[ZendCon 2013] MySQL HA, Recovery and Load Balancing

Using MySQL Server and Percona XtraDB Cluster we will take a look at several possible replication configurations for MySQL high-availability, disaster recovery and load-balancing. After comparing them for administration ease, scalability of reads and writes, robustness of the system for errors and time for data recovery as well as looking at how they impact your code-layer we will discuss the ins-and-outs of each setup, looking at the pros and cons, and what trade-offs you will need to make depending on your needs.

Transcript

1. MySQL HA, DR and
   Load Balancing
   Thursday, October 10, 13
   

   View Slide

2. •Community Engineer at Engine
   Yard
   •Author of Zend PHP 5
   Certification Study Guide,
   Sitepoints PHP Anthology: 101
   Essential Tips, Tricks & Hacks
   & PHP Master: Write Cutting
   Edge Code
   •A contributor to Zend
   Framework 1 & 2, phpdoc, and
   PHP internals
   •@dshafik
   Davey Shafik
   Thursday, October 10, 13
   

   View Slide

3. High Availability
   Thursday, October 10, 13
   

   View Slide

4. High Availability (HA)
   • Not: servers that don’t die
   • Ensures service availability, not server availability
   4
   Thursday, October 10, 13
   

   View Slide

5. Disaster Recovery
   Thursday, October 10, 13
   

   View Slide

6. Disaster Recovery (DR)
   • Process for recovering from different failure
   scenarios
   • Automated or not (or in-between)
   • Planning
   • Something to recover from
   • Something to recover to
   6
   Thursday, October 10, 13
   

   View Slide

7. Load Balancing
   Thursday, October 10, 13
   

   View Slide

8. Load Balancing
   • Balancing the load between multiple machines
   • Smart (based on system resources)
   • Dumb (round-robin)
   • ELB is easy, but slow
   • HAProxy is super faster (and still pretty easy)
   8
   Thursday, October 10, 13
   

   View Slide

9. Why?
   Thursday, October 10, 13
   

   View Slide

10. Why?
    • Curiosity — MySQL admin is fun
    • Goal: Make system predictable — then it can be
    automated (some)
    • Impacts of design decisions
    • Percona Cluster: oooh shiny!
    10
    Thursday, October 10, 13
    

    View Slide

11. What
    Thursday, October 10, 13
    

    View Slide

12. What
    • HA + Performance
    • Minimal code impact (drop-in system)
    • Simple (less moving parts)
    12
    Thursday, October 10, 13
    

    View Slide

13. MySQL 5.5
    Thursday, October 10, 13
    

    View Slide

14. Semi-Sync Replication
    • Improved Data Integrity
    • Master blocks until 1 slave has written to relay log,
    flushed to disk & acknowledged
    • Slower... a little.
    • Failover to asynchronous replication
    14
    Thursday, October 10, 13
    

    View Slide

15. Thursday, October 10, 13
    

    View Slide

16. Semi-Sync Master Setup
    INSTALL PLUGIN rpl_semi_sync_master SONAME
    'semisync_master.so';
    rpl_semi_sync_master_enabled = {0|1};
    rpl_semi_sync_master_timeout = N;
    Thursday, October 10, 13
    

    View Slide

17. Semi-Sync Slave Setup
    INSTALL PLUGIN rpl_semi_sync_slave SONAME
    'semisync_slave.so';
    rpl_semi_sync_slave_enabled = {0|1};
    Thursday, October 10, 13
    

    View Slide

18. Disaster Recovery
    Thursday, October 10, 13
    

    View Slide

19. Disaster Recovery
    • Primary slave is pre-determined failover master
    • Will have the most up-to-date content (not necessarily ALL the
    content)
    • Slaves must re-connect to new master
    • CHANGE MASTER to correct (new) position
    • Switch app to new master
    19
    Thursday, October 10, 13
    

