mariadb/storage/innobase/ut/ut0ut.cc

561 lines
14 KiB
C++
Raw Normal View History

/*****************************************************************************
2018-01-23 17:43:37 +02:00
Copyright (c) 1994, 2017, Oracle and/or its affiliates. All Rights Reserved.
Copyright (c) 2017, 2022, MariaDB Corporation.
This program is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free Software
Foundation; version 2 of the License.
This program is distributed in the hope that it will be useful, but WITHOUT
ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with
this program; if not, write to the Free Software Foundation, Inc.,
51 Franklin Street, Fifth Floor, Boston, MA 02110-1335 USA
*****************************************************************************/
/***************************************************************//**
@file ut/ut0ut.cc
Various utilities for Innobase.
Created 5/11/1994 Heikki Tuuri
********************************************************************/
#include "ha_prototypes.h"
#if HAVE_SYS_TIME_H
#include <sys/time.h>
#endif
#ifndef UNIV_INNOCHECKSUM
#include <mysql_com.h>
#include "ut0ut.h"
#include "trx0trx.h"
#include <string>
#include "log.h"
#include "my_cpu.h"
#ifndef DBUG_OFF
#include "rem0rec.h"
#endif
/**********************************************************//**
Returns the number of milliseconds since some epoch. The
value may wrap around. It should only be used for heuristic
purposes.
@return ms since epoch */
ulint
ut_time_ms(void)
/*============*/
{
return static_cast<ulint>(my_interval_timer() / 1000000);
}
#endif /* !UNIV_INNOCHECKSUM */
/**********************************************************//**
Prints a timestamp to a file. */
void
ut_print_timestamp(
/*===============*/
FILE* file) /*!< in: file where to print */
{
#ifdef _WIN32
SYSTEMTIME cal_tm;
GetLocalTime(&cal_tm);
#else
time_t tm;
struct tm cal_tm;
time(&tm);
localtime_r(&tm, &cal_tm);
#endif
fprintf(file,
IF_WIN("%u-%02u-%02u %02u:%02u:%02u %#zx",
"%d-%02d-%02d %02d:%02d:%02d %#zx"),
#ifdef _WIN32
cal_tm.wYear,
cal_tm.wMonth,
cal_tm.wDay,
cal_tm.wHour,
cal_tm.wMinute,
cal_tm.wSecond,
#else
cal_tm.tm_year + 1900,
cal_tm.tm_mon + 1,
cal_tm.tm_mday,
cal_tm.tm_hour,
cal_tm.tm_min,
cal_tm.tm_sec,
#endif
#ifdef UNIV_INNOCHECKSUM
ulint{0}
#else
ulint(pthread_self())
#endif
);
}
#ifndef UNIV_INNOCHECKSUM
/**********************************************************//**
Sprintfs a timestamp to a buffer, 13..14 chars plus terminating NUL. */
void
ut_sprintf_timestamp(
/*=================*/
char* buf) /*!< in: buffer where to sprintf */
{
#ifdef _WIN32
SYSTEMTIME cal_tm;
GetLocalTime(&cal_tm);
sprintf(buf, "%02u%02u%02u %2u:%02u:%02u",
cal_tm.wYear % 100,
cal_tm.wMonth,
cal_tm.wDay,
cal_tm.wHour,
cal_tm.wMinute,
cal_tm.wSecond);
#else
time_t tm;
struct tm cal_tm;
time(&tm);
localtime_r(&tm, &cal_tm);
sprintf(buf, "%02d%02d%02d %2d:%02d:%02d",
cal_tm.tm_year % 100,
cal_tm.tm_mon + 1,
cal_tm.tm_mday,
cal_tm.tm_hour,
cal_tm.tm_min,
cal_tm.tm_sec);
#endif
}
/*************************************************************//**
Prints the contents of a memory buffer in hex and ascii. */
void
ut_print_buf(
/*=========*/
FILE* file, /*!< in: file where to print */
const void* buf, /*!< in: memory buffer */
ulint len) /*!< in: length of the buffer */
{
const byte* data;
ulint i;
fprintf(file, " len " ULINTPF "; hex ", len);
for (data = (const byte*) buf, i = 0; i < len; i++) {
fprintf(file, "%02x", *data++);
}
fputs("; asc ", file);
data = (const byte*) buf;
for (i = 0; i < len; i++) {
int c = (int) *data++;
putc(isprint(c) ? c : ' ', file);
}
putc(';', file);
}
/*************************************************************//**
Prints the contents of a memory buffer in hex. */
void
ut_print_buf_hex(
/*=============*/
std::ostream& o, /*!< in/out: output stream */
const void* buf, /*!< in: memory buffer */
ulint len) /*!< in: length of the buffer */
{
const byte* data;
ulint i;
static const char hexdigit[16] = {
'0','1','2','3','4','5','6','7','8','9','A','B','C','D','E','F'
};
o << "(0x";
for (data = static_cast<const byte*>(buf), i = 0; i < len; i++) {
byte b = *data++;
o << hexdigit[int(b) >> 4] << hexdigit[b & 15];
}
o << ")";
}
/*************************************************************//**
Prints the contents of a memory buffer in hex and ascii. */
void
ut_print_buf(
/*=========*/
std::ostream& o, /*!< in/out: output stream */
const void* buf, /*!< in: memory buffer */
ulint len) /*!< in: length of the buffer */
{
const byte* data;
ulint i;
for (data = static_cast<const byte*>(buf), i = 0; i < len; i++) {
int c = static_cast<int>(*data++);
o << (isprint(c) ? static_cast<char>(c) : ' ');
}
ut_print_buf_hex(o, buf, len);
}
/** Get a fixed-length string, quoted as an SQL identifier.
