Plmce 14 be a_hero_16x9_final
1 like829 views
The document summarizes a presentation given by Marco Tusa and Aleksandr Kuzminsky on using the Percona Data Recovery Tool for InnoDB to recover data from corrupted or missing InnoDB files. It introduces the speakers and outlines how the tool can be used to extract data from ibdata files, attach .ibd files after corruption, recover deleted records, recover dropped tables, and recover from wrong updates by parsing binary logs. Examples are provided of running the tool's page_parser, constraints_parser and sys_parser components to extract schema and data.
![How recover Drop tables -
ā¢ Different methods if using innodb_file_per_table=[0|1].
ā Must act fast because files can be overwritten quickly
ā¢ In the first case pages are mark free for reuse
ā¢ In the second the file is removed and we need to scan
the device
28](https://image.slidesharecdn.com/plmce14beahero16x9final-140511135605-phpapp02/85/Plmce-14-be-a_hero_16x9_final-27-320.jpg)
Plmce 14 be a_hero_16x9_final
- 1. Percona Live Santa Clara 2014 Be the hero of the day with Data recovery for InnoDB Marco Tusa ā Aleksandr Kuzminsky April 2014
- 2. Who ā¢ Marco āThe Grinchā ā¢ Manager Rapid Response ā¢ Former Pythian MySQL cluster technical leader ā¢ Former MySQL AB PS (EMEA) ā¢ Love programming ā¢ History of religions ā¢ Ski; Snowboard; scuba diving; Mountain trekking 2
- 3. What we will cover ā¢ Recovery toolkit introduction ā¢ Show how to extract data from IBD data file ā¢ Attach ibd files after IBDATA corruption ā¢ Recover deleted records ā¢ Recover drop table ā¢ Bad update 3
- 4. What is Percona Data recovery Tool for InnoDB ā¢ Set of open source tools ā¢ Work directly on the data files ā¢ Recover lost data (no backup available) ā¢ Wrappers (you can help) 4 A
- 5. What files? ā¢ Server wide files ā <Table>.frm ā¢ InnoDB files ā ibdata1 ā¢ InnoDB dictionary ā¢ UNDO segment ā¢ All tables if innodb_file_per_table=OFF ā <Table>.ibd ā Reads raw partition 5 A
- 6. InnoDB file format ā¢ Antelope ā REDUNDANT (versions 4.x) ā COMPACT (5.X) ā¢ Barracuda ā REDUNDANT, COMPACT ā New time formats in BARRACUDA ā COMPRESSED 6 A
- 7. What is a InnoDB Table space? Tablespace consists of pages ā¢ InnoDB page ā 16k by default ā page_id is file offset in 16k chunks ā Records never fragmented ā Types: ā¢ FIL_PAGE_INDEX ā¢ FIL_PAGE_TYPE_BLOB ā¢ FIL_PAGE_UNDO_LOG 7 A
- 8. InnoDB index (B+ Tree) 8 A
- 9. Requirements and minimal skill to use the tool ā¢ You need to know how to compile (make) ā¢ MySQL you know what it is right? ā¢ How to import data from a tab separated file 9 M
- 10. Show the tool in action - Process Process: 10 data extraction Extract data From ibdataX Read SYS_X Tables Generate Table filters files Extract data from ibd Tbspaces Validate Data Import data back Final clean up Production ready
- 11. Show the tool in action - page_parser ā¢ Extract pages from InnoDB files ā (In case of innodb_file_per_table =0 it also extract real data) ā page_parser -f ibdata1 (or table space like employees.ibd) 11 Extract data from idbata
- 12. Show the tool in action ā contraints_parser ā¢ Extract data from InnoDB pages ā¢ IE SYS_TABLE/INDEX/COLUMN ā bin/constraints_parser.SYS_TABLES -4Uf FIL_PAGE_INDEX/0-1 ā bin/constraints_parser.SYS_INDEXES -4Uf FIL_PAGE_INDEX/0-3 12 Extract data from idbata
- 13. Show the tool in action - contraints_parser Output SYS_TABLES "employees/salaries" 811 ļ ā¦ SYS_INDEXES 811 1889 "PRIMARYā ļ SYS_INDEXES 811 1890 "emp_no" Table ID Index Id 13 Read SYS_Tables/Indexes
- 14. Show the tool in action - sys_parser Why: Lost FRM file Two possible ways: ā¢ Easy: copy it from slave/backup ā¢ Less easy: ā Run sys_parser on a new create instance (info not accessible ā require dictionary table load) 14 Lost FRM and IBD (1)
- 15. Show the tool in action - sys_parser Output: ./sys_parser -h192.168.0.35 -u stress -p tool ād <database> employees/salaries CREATE TABLE `salariesR`( `emp_no` INT NOT NULL, `salary` INT NOT NULL, `from_date` DATE NOT NULL, `to_date` DATE NOT NULL, PRIMARY KEY (`emp_no`, `from_date`)) ENGINE=InnoDB; 15 Lost FRM and IBD (2)
- 16. Show the tool in action - ibdconnect ā¢ Accidental removal of the IBDATA file ā¢ IBDATA table space corruption ā¢ Only file per table available (IE employees.ibd) 16 Attach Table from another source (1)
