MySQL – Troubleshooting High CPU Usage for Single Query

MySQLmysql-5.5PHP

I recently moved from standard shared hosting to Digital Ocean.

While I've thoroughly enjoyed the experience – especially learning to manage some (relatively) large DB's – I've run into one major issue.

My Droplet on Digital Ocean has 2GB RAM and 2 CPU's. The only thing I'm running on the server is a fairly low-profile PHP app. However, one page in this app contains a rather extensive query that, at the moment, returns around 10k rows. There are a few joins involved (not all of them indexed).

Whenever I hit this page, it takes a good 60 seconds to load. Mysql's CPU usage shoots up to the server max:

I've used SQL Tuner, converted the tables to InnoDB (per a recommendation I read, though I have at most 25 concurrent users), made and tested a huge number of changes to my.cnf, and spent a significant number of hours trying to decipher this issue. However, I'm still getting the same results (and some frustrated users).

Is this problem 100% the fault of the query? Or, are there further changes I could make to my.cnf (or anywher else) that could help decrease the CPU load and increase the loading speed?

Thanks for any assistance!

MySQL Version information:

The my.cnf, including portions generated by Percona and recommendations made by SQL Tuner:

#
# The MySQL database server configuration file.
#
# You can copy this to one of:
# - "/etc/mysql/my.cnf" to set global options,
# - "~/.my.cnf" to set user-specific options.
# 
# One can use all long options that the program supports.
# Run program with --help to get a list of available options and with
# --print-defaults to see which it would actually understand and use.
#
# For explanations see
# http://dev.mysql.com/doc/mysql/en/server-system-variables.html

# This will be passed to all mysql clients
# It has been reported that passwords should be enclosed with ticks/quotes
# escpecially if they contain "#" chars...
# Remember to edit /etc/mysql/debian.cnf when changing the socket location.
[client]
port        = 3306
socket      = /var/run/mysqld/mysqld.sock

# Here is entries for some specific programs
# The following values assume you have at least 32M ram

# This was formally known as [safe_mysqld]. Both versions are currently parsed.
[mysqld_safe]
socket      = /var/run/mysqld/mysqld.sock
nice        = 0

[mysqld]
#
# * Basic Settings
#
user        = mysql
default_storage_engine         = InnoDB
pid-file    = /var/run/mysqld/mysqld.pid
socket      = /var/run/mysqld/mysqld.sock
port        = 3306
basedir     = /usr
datadir     = /var/lib/mysql
tmpdir      = /tmp
lc-messages-dir = /usr/share/mysql
skip-external-locking

# CACHES AND LIMITS #
tmp_table_size                 = 32M
max_heap_table_size            = 32M
query_cache_type               = 0
query_cache_size               = 0
max_connections                = 50
thread_cache_size              = 50
open_files_limit               = 65535
table_definition_cache         = 1024
table_open_cache               = 2048

# INNODB #
innodb_flush_method            = O_DIRECT
innodb_log_files_in_group      = 2
innodb_log_file_size           = 250M
innodb_flush_log_at_trx_commit = 1
innodb_file_per_table          = 1
innodb_buffer_pool_size        = 1000M
join_buffer_size = 250M

# LOGGING #
log_error                      = /var/log/mysql/error.log
log_queries_not_using_indexes  = 1
slow_query_log                 = 1
slow_query_log_file            = /var/log/mysql/mysql-slow.log
long_query_time                = 1


#
# Instead of skip-networking the default is now to listen only on
# localhost which is more compatible and is not less secure.
bind-address        = 127.0.0.1
#
# * Fine Tuning
#
key_buffer      = 16M
max_allowed_packet  = 16M
thread_stack        = 192K
thread_cache_size       = 8
# This replaces the startup script and checks MyISAM tables if needed
# the first time they are touched
myisam-recover         = BACKUP
#max_connections        = 100
#table_cache            = 64
#thread_concurrency     = 10
#
# * Query Cache Configuration
#
query_cache_limit   = 1M
query_cache_size        = 16M
#
# * Logging and Replication
#
# Both location gets rotated by the cronjob.
# Be aware that this log type is a performance killer.
# As of 5.1 you can enable the log at runtime!
#general_log_file        = /var/log/mysql/mysql.log
#general_log             = 1
#
# Error log - should be very few entries.
#
log_error = /var/log/mysql/error.log
#
# Here you can see queries with especially long duration
#log_slow_queries   = /var/log/mysql/mysql-slow.log
#long_query_time = 2
#log-queries-not-using-indexes
#
# The following can be used as easy to replay backup logs or for replication.
# note: if you are setting up a replication slave, see README.Debian about
#       other settings you may need to change.
#server-id      = 1
#log_bin            = /var/log/mysql/mysql-bin.log
expire_logs_days    = 10
max_binlog_size         = 100M
#binlog_do_db       = include_database_name
#binlog_ignore_db   = include_database_name
#
# * InnoDB
#
# InnoDB is enabled by default with a 10MB datafile in /var/lib/mysql/.
# Read the manual for more InnoDB related options. There are many!
#
# * Security Features
#
# Read the manual, too, if you want chroot!
# chroot = /var/lib/mysql/
#
# For generating SSL certificates I recommend the OpenSSL GUI "tinyca".
#
# ssl-ca=/etc/mysql/cacert.pem
# ssl-cert=/etc/mysql/server-cert.pem
# ssl-key=/etc/mysql/server-key.pem



[mysqldump]
quick
quote-names
max_allowed_packet  = 16M

[mysql]
#no-auto-rehash # faster start of mysql but no tab completition

[isamchk]
key_buffer      = 16M

#
# * IMPORTANT: Additional settings that can override those from this file!
#   The files must end with '.cnf', otherwise they'll be ignored.
#
!includedir /etc/mysql/conf.d/

Edit – In hindsight… I'm assuming it's this generated query. It takes 42s to run in PHPMyAdmin.