If the string contains a slash '/', the string will be
output as two identifiers separated by a period (.),
as in SQL database_name.identifier.
@param [in] trx transaction (NULL=no quotes).
@param [in] name table name.
@retval String quoted as an SQL identifier.
*/
std::string
ut_get_name(
const trx_t* trx,
const char* name)
{
/* 2 * NAME_LEN for database and table name,
and some slack for the #mysql50# prefix and quotes */
char buf[3 * NAME_LEN];
const char* bufend;
bufend = innobase_convert_name(buf, sizeof buf,
name, strlen(name),
trx ? trx->mysql_thd : NULL);
buf[bufend - buf] = '\0';
return(std::string(buf, 0, size_t(bufend - buf)));
}
/**********************************************************************//**
Outputs a fixed-length string, quoted as an SQL identifier.
If the string contains a slash '/', the string will be
output as two identifiers separated by a period (.),
as in SQL database_name.identifier. */
void
ut_print_name(
/*==========*/
FILE* f, /*!< in: output stream */
const trx_t* trx, /*!< in: transaction */
const char* name) /*!< in: name to print */
{
/* 2 * NAME_LEN for database and table name,
and some slack for the #mysql50# prefix and quotes */
char buf[3 * NAME_LEN];
const char* bufend;
bufend = innobase_convert_name(buf, sizeof buf,
name, strlen(name),
trx ? trx->mysql_thd : NULL);
if (fwrite(buf, 1, size_t(bufend - buf), f) != size_t(bufend - buf)) {
perror("fwrite");
}
}
/**********************************************************************//**
Catenate files. */
void
ut_copy_file(
/*=========*/
FILE* dest, /*!< in: output file */
FILE* src) /*!< in: input file to be appended to output */
{
long len = ftell(src);
char buf[4096];
rewind(src);
do {
size_t maxs = len < (long) sizeof buf
? (size_t) len
: sizeof buf;
size_t size = fread(buf, 1, maxs, src);
if (fwrite(buf, 1, size, dest) != size) {
perror("fwrite");
}
len -= (long) size;
if (size < maxs) {
break;
}
} while (len > 0);
}
/** Convert an error number to a human readable text message.
The returned string is static and should not be freed or modified.
