Diagnosing Failures in MySQL Replication

Devananda van der Veen Percona Live London 2011

Introduction

-2-

About Me Sr Consultant at Percona since 2009 Worked with MySQL replication since 2004 Live in NW Washington About Percona Support, Consulting, Training, Development We create & maintain Percona-Server w/ XtraDB and XtraBackup Many other tools too: maatkit & aspersa percona-toolkit

www.percona.com

Table of Contents

-3-

Introduce the Problems Replication Architecture Review Masters, Slaves, and the many files they keep Different ways things break How to diagnose the problems Knowing when to skip it, when and how to fix it, and when to just rebuild it. Questions (if we have time)

www.percona.com

Replication is Ubiquitous

-4-

Who doesn't use MySQL Replication for...

Load distribution Redundancy (local or geographic) Master-Master fail-over Taking backups from slave Archiving & Data Warehousing Many other things too

www.percona.com

Ubiquitous... but not perfect

Google mysql replication failure 130,000+ results

-5-

www.percona.com

Ubiquitous... but not perfect

MySQL lists 51 Active bugs in Replication as of October 15, 2011
26945 48062 43457 58637 62557

-6-

Explicit TEMP TABLE causes replication failure if mysqld restarted Date arithmetic in Stored Procedure breaks replication replicate-ignore-db behaves differently with different binlog formats INSERT..ON DUP KEY UPDATE unsafe if more than one unique key SHOW SLAVE STATUS gives wrong output with master-master and using SET uservars

http://bit.ly/MysqlRepBugs

www.percona.com

Most problems have a Solution

Here are some specific problems that I'll discuss (time permitting)

-7-

Duplicate key errors Temp table does not exist after slave restart Binary and relay log corruption Slave reads from wrong position after crash And more...

www.percona.com

But first
We need to know a little about how MySQL replication works

-8-

Master & Slave threads Binary & Relay log files Status files (or status tables in 5.6)

If you want to know more, see Lars Thalmann's presentations http://bit.ly/LarsRepTalk http://bit.ly/LarsRepTalk1

www.percona.com

A Brief Review
Write to Keeps list at

-9-

Master Host
master-bin.xxx master-bin.index

(binary file) (plain-text file)

Composed of two threads: IO + SQL

IO_Thread SQL_Thread

Slave Host

Status file Reads from Writes to Local list

Status file master.info master-bin.xxx Reads from relay-bin.xxx relay-bin.index

relay.info relay-bin.xxx

http://bit.ly/5-5-slave-log-status
www.percona.com

The Index Files

# cat master-bin.index ./master-bin.000001 ./master-bin.000002 # cat relay-bin.index ./relay-bin.000001 ./relay-bin.000002 Master binary log index file

-10-

Slave relay log index file

You sometimes need to edit these files manually (but only in dire situations)

www.percona.com

The Info Files

Slave:$ cat master.info # of lines in file 15 Master_Log_File master-bin.000001 Read_Master_Log_Pos 1818 Master_Host 127.0.0.1 Master_User repl_user repl_pass (plain-text!!) Master_Password Master_Port 19501 Connect_Retry 60 ... ...

-11-

current position of io_thd on master

www.percona.com

The Info Files (2)

-12-

local read position Slave:$ cat relay-log.info of sql_thd ./relay-bin.000002 Relay_Log_File 1392 Relay_Log_Pos master-bin.000001 Relay_Master_Log_File 1818 Exec_Master_Log_Pos remote read position of sql_thd

These positions correspond to the end_log_pos of the last statement executed by sql_thread

www.percona.com

mysql> show slave status\G

Slave_IO_State: Waiting for master Slave_IO_Running: Yes Slave_SQL_Running: Yes Master_Log_File: master-bin.000001 Read_Master_Log_Pos: 1818 Relay_Log_File: relay-bin.000001 Relay_Log_Pos: 251 Relay_Master_Log_File: master-bin.000001 Exec_Master_Log_Pos: 1818
io_thread remote pos sql_thread local pos sql_thread remote pos

