- 18 Sep, 2017 5 commits
-
-
Sergei Golubchik authored
-
Sergei Golubchik authored
create libmariadbclient.so compatibility symlinks on debian
-
Sergei Golubchik authored
MDEV-13636 ALTER TABLE ... DELAY_KEY_WRITE=1 creates table copy for MyISAM table with DATA DIRECTORY/INDEX DIRECTORY options correct detection of changes in DATA/INDEX DIRECTORY
-
Sergei Golubchik authored
MDEV-13650 Backport fix for MDEV-13060 (crash when both AWS plugin and server_audit are loaded) to 10.1 MDEV-13060 Server Audit Plugin Crashes with AWS KMS plugin don't do auditing if thd is NULL. Collaterals: * copy-paste bugs server_audit.c
-
Marko Mäkelä authored
The purpose of the test is to ensure that redo log apply will extend data files before applying page-level redo log records. The test intermittently failed, because the doublewrite buffer would sometimes contain data for the pages that the test truncated. When the test truncates data files, it must also remove the doublewrite buffer entries, because under normal operation, the doublewrite buffer would only be written to if the data file already has been extended.
-
- 17 Sep, 2017 1 commit
-
-
Marko Mäkelä authored
-
- 16 Sep, 2017 8 commits
-
-
Vladislav Vaintroub authored
Compilation got confused about 2 wsrep.h headers in include path Rename backup's wsrep.h to backup_wsrep.h to fixO
-
Marko Mäkelä authored
Introduce innodb_encrypt_log.combinations and prove that the encryption and decryption take place during both online ADD INDEX (WL#5266) and online table-rebuilding ALTER (WL#6625).
-
Marko Mäkelä authored
-
Marko Mäkelä authored
Apply the same changes to both InnoDB and XtraDB.
-
Vladislav Vaintroub authored
message was written.
-
Vladislav Vaintroub authored
The messages are getting lost because they are written with fprintf() but without fflush(), so abort() would lose buffered text. Applied fix from 10.2, which makes in_logf() use sql_print_information(), which in turn does proper flush after each message
-
Vladislav Vaintroub authored
-
Marko Mäkelä authored
MDEV-13807 mariabackup --apply-log-only does generate redo log by performing rollback and possibly other tasks Skip rollback and other redo-log-generating tasks if srv_apply_log_only is set. Instead of assigning the debug variable recv_no_log_write = FALSE, assign it to srv_apply_log_only, so that any unwanted writes are caught.
-
- 15 Sep, 2017 1 commit
-
-
Alexander Barkov authored
The problem was introduced by the patch for MDEV-7661, which (in addition to the fix itself) included an attempt to make CONVERT/CAST work in the same way with fields (i.e. return NULL in strict mode if a non-convertable character found). It appeared to be a bad idea and some users were affected by this behavior change. Changing CONVERT/CAST not depend on sql_mode (restoring pre-10.1.4 behavior).
-
- 14 Sep, 2017 1 commit
-
-
Jan Lindström authored
…porary file Fixed by removing writing key version to start of every block that was encrypted. Instead we will use single key version from log_sys crypt info. After this MDEV also blocks writen to row log are encrypted and blocks read from row log aren decrypted if encryption is configured for the table. innodb_status_variables[], struct srv_stats_t Added status variables for merge block and row log block encryption and decryption amounts. Removed ROW_MERGE_RESERVE_SIZE define. row_merge_fts_doc_tokenize Remove ROW_MERGE_RESERVE_SIZE row_log_t Add index, crypt_tail, crypt_head to be used in case of encryption. row_log_online_op, row_log_table_close_func Before writing a block encrypt it if encryption is enabled row_log_table_apply_ops, row_log_apply_ops After reading a block decrypt it if encryption is enabled row_log_allocate Allocate temporary buffers crypt_head and crypt_tail if needed. row_log_free Free temporary buffers crypt_head and crypt_tail if they exist. row_merge_encrypt_buf, row_merge_decrypt_buf Removed. row_merge_buf_create, row_merge_buf_write Remove ROW_MERGE_RESERVE_SIZE row_merge_build_indexes Allocate temporary buffer used in decryption and encryption if needed. log_tmp_blocks_crypt, log_tmp_block_encrypt, log_temp_block_decrypt New functions used in block encryption and decryption log_tmp_is_encrypted New function to check is encryption enabled. Added test case innodb-rowlog to force creating a row log and verify that operations are done using introduced status variables.
-
- 07 Sep, 2017 4 commits
-
-
Marko Mäkelä authored
-
Marko Mäkelä authored
recv_reset_logs(): Initialize the redo log buffer, so that no data from the old redo log can be written to the new redo log. This bug has very little impact before MariaDB 10.2. The innodb_log_encrypt option that was introduced in MariaDB 10.1 increases the impact. If the redo log used to be encrypted, and it is being resized and encryption disabled, then previously encrypted data could end up being written to the new redo log in clear text. This resulted in encryption.innodb_encrypt_log test failures in MariaDB 10.2.