@param[in] num InnoDB internal error number
@return string, describing the error */
const char*
ut_strerr(
dberr_t num)
{
switch (num) {
case DB_SUCCESS:
return("Success");
case DB_SUCCESS_LOCKED_REC:
return("Success, record lock created");
case DB_ERROR:
return("Generic error");
case DB_READ_ONLY:
return("Read only transaction");
case DB_INTERRUPTED:
return("Operation interrupted");
case DB_OUT_OF_MEMORY:
return("Cannot allocate memory");
case DB_OUT_OF_FILE_SPACE:
return("Out of disk space");
case DB_LOCK_WAIT:
return("Lock wait");
case DB_DEADLOCK:
return("Deadlock");
MDEV-26642/MDEV-26643/MDEV-32898 Implement innodb_snapshot_isolation https://jepsen.io/analyses/mysql-8.0.34 highlights that the transaction isolation levels in the InnoDB storage engine do not correspond to any widely accepted definitions, such as "Generalized Isolation Level Definitions" https://pmg.csail.mit.edu/papers/icde00.pdf (PL-1 = READ UNCOMMITTED, PL-2 = READ COMMITTED, PL-2.99 = REPEATABLE READ, PL-3 = SERIALIZABLE). Only READ UNCOMMITTED in InnoDB seems to match the above definition. The issue is that InnoDB does not detect write/write conflicts (Section 4.4.3, Definition 6) in the above. It appears that as soon as we implement write/write conflict detection (SET SESSION innodb_snapshot_isolation=ON), the default isolation level (SET TRANSACTION ISOLATION LEVEL REPEATABLE READ) will become Snapshot Isolation (similar to Postgres), as defined in Section 4.2 of "A Critique of ANSI SQL Isolation Levels", MSR-TR-95-51, June 1995 https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/tr-95-51.pdf Locking reads inside InnoDB used to read the latest committed version, ignoring what should actually be visible to the transaction. The added test innodb.lock_isolation illustrates this. The statement UPDATE t SET a=3 WHERE b=2; is executed in a transaction that was started before a read view or a snapshot of the current transaction was created, and committed before the current transaction attempts to execute UPDATE t SET b=3; If SET innodb_snapshot_isolation=ON is in effect when the second transaction was started, the second transaction will be aborted with the error ER_CHECKREAD. By default (innodb_snapshot_isolation=OFF), the second transaction would execute inconsistently, displaying an incorrect SELECT COUNT(*) FROM t in its read view. If innodb_snapshot_isolation=ON, if an attempt to acquire a lock on a record that does not exist in the current read view is made, an error DB_RECORD_CHANGED (HA_ERR_RECORD_CHANGED, ER_CHECKREAD) will be raised. This error will be treated in the same way as a deadlock: the transaction will be rolled back. lock_clust_rec_read_check_and_lock(): If the current transaction has a read view where the record is not visible and innodb_snapshot_isolation=ON, fail before trying to acquire the lock. row_sel_build_committed_vers_for_mysql(): If innodb_snapshot_isolation=ON, disable the "semi-consistent read" logic that had been implemented by myself on the directions of Heikki Tuuri in order to address https://bugs.mysql.com/bug.php?id=3300 that was motivated by a customer wanting UPDATE to skip locked rows that do not match the WHERE condition. It looks like my changes were included in the MySQL 5.1.5 commit ad126d90e019f223470e73e1b2b528f9007c4532; at that time, employees of Innobase Oy (a recent acquisition of Oracle) had lost write access to the repository. The only reason why we set innodb_snapshot_isolation=OFF by default is backward compatibility with applications, such as the one that motivated the implementation of "semi-consistent read" back in 2005. In a later major release, we can default to innodb_snapshot_isolation=ON. Thanks to Peter Alvaro, Kyle Kingsbury and Alexey Gotsman for their work on https://github.com/jepsen-io/ and to Kyle and Alexey for explanations and some testing of this fix. Thanks to Vladislav Lesin for the initial test for MDEV-26643, as well as reviewing these changes.
2024-03-20 09:48:03 +02:00
case DB_RECORD_CHANGED:
return("Record changed");
#ifdef WITH_WSREP
case DB_ROLLBACK:
return("Rollback");