Fields re-ordered for display www.percona.com

Running Example
~/sandboxes$ make_repl_sandbox --circular 2 5.1.55 Percona-Server-5.1.55-rel12.6-200-Linux-x86_64 Node1 = Active master | Node2 = Passive Master node1 > create table foo ( id int not null auto_increment primary key, v varchar(255) not null default '', unique (v)) engine=innodb; Query OK, 0 rows affected (0.01 sec) node1 > insert into foo (v) values ('a'), ('b'); Query OK, 2 rows affected (0.01 sec) Records: 2 Duplicates: 0 Warnings: 0

www.percona.com

mysqlbinlog on masters
# at 1610 #110318 14:20:12 server id 101 end_log_pos 1683 Query thread_id=8 exec_time=0 error_code=0 SET TIMESTAMP=1300483212/*!*/; BEGIN/*!*/; # at 1683 #110318 14:20:12 server id 101 end_log_pos 1791 Query thread_id=8 exec_time=0 error_code=0 SET TIMESTAMP=1300483212/*!*/; INSERT INTO `foo` VALUES (1,'a'),(2,'b')/*!*/; # at 1791 #110318 14:20:12 server id 101 Xid = 32 COMMIT/*!*/; end_log_pos 1818

www.percona.com

mysqlbinlog on relay logs

# at 1184 #110318 14:20:12 server id 101 end_log_pos 1683 Query thread_id=8 exec_time=0 error_code=0 SET TIMESTAMP=1300483212/*!*/; BEGIN/*!*/; end_log_pos is copied from master! # at 1257 #110318 14:20:12 server id 101 end_log_pos 1791 Query thread_id=8 exec_time=0 error_code=0 SET TIMESTAMP=1300483212/*!*/; INSERT INTO `foo` VALUES (1,'a'),(2,'b')/*!*/; # at 1365 #110318 14:20:12 server id 101 Xid = 32 COMMIT/*!*/; end_log_pos 1818

www.percona.com

mysqlbinlog on slave binlog

# at 1610 #110318 14:20:12 server id 101 end_log_pos 1674 Query thread_id=8 exec_time=15 error_code=0 SET TIMESTAMP=1300483212/*!*/; BEGIN/*!*/; Here, exec_time means slave lag! # at 1674 #110318 14:20:12 server id 101 end_log_pos 1782 Query thread_id=8 exec_time=15 error_code=0 SET TIMESTAMP=1300483212/*!*/; INSERT INTO `foo` VALUES (1,'a'),(2,'b')/*!*/; # at 1782 #110318 14:20:12 server id 101 Xid = 35 COMMIT/*!*/; end_log_pos 1809

www.percona.com

End of the Review

Any Questions? (find me after the talk)

www.percona.com

Finally, the meat of the talk!

-19-

www.percona.com

My Approach
When replication stops...

-20-

Why did it stop? This usually tells you how to fix it Generally does not take much time Where did it break? Determine extent of corruption / data loss Were statements skipped or run twice? What is the business impact? Don't over-analyze. Estimate is probably OK at this step.

www.percona.com

My Approach
Now that you know why it stopped...

-21-

Choose your restore method Do you prioritize site availability or data integrity? Validate slave consistency If possible, before allowing traffic if necessary, use tools to resync the slave Have a plan B Sometimes you just need to rebuild

www.percona.com

Understand What Failed

Where and why the failure happens matters a lot!

Understand the basics. Know your architecture.

How you solve it is

Learned from experience Based on common principles Similar even for different organizations

www.percona.com

Two Times when Replication Fails

Server OK but replication stopped

-23-

Crash caused replication to stop

Writes on the slave File corruption in binary or relay-log Documented limitation in MySQL Un/known Bugs?

Hardware fails mysqld crash Kernel panic / OOM killer kill -9 `pidof mysqld` Host recently restarted and rolled back status file

www.percona.com

Two Times when Replication Fails

Server OK but replication stopped Dedicated Monitoring!! Seconds_Behind_Master lies. Watch IO and SQL thread state and position. Use a heartbeat table.

