server crash when calling spider UDF after aria_encrypt_tables is
       enabled (MDEV-34682)
     * SELECT MIN on Spider table no longer returns more rows than expected
       (MDEV-26345)
     * Fix Spider group by handler wrong result on order by aggregate
       (MDEV-29546)

    S3

     * Fix Storage Engine S3 that caused 500 error when using Huawai Cloud S3
       and possibly other S3 providers (MDEV-34867)
          * A new option s3-provider has been added. s3-provider=Huawai needs
            to be set for Huawai Cloud S3

  Optimizer

     * Fix crash caused by query containing constant having clause
       (MDEV-23983)
     * Fix assertion with optimizer_join_limit_pref_ratio and 1-table select
       (MDEV-35072)

  Data Definition - Alter Table

     * Modification of the column now correctly checks foreign key constraint
       (MDEV-34392)
     * Fix incorrect NULL value handling for instantly dropped BLOB columns
       (MDEV-35122)

  Galera

     * Galera updated to 26.4.20
     * Fix server crash when setting wsrep_cluster_address after adding
       invalid value to wsrep_allowlist table (MDEV-31173)
     * TOI (ALTER) no longer hangs on a parent table if SR transaction is in
       progress on a child table (MDEV-34836)
     * Fix INSERT...SELECT' on MyISAM table suddenly replicated by alera
       (MDEV-34647)
     * Fix case where with wsrep_mode=REPLICATE_ARIA, only part of the
       mixed-engine transactions was replicated (MDEV-30653)
     * Fix galera_ssl_reload failure with warning message (MDEV-32778)
     * When InnoDB gets an assertion failure, WSREP layer is now handled
       gracefully (MDEV-32363)

  Backup

     * Recovery now correctly notes some log corruption (MDEV-34802)
     * InnoDB now correctly merges the change buffer to ROW_FORMAT=COMPRESSED
       tables (MDEV-34879)
     * Can now selectively restore sequences using innodb tables from backup
       (MDEV-32350)

  JSON

     * JSON_TABLE now properly unquotes strings (MDEV-27412)
     * JSON_TABLE default values now allows non-string literals (MDEV-25822)

  General

     * mariadbd no longer hangs on startup when --init-file target does not
       exist (MDEV-34814)
     * LOAD DATA INFILE with geometry data no longer fails (MDEV-34883)
     * Fix heap-use-after-free in group_concat with compressed or GIS columns
       (MDEV-16699)
     * Fix assertion `!is_cond()' failed in Item_bool_func::val_int /
       do_select (MDEV-35135)
     * Trigger now works correctly with bulk update (MDEV-34718)
     * Fix assertion failure in find_producing_item upon a query from a view
       (MDEV-35276)
     * Add new setting, --quick-max-column-width to the mariadb client for
       use in --quick mode (MDEV-34704)

Log message:
*: recursive bump for icu 76 shlib major version bump

Log message:
*: revbump for icu downgrade

Log message:
*: recursive bump for icu 76.1 shlib bump

Log message:
*: replace CMAKE_ARGS with CMAKE_CONFIGURE_ARGS

Log message:
mariadb106-*: Update to 10.6.19

                         MariaDB 10.6.19 Release Notes

Notable Items

  Storage Engines

    InnoDB

     * We fail to terminate transaction early with ER_LOCK_TABLE_FULL when
       lock memory is growing (MDEV-34167)
     * Alter operation on redundant table aborts the server (MDEV-34222)
     * MariaDB crashes with SIGILL because the OS does not support AVX512
       (MDEV-34565)
     * InnoDB: Failing assertion: stat_n_leaf_pages > 0 in
       ha_innobase::estimate_rows_upper_bound (MDEV-34474)
     * wait_for_read in buf_page_get_low hurts performance (MDEV-34458)
     * InnoDB: Assertion failure in file ./storage/innobase/page/page0zip.cc
       line 4211 (MDEV-34357)

    Aria

     * Index for (specific) Aria table is created as corrupted (MDEV-34522)

    Spider

     * UBSAN: runtime error: applying non-zero offset in spider_free_mem and
       SIGSEGV in spider_free_mem on SELECT (MDEV-31475)
     * SIGSEGV in ha_resolve_storage_engine_name, UBSAN: runtime error: index
       4294967295 out of bounds for type 'st_plugin_int *[64]' (MDEV-32487)
     * SIGSEGV in spider_db_conn::fin_loop_check, and ASAN:
       heap-use-after-free in spider_db_mbase::fin_loop_check on SHOW TABLE
       STATUS (MDEV-34541)
     * SIGSEGV in ha_spider::lock_tables on BEGIN after table lock
       (MDEV-29962)
     * SIGSEGV in spider_conn_first_link_idx and others on DELETE, INSERT and
       SELECT (MDEV-32492)
     * Spider: Crashes, asserts, hangs, memory corruptions and ASAN
       heap-use-after-free's (MDEV-27902)
     * Spider: @@insert_id 128 to TINYINT: Assertion `!is_set() || (m_status
       == DA_OK_BULK && is_bulk_op())' failed. (MDEV-28105)
     * ASAN errors in spider_db_free_result after partition DDL (MDEV-29027)
     * Server crashes when calling spider UDF after aria_encrypt_tables is
       enabled (MDEV-34682)