MDEV-26642/MDEV-26643/MDEV-32898 Implement innodb_snapshot_isolation https://jepsen.io/analyses/mysql-8.0.34 highlights that the transaction isolation levels in the InnoDB storage engine do not correspond to any widely accepted definitions, such as "Generalized Isolation Level Definitions" https://pmg.csail.mit.edu/papers/icde00.pdf (PL-1 = READ UNCOMMITTED, PL-2 = READ COMMITTED, PL-2.99 = REPEATABLE READ, PL-3 = SERIALIZABLE). Only READ UNCOMMITTED in InnoDB seems to match the above definition. The issue is that InnoDB does not detect write/write conflicts (Section 4.4.3, Definition 6) in the above. It appears that as soon as we implement write/write conflict detection (SET SESSION innodb_snapshot_isolation=ON), the default isolation level (SET TRANSACTION ISOLATION LEVEL REPEATABLE READ) will become Snapshot Isolation (similar to Postgres), as defined in Section 4.2 of "A Critique of ANSI SQL Isolation Levels", MSR-TR-95-51, June 1995 https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/tr-95-51.pdf Locking reads inside InnoDB used to read the latest committed version, ignoring what should actually be visible to the transaction. The added test innodb.lock_isolation illustrates this. The statement UPDATE t SET a=3 WHERE b=2; is executed in a transaction that was started before a read view or a snapshot of the current transaction was created, and committed before the current transaction attempts to execute UPDATE t SET b=3; If SET innodb_snapshot_isolation=ON is in effect when the second transaction was started, the second transaction will be aborted with the error ER_CHECKREAD. By default (innodb_snapshot_isolation=OFF), the second transaction would execute inconsistently, displaying an incorrect SELECT COUNT(*) FROM t in its read view. If innodb_snapshot_isolation=ON, if an attempt to acquire a lock on a record that does not exist in the current read view is made, an error DB_RECORD_CHANGED (HA_ERR_RECORD_CHANGED, ER_CHECKREAD) will be raised. This error will be treated in the same way as a deadlock: the transaction will be rolled back. lock_clust_rec_read_check_and_lock(): If the current transaction has a read view where the record is not visible and innodb_snapshot_isolation=ON, fail before trying to acquire the lock. row_sel_build_committed_vers_for_mysql(): If innodb_snapshot_isolation=ON, disable the "semi-consistent read" logic that had been implemented by myself on the directions of Heikki Tuuri in order to address https://bugs.mysql.com/bug.php?id=3300 that was motivated by a customer wanting UPDATE to skip locked rows that do not match the WHERE condition. It looks like my changes were included in the MySQL 5.1.5 commit ad126d90e019f223470e73e1b2b528f9007c4532; at that time, employees of Innobase Oy (a recent acquisition of Oracle) had lost write access to the repository. The only reason why we set innodb_snapshot_isolation=OFF by default is backward compatibility with applications, such as the one that motivated the implementation of "semi-consistent read" back in 2005. In a later major release, we can default to innodb_snapshot_isolation=ON. Thanks to Peter Alvaro, Kyle Kingsbury and Alexey Gotsman for their work on https://github.com/jepsen-io/ and to Kyle and Alexey for explanations and some testing of this fix. Thanks to Vladislav Lesin for the initial test for MDEV-26643, as well as reviewing these changes.
2024-03-20 09:48:03 +02:00
#endif
case DB_DUPLICATE_KEY:
return("Duplicate key");
case DB_MISSING_HISTORY:
return("Required history data has been deleted");
case DB_TABLE_NOT_FOUND:
return("Table not found");
case DB_TOO_BIG_RECORD:
return("Record too big");
case DB_TOO_BIG_INDEX_COL:
return("Index columns size too big");
case DB_LOCK_WAIT_TIMEOUT:
return("Lock wait timeout");
case DB_NO_REFERENCED_ROW:
return("Referenced key value not found");
case DB_ROW_IS_REFERENCED:
return("Row is referenced");
case DB_CANNOT_ADD_CONSTRAINT:
return("Cannot add constraint");
case DB_CORRUPTION:
return("Data structure corruption");
case DB_CANNOT_DROP_CONSTRAINT:
return("Cannot drop constraint");
case DB_NO_SAVEPOINT:
return("No such savepoint");
case DB_TABLESPACE_EXISTS:
return("Tablespace already exists");
case DB_TABLESPACE_DELETED:
return("Tablespace deleted or being deleted");