- 17. Show the tool in action - ibdconnect What to do? 1. Start a new clean MySQL 2. Create empty structure (same tables definition) 3. Copy over the table spaces 4. Run ibdconnect 5. Run innochecksum_changer 17 Attach Table from another source (2) ./ibdconnect -o ibdata1 -f salaries.ibd -d employees -t salaries salaries.ibd belongs to space #15 Initializing table definitions... Updating employees/salaries (table_id 797) SYS_TABLES is updated successfully Initializing table definitions... Processing table: SYS_TABLES ā¦ Processing table: SYS_INDEXES Setting SPACE=15 in SYS_INDEXES for TABLE_ID = 797
- 18. Show the tool in action ā fix filters Table filters use for: ā¢ Identify the data inside the ibd ā¢ Filter out the damage records ā¢ Bound to table definition ā¢ Must recompile for each table definition ā¢ Generated with the tool create_def.pl 18 Generate Table filters (1)
- 19. Show the tool in action - page_parser filters ā¢ Generated with the tool create_def.pl create_defs.pl --db=$schema --table=$table > $myPath/include/table_defs.${schema}.$table.defreco very ā¢ Create symbolic link to include/table_defs.h ā¢ Compile again 19 Generate Table filters (1)
- 20. Show the tool in action - constraints_parser The data is extracted by the tool specifying the table space an possible BLOB directory. /constraints_parser -5Uf FIL_PAGE_INDEX/0-${INDEXID} āb FIL_PAGE_TYPE_BLOB/ > $DESTDIR/${SCHEMA}_${TABLE}.csvā FIL_PAGE_INDEX/0-${INDEXID} is the ID of the PK FIL_PAGE_TYPE_BLOB is the directory containing the BLOB 21 Extract data from Table space (1)
- 21. Show the tool in action - constraints_parser Example of the data: -- Page id: 4, Format: COMPACT, Records list: Valid, Expected records: (164 164) 00000000150B 88000002260084 employees 10001 "1953-09-02ā "G" "eorgiF" "(null)" "12722-11-12" 00000000150B 88000002260091 employees 10002 "1964-06-02" "B" "ezalelSā "(null)" "14006-11-05" 00000000150B 8800000226009E employees 10003 "1959-12-03" "P" "artoB" "(null)" "14003-03-15" 00000000150B 880000022600AB employees 10004 "1954-05-01" "C" "hirstianK""(null)ā "12598-03-09" 00000000150B 880000022600B8 employees 10005 "1955-01-21" "K" "yoichiM" "(null)" "13876-11-14" <snip> 00000000150B 880000022608EE employees 10164 "1956-01-19" "J" "agodaB" "(null)" "12474-11-14" -- Page id: 4, Found records: 164, Lost records: NO, Leaf page: YES 22 Validate data
- 22. Show the tool in action - LOAD DATA INFILE How to import the data back? Easy as: LOAD DATA INFILE āPLMC_employees/employees' REPLACE INTO TABLE `employees` FIELDS TERMINATED BY 't' OPTIONALLY ENCLOSED BY '"' LINES STARTING BY 'employeest' (`emp_no`, `birth_date`, `first_name`, `last_name`, `gender`, `hire_date`); Done 23 Import data back
- 23. How recover deleted record - Identify the records just for this exercise: select count(emp_no) from employeesR where hire_date > '1999-08-24'; +---------------+ | count(emp_no) | +---------------+ | 279 | +---------------+ 1 row in set (0.23 sec) And bravely delete them: delete from employeesR where hire_date > '1999-08-24'; Query OK, 279 rows affected (0.55 sec) 24 Delete records
- 24. How recover deleted record - To recover deleted record we must use the āD flag: constraints_parser -5Df /FIL_PAGE_INDEX/0-1975 -b /FIL_PAGE_TYPE_BLOB/ > employees_employeesDeleted.csv cat employees_employeesDeleted.csv |grep -i -v -e "-- Page id"|wc -l 55680 ļ Too many because I take unfiltered records 25 Recover delete records
- 25. How recover deleted record - name: "employeesR", { { /* int(11) */ name: "emp_no", type: FT_INT, fixed_length: 4, has_limits: TRUE, limits: { can_be_null: FALSE, int_min_val: 10001, int_max_val: 499999 }, 26 Use filters to clean up results name: "first_name", type: FT_CHAR, min_length: 0, max_length: 42, has_limits: TRUE, limits: { can_be_null: FALSE, char_min_len: 3, char_max_len: 42, char_ascii_only: TRUE }, can_be_null: FALSE }, name: "last_name", type: FT_CHAR, min_length: 0, max_length: 48, has_limits: TRUE, limits: { can_be_null: FALSE, char_min_len: 3, char_max_len: 48, char_ascii_only: TRUE },