  Partitioning

     * MariaDB Server crashes with ill-formed partitions (MDEV-32155)
     * SIGSEGV in parse_engine_part_options on INSERT, SELECT or ALTER
       (MDEV-34421)
     * Assertion `auto_increment_value' failed in ha_partition::info on
       INSERT into MEMORY table (MDEV-24610)

  Character Sets

     * On startup: UBSAN: applying zero offset to null pointer in
       my_copy_fix_mb from strings/ctype-mb.c and other locations
       (MDEV-34226)
     * On startup: UBSAN: runtime error: applying zero offset to null pointer
       in skip_trailing_space and
       my_hash_sort_utf8mb3_general1400_nopad_as_ci (MDEV-34187)
     * SHOW CREATE DATABASE statement crashes the server when db name
       contains some unicode characters, ASAN stack-buffer-overflow
       (MDEV-32376)
     * Wrong result set with utf8mb4_danish_ci and BNLH join (MDEV-34417)

  Optimizer

     * On startup: UBSAN: runtime error: applying non-zero offset in
       JOIN::make_aggr_tables_info in sql/sql_select.cc (MDEV-34227)
     * Crash after killing query while it is processed by test_quick_select
       (MDEV-30651)
     * Extend condition normalization to include 'NOT a' (MDEV-19520)
     * Constant subquery causing a crash in pushdown optimization
       (MDEV-29363)
     * Crash when pushing condition with CHARSET()/COERCIBILITY() into
       derived table (MDEV-33010)
     * 2nd execution name resolution problem with pushdown into unions
       (MDEV-34506)
     * Assertion `(key_part->key_part_flag & 4) == 0' failed key_hashnr
       (MDEV-34580)
     * Crash caused by query containing constant having clause (MDEV-23983)
     * Using NAME_CONST() (or executing query from the stored procedure and
       referring to a local variable) changes the plan and may make execution
       slower (MDEV-33971)
     * ORDER BY DESC causes ROWID Filter optimization performance degradation
       (MDEV-33875)

  Replication

     * Auto-generated DELETE from HEAP table can break replication
       (MDEV-25607)
     * Replication fails when XA transactions are used where the slave has
       replicate_do_db set and the client has touched a different database
       when running DML such as inserts. (MDEV-33921)
     * replication master password length (MDEV-23857)
     * an option to enable semisync recovery (MDEV-33465)

  Backup

     * Hide password passed on commandline from xtrabackup_info (MDEV-34434)

  Galera

     * Galera updated to 26.4.19
     * galera_gtid_2_cluster: Assertion `thd->wsrep_next_trx_id() !=
       (0x7fffffffffffffffLL * 2ULL + 1)' (MDEV-32633)
     * table gtid_slave_pos entries never been deleted with wsrep_gtid_mode =
       0 (MDEV-34170)
     * Deadlock found when trying to get lock during applying (MDEV-31658)
     * Change error code for Galera unkillable threads (MDEV-12008)
     * 10.11.8 cluster becomes inconsistent when using composite primary key
       and partitioning (MDEV-34269)
     * wsrep_sst_mariabackup use /tmp dir during SST rather then user defined
       tmpdir (MDEV-32158)

  General

     * As per the MariaDB Deprecation Policy, this will be the last release
       of MariaDB 10.6 for Debian 10 "Buster", RHEL/CentOS 7, and Fedora 38
     * IMPORT TABLESPACE unnecessarily traverses tablespaces list
       (MDEV-34670)
     * unknown variable defaults-group-suffix= with
       mariadb-secure-installation (MDEV-33265)
     * mariadb-install-db appears to hang on macOS (MDEV-34129)
     * We fail to terminate transaction early with ER_LOCK_TABLE_FULL when
       lock memory is growing (MDEV-34167)
     * Disable new connections in case of fatal signal (MDEV-34475)
     * Control over memory allocated for SP/PS (MDEV-14959)
     * Trigger doesn't work correctly with bulk insert (MDEV-24411)
     * Assertion `table->field[0]->ptr >= table->record[0] &&
       table->field[0]->ptr <= table->record[0] + \ 
table->s->reclength' failed
       in void handler::assert_icp_limitations(uchar*) (MDEV-34632)
     * Sandbox mode \- is not compatible with --binary-mode (MDEV-34203)

Log message:
Apply fix for hang and data loss on mariadb 10.6 shutdown

Apply https://github.com/MariaDB/server/pull/3253 to fix a hang on
at least NetBSD resulting in unrecoverable database corruption when
attempting to shutdown the mariadb server

Log message:
revbump after icu and protobuf updates