    View Slide

20. Performance
    Thursday, October 10, 13
    

    View Slide

21. Performance
    • Benchmarker: 1 XL AWS Instance
    • Master: 1 AWS Medium Instance
    • Slaves: 3 AWS Large Instances (one semi-sync)
    • HAProxy: 1 AWS Large Instance
    21
    Thursday, October 10, 13
    

    View Slide

22. Benchmarks
    • Sysbench OLTP Benchmark
    • 1.5 Million Rows
    • 64 Connection
    • InnoDB Storage Engine
    • 50,000 Queries
    • Optionally: Read-only
    22
    Thursday, October 10, 13
    

    View Slide

23. Performance
    • Writing 1.5 Million Records, with indexes:
    • 1m 13s seconds (replicated)
    • Sysbench OLTP Benchmark:
    • Read/Write (with semi-sync): 2850/qps
    • Read/Write (without semi-sync): 2900/qps
    • Read only: 10850/qps
    23
    Thursday, October 10, 13
    

    View Slide

24. MySQL 5.6
    Thursday, October 10, 13
    

    View Slide

25. Global Transaction IDs (GTIDs)
    • Unique identifier across all server (binlogs)
    • combines server UUID:Transaction ID
    • Transaction based
    • No need for log files or positions with CHANGE
    MASTER
    • Consistent group-wide
    • Easier failover
    25
    Thursday, October 10, 13
    

    View Slide

26. Thursday, October 10, 13
    

    View Slide

27. GTID Master & Slave Setup
    gtid-mode = ON
    log-bin
    enforce-gtid-consistency = ON
    log-slave-updates = ON
    Thursday, October 10, 13
    

    View Slide

28. GTID Slave Setup
    • STOP SLAVE;
    • CHANGE MASTER TO MASTER_HOST=' ',
    MASTER_USER='',
    MASTER_PASSWORD='',
    MASTER_AUTO_POSITION=1;
    • START SLAVE;
    28
    Thursday, October 10, 13
    

    View Slide

29. Disaster Recovery
    Thursday, October 10, 13
    

    View Slide

30. Disaster Recovery: mysqlfailover
    • Performs replication health monitoring
    • handles automatic failover
    • automatic !== immediate (interval in seconds >= 5s)
    • Hook scripts for failed check, before/after failover,
    after complete recovery,
    30
    Thursday, October 10, 13
    

    View Slide

31. Hooks
    • Failed check:
    • Alert somebody
    • Set app to maintenance
    • Before failover:
    • Alert somebody
    • Log Something
    • Take backups
    31
    Thursday, October 10, 13
    

    View Slide

32. Hooks
    • After Failover:
    • Alert Somebody
    • After complete recovery:
    • Modify Application Config
    • Switchover elastic IPs (amazon)
    • Verify Application
    • Alert Somebody!
    32
    Thursday, October 10, 13
    

    View Slide

33. Performance
    Thursday, October 10, 13
    

    View Slide

34. Performance
    • Benchmarker: 1 XL AWS Instance
    • Master: 1 AWS Medium Instance
    • Slaves: 3 AWS Large Instances
    • HAProxy: 1 AWS Large Instance
    34
    Thursday, October 10, 13
    

    View Slide

35. Performance
    • Writing 1.5 Million Records, with indexes:
    • 60 seconds (replicated)
    • Sysbench OLTP Benchmark:
    • Read/Write: 2950/qps
    • Read only: 9500/qps
    35
    Thursday, October 10, 13
    

    View Slide

36. Percona XtraDB Cluster
    Thursday, October 10, 13
    

    View Slide

37. Multi-Master Replication
    • Everything replicates to everything
    • Uses XtraDB table engine (innodb compatible)
    — not NDB
    • Automatically transfer state to newly introduced
    nodes
    • Still has a “primary” server
    • Based on Galera for multi-master writes
    37
    Thursday, October 10, 13
    