-24-

Crash caused replication to stop skip-slave-start innodb_overwrite_relay _log_info (only in Percona-Server) 5.6-labs features http://bit.ly/crash-safe-replication

www.percona.com

Common Situations
Documented limitations & bugs Writes on the slave File Corruption Hardware Failures

www.percona.com

Documented Limitations

www.percona.com

Documented limitations

-27-

Statement-Based Replication (SBR) has many limitations Documentation: http://bit.ly/5-5-rep-limits 33 subsections Still the default format, even in 5.5 Row-Based Replication (RBR) avoids most limitations But not all: http://bit.ly/5-5-rbr-limits All tables require a PRIMARY KEY Can be more difficult to do online alter table than SBR

www.percona.com

Documented limitations

-28-

Get to know the limitations & open bugs Some are documented as bug reports Avoid them like the plague Watch the patch notes like a hawk (even if you don't upgrade) Change your application as necessary

www.percona.com

Some limitations in SBR...

| Note | 1592 | Statement may not be safe to log in statement format.|

-29-

Non-Deterministic functions or statements UUID, NOW, CURRENT_USER, FOUND_ROWS UPDATE|DELETE with LIMIT but no ORDER BY TEMPORARY and MEMORY tables SET TX_ISOLATION = READ_COMMITTED Routines / Triggers with logical expressions Dynamic SQL And lots more ...

www.percona.com

Some limitations in RBR...

Can't read SQL in binary log Different DDL on master / slave breaks RBR (*) Slave calls fsync() for every row, not every transaction

(this is really a performance issue, not a limitation)

Mixing InnoDB and MyISAM in one transaction ... It's just not a good idea! Order of binlog events changed multiple times in 5.1

(*) Percona is working to enable slaves with different table definitions to replay RBR events that would normally fail.

www.percona.com

Writes on the Slave

(1) When it's just a slave (2) When it's also a master

www.percona.com

Write on Slave (1)

(when a slave is just a slave)

-32-

node2 > insert into foo (v) values ('bad insert'); node2 > select * from foo; | 1 | a | | 2 | b | | 3 | bad insert | node1 > insert into foo (v) values ('c'); node2 > show slave status\G Last_Errno: 1062 Last_Error: Error 'Duplicate entry '3' for key 'PRIMARY'' on query. Default database: 'test'. Query: 'insert into foo (v) values ('c')'

www.percona.com

Write on Slave (1)

(when a slave is just a slave) Sample binlog from slave (node2) #110328 13:55:52 server id 101 end_log_pos 640 Query thread_id=7 exec_time=0 error_code=0 INSERT INTO `foo` VALUES (1,'a'),(2,'b') #110328 13:56:34 server id 102 end_log_pos 877 Query thread_id=7 exec_time=0 error_code=0 SET TIMESTAMP=1301345794/*!*/; insert into foo (v) values ('bad insert')

-33-

www.percona.com

Write on Slave (1)

(when a slave is just a slave)

-34-

DELETE record from slave; optionally, insert to master may preserve data integrity insert to master allows you to keep the record more time-consuming than SKIP_COUNTER foreign keys and triggers make this very complicated!

www.percona.com

Write on Slave (1)

(when a slave is just a slave)

-35-

SET GLOBAL SQL_SLAVE_SKIP_COUNTER=1; easiest method data inconsistency may propagate to other tables, eg. if you use INSERT..SELECT or triggers must sync later with pt-table-sync

www.percona.com

Write on Slave (1)

(when a slave is just a slave)

-36-

auto_increment_increment & _offset would have prevented failure in this example safe to do even if you never write to slave master = even, slave = odd makes it trivial to identify bad writes - `id` will be odd

www.percona.com

Write on Slave (2)

(when a slave is also a master) node1 > insert into foo (v) values ('c'); node1 > SHOW SLAVE STATUS\G Error 'Duplicate entry '3' for key 'PRIMARY'' on query. Default database: 'test'. Query: 'insert into foo (v) values ('bad insert')'

-37-

node2 > insert into foo (v) values ('bad insert'); node2 > SHOW SLAVE STATUS\G Error 'Duplicate entry '3' for key 'PRIMARY'' on query. Default database: 'test'. Query: 'insert into foo (v) values ('c')'
Note: I used SLAVE STOP; on both hosts to simulate simultaneous writes

www.percona.com

Write on Slave (2)

(when a slave is also a master)

-38-

Fixing master-master is same principle as master-slave... But you might accidentally corrupt your primary master! So be extra careful...