-
Marko Mäkelä authored
The SEARCH_ABORT logic was not working as intended (tests were not being aborted).
-
Marko Mäkelä authored
In XtraDB, buf_block_get_frame() can return NULL, and this value can be passed to buf_page_print(). Do not declare the parameter as nonnull.
-
- 06 Sep, 2017 4 commits
-
-
Marko Mäkelä authored
The bug only affects ROW_FORMAT=DYNAMIC tables. It is reproducible already with 2 records in the table. Keep testing with ROW_FORMAT=REDUNDANT just in case.
-
Marko Mäkelä authored
The bug only affected innodb_page_size=64k. There is no point running the test with smaller page sizes.
-
Marko Mäkelä authored
-
Marko Mäkelä authored
buf_page_print(): Remove the parameter 'flags', and when a server abort is intended, perform that in the caller. In this way, page corruption reports due to different reasons can be distinguished better. This is non-functional code refactoring that does not fix any page corruption issues. The change is only made to avoid falsely grouping together unrelated causes of page corruption.
-
- 04 Sep, 2017 3 commits
-
-
Andrei Elkin authored
A new $MYSQLD_LAST_CMD evaluation was too late in case --manual-gdb. Now it is done before the server restart type branches which is safe and the args value has been fully computed by the new point of evaluation.
-
Oleksandr Byelkin authored
save thd->select_number between parsing and executions (in case it was not complete executed due to errors (for example epsent table))
-
Marko Mäkelä authored
This is a backport of the following: MDEV-13009 10.1.24 does not compile on architectures without 64-bit atomics Add a missing #include "sync0types.h" that was removed in MDEV-12674.
-
- 01 Sep, 2017 1 commit
-
-
Marko Mäkelä authored
metadata_lock_info_duration[]: Remove the unused variable. Add some comments /* fall through */ to silence -Wimplicit-fallthrough
-
- 31 Aug, 2017 6 commits
-
-
Marko Mäkelä authored
-
Vladislav Vaintroub authored
char* parameter is expected by the message ER_KEY_COLUMN_DOES_NOT_EXITS, thus pass char*, rather than LEX_STRING.
-
Vladislav Vaintroub authored
ERROR_FILE_SYSTEM_LIMITATION was seen by support when backing up large file. However mariabackup error message was not very helpful, since it mapped the error to generic catch-all EINVAL. With the patch, ERROR_FILE_SYSTEM_LIMITATION will be mapped to more appropriate EFBIG. Also add mapping from ERROR_NO_SYSTEM_RESOURCES to ENOMEM.
-
Marko Mäkelä authored
There is a race condition in InnoDB startup. A number of fil_crypt_thread are created by fil_crypt_threads_init(). These threads may call btr_scrub_complete_space() before btr_scrub_init() was called. Those too early calls would be accessing an uninitialized scrub_stat_mutex. innobase_start_or_create_for_mysql(): Invoke btr_scrub_init() before fil_crypt_threads_init(). fil_crypt_complete_rotate_space(): Only invoke btr_scrub_complete_space() if scrubbing is enabled. There is no need to update the statistics if it is not enabled.
-
Jan Lindström authored
After MDEV-13583: Improvements for MTR rebootstrap introduced in MDEV-12042 bootsrap correctly creates mysql/innodb_table_stats and mysql/innodb_index_stats InnoDB tables before innodb_encryption test starts. These tables are also encrypted or decrypted, thus we need to wait also these tables (if not we could randomly get different results as system tablespace and these tables are encrypted or decrypted in parallel).
-
Jan Lindström authored
Problem was incorrect definition of wsrep_recovery, trx_sys_update_wsrep_checkpoint and trx_sys_read_wsrep_checkpoint functions causing innodb_plugin not to load as there was undefined symbols.
-
- 30 Aug, 2017 4 commits
-
-
Marko Mäkelä authored
-
Marko Mäkelä authored
-
Marko Mäkelä authored
wsrep_drop_table_query(): Remove the definition of this ununsed function. row_upd_sec_index_entry(), row_upd_clust_rec_by_insert(): Evaluate the simplest conditions first. The merge could have slightly hurt performance by causing extra calls to wsrep_on().
-
Marko Mäkelä authored
This is not affecting correctness; delete NULL is a valid operation.
-
- 29 Aug, 2017 2 commits
-
-
Marko Mäkelä authored
recv_find_max_checkpoint(): Refer to MariaDB 10.2.2 instead of MySQL 5.7.9. Do not hint that a binary downgrade might be possible, because there are many changes in InnoDB 5.7 that could make downgrade impossible: a column appended to SYS_INDEXES, added SYS_* tables, undo log format changes, and so on.
-
Jan Lindström authored
-