case DB_TABLESPACE_NOT_FOUND:
return("Tablespace not found");
case DB_LOCK_TABLE_FULL:
return("Lock structs have exhausted the buffer pool");
case DB_FOREIGN_DUPLICATE_KEY:
return("Foreign key activated with duplicate keys");
case DB_FOREIGN_EXCEED_MAX_CASCADE:
return("Foreign key cascade delete/update exceeds max depth");
case DB_TOO_MANY_CONCURRENT_TRXS:
return("Too many concurrent transactions");
case DB_UNSUPPORTED:
return("Unsupported");
case DB_INVALID_NULL:
return("NULL value encountered in NOT NULL column");
case DB_STATS_DO_NOT_EXIST:
return("Persistent statistics do not exist");
case DB_FAIL:
return("Failed, retry may succeed");
case DB_OVERFLOW:
return("Overflow");
case DB_UNDERFLOW:
return("Underflow");
case DB_STRONG_FAIL:
return("Failed, retry will not succeed");
case DB_ZIP_OVERFLOW:
return("Zip overflow");
case DB_RECORD_NOT_FOUND:
return("Record not found");
case DB_CHILD_NO_INDEX:
return("No index on referencing keys in referencing table");
case DB_PARENT_NO_INDEX:
return("No index on referenced keys in referenced table");
case DB_FTS_INVALID_DOCID:
return("FTS Doc ID cannot be zero");
case DB_INDEX_CORRUPT:
return("Index corrupted");
case DB_UNDO_RECORD_TOO_BIG:
return("Undo record too big");
case DB_END_OF_INDEX:
return("End of index");
case DB_IO_ERROR:
return("I/O error");
case DB_NOT_FOUND:
return("not found");
case DB_ONLINE_LOG_TOO_BIG:
return("Log size exceeded during online index creation");
case DB_IDENTIFIER_TOO_LONG:
return("Identifier name is too long");
case DB_FTS_EXCEED_RESULT_CACHE_LIMIT:
return("FTS query exceeds result cache limit");
case DB_TEMP_FILE_WRITE_FAIL:
2014-02-26 19:23:04 +01:00
return("Temp file write failure");
case DB_CANT_CREATE_GEOMETRY_OBJECT:
return("Can't create specificed geometry data object");
case DB_CANNOT_OPEN_FILE:
return("Cannot open a file");
case DB_TABLE_CORRUPT:
return("Table is corrupted");
2014-05-06 21:13:16 +02:00
case DB_FTS_TOO_MANY_WORDS_IN_PHRASE:
return("Too many words in a FTS phrase or proximity search");
case DB_DECRYPTION_FAILED:
return("Table is compressed or encrypted but uncompress or decrypt failed.");
case DB_IO_PARTIAL_FAILED:
return("Partial IO failed");
case DB_COMPUTE_VALUE_FAILED:
return("Compute generated column failed");
case DB_NO_FK_ON_S_BASE_COL:
return("Cannot add foreign key on the base column "
"of stored column");
MDEV-11254: innodb-use-trim has no effect in 10.2 Problem was that implementation merged from 10.1 was incompatible with InnoDB 5.7. buf0buf.cc: Add functions to return should we punch hole and how big. buf0flu.cc: Add written page to IORequest fil0fil.cc: Remove unneeded status call and add test is sparse files and punch hole supported by file system when tablespace is created. Add call to get file system block size. Used file node is added to IORequest. Added functions to check is punch hole supported and setting punch hole. ha_innodb.cc: Remove unneeded status variables (trim512-32768) and trim_op_saved. Deprecate innodb_use_trim and set it ON by default. Add function to set innodb-use-trim dynamically. dberr.h: Add error code DB_IO_NO_PUNCH_HOLE if punch hole operation fails. fil0fil.h: Add punch_hole variable to fil_space_t and block size to fil_node_t. os0api.h: Header to helper functions on buf0buf.cc and fil0fil.cc for os0file.h os0file.h: Remove unneeded m_block_size from IORequest and add bpage to IORequest to know actual size of the block and m_fil_node to know tablespace file system block size and does it support punch hole. os0file.cc: Add function punch_hole() to IORequest to do punch_hole operation, get the file system block size and determine does file system support sparse files (for punch hole). page0size.h: remove implicit copy disable and use this implicit copy to implement copy_from() function. buf0dblwr.cc, buf0flu.cc, buf0rea.cc, fil0fil.cc, fil0fil.h, os0file.h, os0file.cc, log0log.cc, log0recv.cc: Remove unneeded write_size parameter from fil_io calls. srv0mon.h, srv0srv.h, srv0mon.cc: Remove unneeded trim512-trim32678 status variables. Removed these from monitor tests.
2017-01-24 14:40:58 +02:00
case DB_IO_NO_PUNCH_HOLE:
return ("File system does not support punch hole (trim) operation.");