For example: DELETE record from secondary; optionally, insert to primary Use SET SESSION SQL_LOG_BIN = 0 for DELETE But allow INSERT to replicate normally

www.percona.com

Write on Slave (2)

(when a slave is also a master)

-39-

SET SQL_SLAVE_SKIP_COUNTER = 1 On both masters Then sync with pt-table-sync auto_increment_increment & _offset A must have if you use AUTO_INC + master-master replication + fail-over

www.percona.com

Common mistake re: SKIP

There is a real open source project that [W]ill not only check on the health of your MySQL Replication setup, but it can also automatically heal and restart the MySQL Replication. Self-Healing MySQL Replication is not a new idea. http://bit.ly/replication-suicide but it's a really bad idea!!

www.percona.com

-41-

File Corruption

www.percona.com

File Corruption
Slave may stop if any of these files become corrupt Binary log | Relay log | Table data | Index data Examples of SHOW SLAVE STATUS
Last_IO_Error: 1236 Last_IO_Error: Got fatal error 1236 from master when reading data from binary log Last_SQL_Errno: 1594 Last_SQL_Error: Relay log read failure: Could not parse relay log event entry.

www.percona.com

If Master Binlog is Corrupt...

Verify it with mysqlbinlog [--verbose --base64] Find extent of corruption CHANGE MASTER TO Master_Log_Pos = <good_pos> Analyze bad section of log. Identify all affected tables. Use pt-table-sync when replication catches up. Resolve underlying (hardware) cause of the corruption!

www.percona.com

If Slave Relay Log is Corrupt...

Verify it with mysqlbinlog <relay-log-file> Check master binlog for corruption. If it is, see previous slide. Re-fetch the corrupted relay log CHANGE MASTER TO Master_Log_Pos = <Exec_master_log_pos>; Very rare If this occurs frequently, check for network problems.

www.percona.com

If a Table is Corrupt...

Stop replication and stop all traffic to that server Mount the file system read-only Check mysql error log - it may contain more info Even after fixing the corruption, pt-table-sync Sometimes, only option is restore from a backup

www.percona.com

Hardware Failures
A SAN is just a bigger SPoF -unknown

www.percona.com

HW Failure #1
Unplanned master restart all slaves stop with error: 'Client requested master to start replication from impossible position' Process: Compare each SLAVE STATUS to master binlog Realize that Exec_Master_Log_Pos different on each slave, and greater than master's log file size! Panic for a minute...

www.percona.com

HW Failure #1
Possible solutions: Promote slave that read the most Isolate extra events from slave binlog, replay on master Force slaves to resume from next binlog, then pt-table-sync to remove extra events Prevent with: RAID BBU + HDD cache disabled innodb_flush_log_at_trx_commit = 1 sync_binlog = 1

www.percona.com

HW Failure #2
Replaced a failed disk in a replica. After restart, replication fails with duplicate key error. Process: Check error log... there's no slave stop coordinates! Realize file system in read-only mode before shut down Was master.info rolled back by file system journal?

www.percona.com

HW Failure #2
Possible solutions: Guess where slave stopped, then CHANGE MASTER But what if data files also rolled back? SQL_SLAVE_SKIP_COUNTER + prayers + pt-table-sync Rebuild the slave Prevent with: Percona Server + innodb_overwrite_relay_log_info

www.percona.com

War Stories

-51-

Slave generates different auto_inc values for INSERT inside an ON_INSERT TRIGGER. This goes unnoticed for months, then starts causing problems. During MMM migration, a few uncommitted transactions get left behind and cause duplicate key errors on both nodes Major hardware failure in a geo-distributed three-node replication ring. Each host is written to by local processes. There are some tables only written on single host, but also some shared tables that all write to. How do you determine which of remaining two masters is good?

www.percona.com

Thank You to Our Sponsors

Platinum Sponsor

Gold Sponsor

Silver Sponsors

www.percona.com

Percona Live London Sponsors

Exhibitor Sponsors

Friends of Percona Sponsors

Media Sponsors

www.percona.com

Annual MySQL Users Conference

Presented by Percona Live
The Hyatt Regency Hotel, Santa Clara, CA April 10th-12th, 2012 Featured Speakers
Mark Callaghan, Facebook Jeremy Zawodny, Craigslist Marten Mickos, Eucalyptus Systems Sarah Novotny, Blue Gecko Peter Zaitsev, Percona Baron Schwartz, Percona

The Call for Papers is Now Open!

Visit www.percona.com/live/mysql-conference-2012/
www.percona.com

[email protected] devananda.vdv@gmail

We're Hiring! percona.com/about-us/careers/

www.percona.com/live