- 26. How recover deleted record - Now let us recompile and run the extraction again: cat employees_employeesDeleted.csv |grep -i -v -e "-- Page id"|wc -l 279 <------ Bingo! 27 Check if it fits and reload
- 27. How recover Drop tables - ā¢ Different methods if using innodb_file_per_table=[0|1]. ā Must act fast because files can be overwritten quickly ā¢ In the first case pages are mark free for reuse ā¢ In the second the file is removed and we need to scan the device 28
- 28. How recover Drop tables - What we need then ? ā¢ The table definition ā¢ The PK index ā Parse the dictionary with the āDā flag ā¢ constraints_parser.SYS_TABLES -4D ā¢ Extract the InnoDB pages 29
- 29. How recover Drop tables - Not using innodb_file_per_table method: 1. Extract the ibdataX as usual 2. Run contraints_parser constraints_parser -5Uf ./FIL_PAGE_INDEX/0-1975 -b ./FIL_PAGE_TYPE_BLOB/ > employees_employeesDroped.csv cat employees_employeesDroped.csv |grep -i -v -e "-- Page id"|wc -l 300024 <---- done 30 Not using file per table
- 30. How recover Drop tables - Innodb_file_per_table=1 method: What we need more? ā¢ To know what device is containing the dropped table 31 using file per table
- 31. How recover Drop tables - Identify the PK id from dictionary: cat SYS_TABLE.csv|grep employeesR SYS_TABLES "employees/employeesR" 855 cat SYS_INDEX.csv|grep 855 SYS_INDEXES 855 1979 "PRIMARYā 32 using file per table
- 32. How recover Drop tables - This time we must run the page_parser against the device not the file using the T option. -T -- retrieves only pages with index id = NM (N - high word, M - low word of id) page_parser -t 100G -T 0:1979 -f /dev/mapper/EXT_VG-extlv Parse in this case take longer. 33 Run the page extraction
- 33. How recover wrong updates - Doing with the tool is possible when new data is larger then the original, and will not fit in the original page, otherwise the old one will be replaced, as such the only way is to parse the undo segment. Tools: s_indexes, s_tables recover the dictionary from the undo. 34
- 34. How recover wrong updates - Other method: Possible to use the binary log for that When in ROW format . 35
- 35. How recover wrong updates - You can use binary log to recover your data if: ā¢ Binlog format = ROW ā¢ binlog_row_image = FULL (from 5.6 you can change it) 36 Prerequisite
- 36. How recover wrong updates - Assume a table like : +--------+------------+------------+-----------+--------+------------+ | emp_no | birth_date | first_name | last_name | gender | hire_date | +--------+------------+------------+-----------+--------+------------+ | 10001 | 1953-09-02 | Georgi | Facello | M | 1986-06-26 | | 10002 | 1964-06-02 | Bezalel | Simmel | F | 1985-11-21 | | 10003 | 1959-12-03 | Parto | Bamford | M | 1986-08-28 | +--------+------------+------------+-----------+--------+------------+ An action like : update employeesR set last_name="WRONG-NAME" where emp_no < 10010; 37 Scenario (1)
- 37. How recover wrong updates - You will have to recover something like: +--------+------------+------------+------------+--------+------------+ | emp_no | birth_date | first_name | last_name | gender | hire_date | +--------+------------+------------+------------+--------+------------+ | 10001 | 1953-09-02 | Georgi | WRONG-NAME | M | 1986-06-26 | | 10002 | 1964-06-02 | Bezalel | WRONG-NAME | F | 1985-11-21 | | 10003 | 1959-12-03 | Parto | WRONG-NAME | M | 1986-08-28 | ā¦ 38 Scenario (2)
- 38. How recover wrong updates - with a simple command like : mysqlbinlog -vvv logs/binlog.000034 --start-datetime="2014-03-19 11:00:07"|grep -e "@1" -e "@4"|awk -F '/*' '{print $1}'|awk '{print $2}' @1=10001 @4='Facello' @1=10001 @4='WRONG-NAME' @1=10002 @4='Simmel' @1=10002 @4='WRONG-NAME'.. 39 Recover from binary log (2)
- 39. Reference and repositories Main Percona branch: bzr branch lp:percona-data-recovery-tool-for-innodb Marco branch: https://github.com/Tusamarco/drtools 40 Recover from binary log (2)
- 40. Q&A 41
- 41. Contacts 42 To contact Marco marco.tusa@percona.com marcotusa@tusacentral.net To follow me http://www.tusacentral.net/ https://www.facebook.com/marco.tusa.94 @marcotusa http://it.linkedin.com/in/marcotusa/ To contact Aleksander aleksandr.kuzminsky@percona.com To follow me http://www.mysqlperformanceblog.com/author/akuzminsk y/ https://www.linkedin.com/in/akuzminsky