    View Slide

38. Thursday, October 10, 13
    

    View Slide

39. Primary Server Setup
    [mysqld]
    wsrep_cluster_address=gcomm://
    datadir=/var/lib/mysql
    user=mysql
    binlog_format=ROW
    wsrep_provider=/usr/lib/libgalera_smm.so
    wsrep_slave_threads=2
    wsrep_cluster_name=benchmarking
    wsrep_sst_method=xtrabackup
    wsrep_node_name=
    innodb_locks_unsafe_for_binlog=1
    innodb_autoinc_lock_mode=2
    Thursday, October 10, 13
    

    View Slide

40. Other Server Setup
    [mysqld]
    wsrep_cluster_address=gcomm://
    datadir=/var/lib/mysql
    user=mysql
    binlog_format=ROW
    wsrep_provider=/usr/lib/libgalera_smm.so
    wsrep_slave_threads=2
    wsrep_cluster_name=benchmarking
    wsrep_sst_method=xtrabackup
    wsrep_node_name=
    innodb_locks_unsafe_for_binlog=1
    innodb_autoinc_lock_mode=2
    Thursday, October 10, 13
    

    View Slide

41. Disaster Recovery
    Thursday, October 10, 13
    

    View Slide

42. Disaster Recovery
    • Choose a new Primary server
    • Update config
    • Restart all servers
    42
    Thursday, October 10, 13
    

    View Slide

43. Performance
    Thursday, October 10, 13
    

    View Slide

44. Performance
    • Unable to test!
    • More than 4 concurrent “users” resulted in auto
    increment primary key collisions
    • Unable to complete the benchmark at all with >= 16
    threads
    • ALERT: failed to execute mysql_stmt_execute():
    Err1062 Duplicate entry
    '500519' for key 'PRIMARY'
    FATAL: database error, exiting...
    44
    Thursday, October 10, 13
    

    View Slide

45. Other Issues
    Thursday, October 10, 13
    

    View Slide

46. Other Issues
    • Replication was very brittle, restarting the daemon
    would sometimes break it
    • Restarting the server almost always break it
    • Adding new nodes very often failed
    • Completely unable to get the system up and running
    with the latest version for this talk.
    46
    Thursday, October 10, 13
    

    View Slide

47. Read/Write Splitting
    mysqlnd_ms
    Thursday, October 10, 13
    

    View Slide

48. Read/Write Splitting
    $ pecl install mysqlnd_ms
    mysqlnd_ms.enable=1
    mysqlnd_ms.config_file=
    /path/to/mysqlnd_ms.json
    Thursday, October 10, 13
    

    View Slide

49. Read/Write Splitting
    {
    "myapp": {
    "master": {
    "master_0": {
    "host": "localhost",
    "socket": "\/tmp\/mysql.sock"
    }
    },
    "slave": {
    "slave_0": {
    "host": "192.168.2.27",
    "port": "3306"
    }
    }
    }
    }
    Thursday, October 10, 13
    

    View Slide

50. Read/Write Splitting
    SELECT * FROM foo; -- Use Slave
    /* MYSQLND_MS_SLAVE_SWITCH */SELECT * FROM foo; -- Use Slave
    SELECT * FROM foo; INSERT INTO bar VALUES ('baz'); -- Use Slave!
    /* MYSQLND_MS_MASTER_SWITCH */SELECT * FROM foo; -- Use Master
    /* MYSQLND_MS_LAST_USED_SWITCH */SELECT * FROM foo; -- Use whatever
    -- was last used
    Thursday, October 10, 13
    

    View Slide

51. Other Resources
    • Semi-Sync Replication:
    • http://ey.io/18PllQP
    • Replication with GTIDs:
    • http://ey.io/15nTo20
    • MySQL Utilities (include mysqlfailover):
    • http://ey.io/GGP2Kh
    • Percona XtraDB Cluster:
    • http://ey.io/1gqyyUn
    51
    http://ey.io/zc13-mysql
    Thursday, October 10, 13
    

    View Slide

52. Feedback & Questions:
    Feedback: https://joind.in/9093
    Twitter: @dshafik
    Email: davey@engineyard.com
    http://ey.io/zc13-mysql
    Thursday, October 10, 13
    

    View Slide