MDEV-12253: Buffer pool blocks are accessed after they have been freed Problem was that bpage was referenced after it was already freed from LRU. Fixed by adding a new variable encrypted that is passed down to buf_page_check_corrupt() and used in buf_page_get_gen() to stop processing page read. This patch should also address following test failures and bugs: MDEV-12419: IMPORT should not look up tablespace in PageConverter::validate(). This is now removed. MDEV-10099: encryption.innodb_onlinealter_encryption fails sporadically in buildbot MDEV-11420: encryption.innodb_encryption-page-compression failed in buildbot MDEV-11222: encryption.encrypt_and_grep failed in buildbot on P8 Removed dict_table_t::is_encrypted and dict_table_t::ibd_file_missing and replaced these with dict_table_t::file_unreadable. Table ibd file is missing if fil_get_space(space_id) returns NULL and encrypted if not. Removed dict_table_t::is_corrupted field. Ported FilSpace class from 10.2 and using that on buf_page_check_corrupt(), buf_page_decrypt_after_read(), buf_page_encrypt_before_write(), buf_dblwr_process(), buf_read_page(), dict_stats_save_defrag_stats(). Added test cases when enrypted page could be read while doing redo log crash recovery. Also added test case for row compressed blobs. btr_cur_open_at_index_side_func(), btr_cur_open_at_rnd_pos_func(): Avoid referencing block that is NULL. buf_page_get_zip(): Issue error if page read fails. buf_page_get_gen(): Use dberr_t for error detection and do not reference bpage after we hare freed it. buf_mark_space_corrupt(): remove bpage from LRU also when it is encrypted. buf_page_check_corrupt(): @return DB_SUCCESS if page has been read and is not corrupted, DB_PAGE_CORRUPTED if page based on checksum check is corrupted, DB_DECRYPTION_FAILED if page post encryption checksum matches but after decryption normal page checksum does not match. In read case only DB_SUCCESS is possible. buf_page_io_complete(): use dberr_t for error handling. buf_flush_write_block_low(), buf_read_ahead_random(), buf_read_page_async(), buf_read_ahead_linear(), buf_read_ibuf_merge_pages(), buf_read_recv_pages(), fil_aio_wait(): Issue error if page read fails. btr_pcur_move_to_next_page(): Do not reference page if it is NULL. Introduced dict_table_t::is_readable() and dict_index_t::is_readable() that will return true if tablespace exists and pages read from tablespace are not corrupted or page decryption failed. Removed buf_page_t::key_version. After page decryption the key version is not removed from page frame. For unencrypted pages, old key_version is removed at buf_page_encrypt_before_write() dict_stats_update_transient_for_index(), dict_stats_update_transient() Do not continue if table decryption failed or table is corrupted. dict0stats.cc: Introduced a dict_stats_report_error function to avoid code duplication. fil_parse_write_crypt_data(): Check that key read from redo log entry is found from encryption plugin and if it is not, refuse to start. PageConverter::validate(): Removed access to fil_space_t as tablespace is not available during import. Fixed error code on innodb.innodb test. Merged test cased innodb-bad-key-change5 and innodb-bad-key-shutdown to innodb-bad-key-change2. Removed innodb-bad-key-change5 test. Decreased unnecessary complexity on some long lasting tests. Removed fil_inc_pending_ops(), fil_decr_pending_ops(), fil_get_first_space(), fil_get_next_space(), fil_get_first_space_safe(), fil_get_next_space_safe() functions. fil_space_verify_crypt_checksum(): Fixed bug found using ASAN where FIL_PAGE_END_LSN_OLD_CHECKSUM field was incorrectly accessed from row compressed tables. Fixed out of page frame bug for row compressed tables in fil_space_verify_crypt_checksum() found using ASAN. Incorrect function was called for compressed table. Added new tests for discard, rename table and drop (we should allow them even when page decryption fails). Alter table rename is not allowed. Added test for restart with innodb-force-recovery=1 when page read on redo-recovery cant be decrypted. Added test for corrupted table where both page data and FIL_PAGE_FILE_FLUSH_LSN_OR_KEY_VERSION is corrupted. Adjusted the test case innodb_bug14147491 so that it does not anymore expect crash. Instead table is just mostly not usable. fil0fil.h: fil_space_acquire_low is not visible function and fil_space_acquire and fil_space_acquire_silent are inline functions. FilSpace class uses fil_space_acquire_low directly. recv_apply_hashed_log_recs() does not return anything.
2017-04-26 15:19:16 +03:00
case DB_PAGE_CORRUPTED:
return("Page read from tablespace is corrupted.");
/* do not add default: in order to produce a warning if new code
is added to the enum but not added here */
}
/* we abort here because if unknown error code is given, this could
mean that memory corruption has happened and someone's error-code
variable has been overwritten with bogus data */
ut_error;
/* NOT REACHED */
return("Unknown error");
}
namespace ib {
MDEV-27158: humanize the bytes in innodb info/error messages Log messages like total size = 17179869184, chunk size = 134217728 get hard to read. If we normalize it down to IEC units is easier. Idea thanks to Axel Schwenke. Review thanks to Eugene Kosov and Marko Mäkelä $ mariadblocal --innodb-buffer-pool-size=30G --innodb-log-file-size=128M Installing MariaDB/MySQL system tables in '/tmp/build-mariadb-server-10.7-datadir' ... 2021-12-09 9:54:04 0 [Note] /home/dan/repos/build-mariadb-server-10.7/sql/mysqld (server 10.7.2-MariaDB) starting as process 250473 ... 2021-12-09 9:54:04 0 [Note] InnoDB: The first data file './ibdata1' did not exist. A new tablespace will be created! 2021-12-09 9:54:04 0 [Note] InnoDB: Compressed tables use zlib 1.2.11 2021-12-09 9:54:04 0 [Note] InnoDB: Number of transaction pools: 1 2021-12-09 9:54:04 0 [Note] InnoDB: Using crc32 + pclmulqdq instructions 2021-12-09 9:54:04 0 [Note] InnoDB: Using liburing 2021-12-09 9:54:04 0 [Note] InnoDB: Initializing buffer pool, total size = 128.000MiB, chunk size = 128.000MiB 2021-12-09 9:54:04 0 [Note] InnoDB: Completed initialization of buffer pool 2021-12-09 9:54:04 0 [Note] InnoDB: Setting O_DIRECT on file ./ibdata1 failed 2021-12-09 9:54:04 0 [Note] InnoDB: Setting file './ibdata1' size to 12.000MiB. Physically writing the file full; Please wait ... 2021-12-09 9:54:04 0 [Note] InnoDB: File './ibdata1' size is now 12.000MiB. 2021-12-09 9:54:04 0 [Note] InnoDB: Setting log file ./ib_logfile101 size to 96.000MiB 2021-12-09 9:54:04 0 [Note] InnoDB: Renaming log file ./ib_logfile101 to ./ib_logfile0 2021-12-09 9:54:04 0 [Note] InnoDB: New log file created, LSN=10317 2021-12-09 9:54:04 0 [Note] InnoDB: Doublewrite buffer not found: creating new 2021-12-09 9:54:04 0 [Note] InnoDB: Doublewrite buffer created 2021-12-09 9:54:04 0 [Note] InnoDB: 128 rollback segments are active. 2021-12-09 9:54:04 0 [Note] InnoDB: Creating shared tablespace for temporary tables 2021-12-09 9:54:04 0 [Note] InnoDB: Setting file './ibtmp1' size to 12.000MiB. Physically writing the file full; Please wait ... 2021-12-09 9:54:04 0 [Note] InnoDB: File './ibtmp1' size is now 12.000MiB. 2021-12-09 9:54:04 0 [Note] InnoDB: 10.7.2 started; log sequence number 0; transaction id 3 OK 2021-12-09 9:54:04 0 [Note] sql/mysqld (server 10.7.2-MariaDB) starting as process 250501 ... 2021-12-09 9:54:04 0 [Note] InnoDB: Compressed tables use zlib 1.2.11 2021-12-09 9:54:04 0 [Note] InnoDB: Number of transaction pools: 1 2021-12-09 9:54:04 0 [Note] InnoDB: Using crc32 + pclmulqdq instructions 2021-12-09 9:54:04 0 [Note] InnoDB: Using liburing 2021-12-09 9:54:04 0 [Note] InnoDB: Initializing buffer pool, total size = 30.000GiB, chunk size = 128.000MiB 2021-12-09 9:54:04 0 [Note] InnoDB: Completed initialization of buffer pool 2021-12-09 9:54:04 0 [Note] InnoDB: Setting O_DIRECT on file ./ibdata1 failed 2021-12-09 9:54:04 0 [Note] InnoDB: Resizing redo log from 96.000MiB to 128.000MiB; LSN=41361 2021-12-09 9:54:04 0 [Note] InnoDB: Starting to delete and rewrite log file. 2021-12-09 9:54:04 0 [Note] InnoDB: Setting log file ./ib_logfile101 size to 128.000MiB 2021-12-09 9:54:04 0 [Note] InnoDB: Renaming log file ./ib_logfile101 to ./ib_logfile0 2021-12-09 9:54:04 0 [Note] InnoDB: New log file created, LSN=41361 2021-12-09 9:54:04 0 [Note] InnoDB: 128 rollback segments are active. 2021-12-09 9:54:04 0 [Note] InnoDB: Creating shared tablespace for temporary tables 2021-12-09 9:54:04 0 [Note] InnoDB: Setting file './ibtmp1' size to 12.000MiB. Physically writing the file full; Please wait ... 2021-12-09 9:54:04 0 [Note] InnoDB: File './ibtmp1' size is now 12.000MiB. 2021-12-09 9:54:04 0 [Note] InnoDB: 10.7.2 started; log sequence number 41349; transaction id 14 2021-12-09 9:54:04 0 [Note] InnoDB: Loading buffer pool(s) from /tmp/build-mariadb-server-10.7-datadir/ib_buffer_pool 2021-12-09 9:54:04 0 [Note] Plugin 'FEEDBACK' is disabled. 2021-12-09 9:54:04 0 [Note] InnoDB: Buffer pool(s) load completed at 211209 9:54:04 2021-12-09 9:54:04 0 [Note] sql/mysqld: ready for connections. Version: '10.7.2-MariaDB' socket: '/tmp/build-mariadb-server-10.7.sock' port: 0 Source distribution 2021-12-09 9:56:57 0 [Note] sql/mysqld (initiated by: unknown): Normal shutdown 2021-12-09 9:56:57 0 [Note] InnoDB: FTS optimize thread exiting. 2021-12-09 9:56:57 0 [Note] InnoDB: Starting shutdown... 2021-12-09 9:56:57 0 [Note] InnoDB: Dumping buffer pool(s) to /tmp/build-mariadb-server-10.7-datadir/ib_buffer_pool 2021-12-09 9:56:57 0 [Note] InnoDB: Buffer pool(s) dump completed at 211209 9:56:57 2021-12-09 9:56:57 0 [Note] InnoDB: Removed temporary tablespace data file: "./ibtmp1" 2021-12-09 9:56:57 0 [Note] InnoDB: Shutdown completed; log sequence number 42602; transaction id 15 2021-12-09 9:56:57 0 [Note] sql/mysqld: Shutdown complete
2021-12-03 12:12:14 +11:00
std::ostream &operator<<(std::ostream &lhs, const bytes_iec &rhs)
{
static const char *sizes[]= {"B", "KiB", "MiB", "GiB", "TiB", "PiB",
"EiB", "ZiB", "YiB"};
size_t i= 0;
double d= rhs.get_double();
for (; d > 512.0 && i < array_elements(sizes); i++, d/= 1024.0);
lhs.precision(3);
lhs << std::fixed << d << sizes[i];
return lhs;
}
ATTRIBUTE_COLD logger& logger::operator<<(dberr_t err)
{
m_oss << ut_strerr(err);
return *this;
}
info::~info()
{
sql_print_information("InnoDB: %s", m_oss.str().c_str());
}
warn::~warn()
{
sql_print_warning("InnoDB: %s", m_oss.str().c_str());
}
/** true if error::~error() was invoked, false otherwise */
bool error::logged;
error::~error()
{
sql_print_error("InnoDB: %s", m_oss.str().c_str());
logged = true;
}
#ifdef _MSC_VER
/* disable warning
"ib::fatal::~fatal': destructor never returns, potential memory leak"
on Windows.
*/
#pragma warning (push)
#pragma warning (disable : 4722)
#endif
ATTRIBUTE_NORETURN
fatal::~fatal()
{
sql_print_error("[FATAL] InnoDB: %s", m_oss.str().c_str());
abort();
}
#ifdef _MSC_VER
#pragma warning (pop)
#endif
error_or_warn::~error_or_warn()
{
if (m_error) {
sql_print_error("InnoDB: %s", m_oss.str().c_str());
} else {
sql_print_warning("InnoDB: %s", m_oss.str().c_str());
}
}
fatal_or_error::~fatal_or_error()
{
sql_print_error(m_fatal ? "[FATAL] InnoDB: %s" : "InnoDB: %s",
m_oss.str().c_str());
if (m_fatal) {
abort();
}
}
} // namespace ib
#ifndef DBUG_OFF
static char dbug_print_buf[1024];
const char * dbug_print_rec(const rec_t* rec, const rec_offs* offsets)
{
rec_printer r(rec, offsets);
strmake(dbug_print_buf, r.str().c_str(), sizeof(dbug_print_buf) - 1);
return dbug_print_buf;
}
const char * dbug_print_rec(const rec_t* rec, ulint info, const rec_offs* offsets)
{
rec_printer r(rec, info, offsets);
strmake(dbug_print_buf, r.str().c_str(), sizeof(dbug_print_buf) - 1);
return dbug_print_buf;
}
const char * dbug_print_rec(const dtuple_t* tuple)
{
rec_printer r(tuple);
strmake(dbug_print_buf, r.str().c_str(), sizeof(dbug_print_buf) - 1);
return dbug_print_buf;
}
const char * dbug_print_rec(const dfield_t* field, ulint n)
{
rec_printer r(field, n);
strmake(dbug_print_buf, r.str().c_str(), sizeof(dbug_print_buf) - 1);
return dbug_print_buf;
}
const char * dbug_print_rec(const rec_t* rec, dict_index_t* index)
{
rec_offs offsets_[REC_OFFS_NORMAL_SIZE];
rec_offs* offsets = offsets_;
rec_offs_init(offsets_);
mem_heap_t* tmp_heap = NULL;
offsets = rec_get_offsets(rec, index, offsets, index->n_core_fields,
ULINT_UNDEFINED, &tmp_heap);
rec_printer r(rec, offsets);
strmake(dbug_print_buf, r.str().c_str(), sizeof(dbug_print_buf) - 1);
return dbug_print_buf;
}
#endif /* !DBUG_OFF */
#endif /* !UNIV_INNOCHECKSUM */