SELECT  x.*,
if(fielding_rating_pos2=0,'-',if(TRUNCATE((fielding_rating_pos2+40)/40,
                        0)<1,1,
                                        0)))) as fielding_rating_pos2, if(TRUNCATE((fielding_rating_pos2+40
  )/40,
                0)>5, 5,TRUNCATE((fielding_rating_pos2+40)/40,
if(fielding_rating_pos3=0,'-',if(TRUNCATE((fielding_rating_pos3+40)/40,
                        0)<1,1,
                                        0)))) as fielding_rating_pos3, if(fielding_rating_pos4=0,
'-',
if(TRUNCATE((fielding_rating_pos3+40)/40,0)>5, 5,TRUNCATE((fielding_rating_pos3+40)/40,
        0
  )<1,1,
        0)))) as fielding_rating_pos4, if(TRUNCATE((fielding_rating_pos4+40)/40,
if(fielding_rating_pos5=0,'-',if(TRUNCATE((fielding_rating_pos5+40)/40,
        0)<1,1,
0
  )>5, 5,
0)))) as fielding_rating_pos5, if(fielding_rating_pos6=0,'-',
if(TRUNCATE((fielding_rating_pos6+40)/40, 0)<1,1,if(TRUNCATE((fielding_rating_pos6+40)/40,
0
  )>5,
5,TRUNCATE((fielding_rating_pos6+40)/40,0)))) as fielding_rating_pos6,
TRUNCATE((fielding_rating_pos4+40)/40, 5,TRUNCATE((fielding_rating_pos5+40)/40,
if(fielding_rating_pos7=0,'-',if(TRUNCATE((fielding_rating_pos7+40)/40,
0)<1,1,
0)))) as fielding_rating_pos7, if(TRUNCATE((fielding_rating_pos5+40
  )/40,
0
  )>5,
'-',
0)>5, 5,TRUNCATE((fielding_rating_pos8+40)/40,0)))) as fielding_rating_pos8,
if(fielding_rating_pos9=0,'-',if(TRUNCATE((fielding_rating_pos9+40)/40,
if(TRUNCATE((fielding_rating_pos7+40)/40,0)>5, 5,TRUNCATE((fielding_rating_pos7+40)/40,
0)<1,1,
0)))) as fielding_rating_pos9, if(TRUNCATE((fielding_rating_pos4+40)/40,
if(fielding_rating_pos8=0, if(fielding_ratings_catcher_arm=0,
'-',
0
  )<1,1,
0)>5,5,TRUNCATE((fielding_ratings_catcher_arm+40)/40, 0)))) as fielding_ratings_catcher_arm,
if(fielding_ratings_infield_arm=0, '-',if(TRUNCATE((fielding_ratings_infield_arm+40)/40,
if(TRUNCATE((fielding_rating_pos9+40)/40,0)>5, 5,TRUNCATE((fielding_rating_pos9+40)/40,
0)<1,1,if(TRUNCATE((fielding_ratings_infield_arm+40)/40, 0)>5,
5,TRUNCATE((fielding_ratings_infield_arm+40)/40,0)))) as fielding_ratings_infield_arm,
if(fielding_ratings_outfield_arm=0,'-',if(TRUNCATE((fielding_ratings_outfield_arm+40)/40,
0)<1,1,if(TRUNCATE((fielding_ratings_outfield_arm+40)/40,0)>5,
5,TRUNCATE((fielding_ratings_outfield_arm+40)/40,0)))
  ) as fielding_ratings_outfield_arm FROM  players_fielding as pf,
if(TRUNCATE((fielding_ratings_catcher_arm+40)/40, if(TRUNCATE((fielding_rating_pos8+40)/40,
    ( if(TRUNCATE((fielding_rating_pos8+40)/40, 0)<1,1,if(TRUNCATE((fielding_ratings_catcher_arm+40)/40,
        SELECT  px.player_id,first_name,last_name,position,role,age,porder,
                slot,batting_ratings_talent_contact,batting_ratings_talent_gap,
                batting_ratings_talent_power,batting_ratings_talent_eye,
                batting_ratings_talent_strikeouts,running_ratings_speed,
        running_ratings_stealing,
                if(TRUNCATE((batting_ratings_talent_contact+40)/40, 0)<1,
                1,
                                0)>5,5,TRUNCATE((batting_ratings_talent_contact+40)/40,
                                0))) as batting_ratings_talent_contact,if(TRUNCATE((batting_ratings_talent_gap+40
                  )/40,
        0)<1,1,
        0)>5,5,TRUNCATE((batting_ratings_talent_gap+40)/40,0))
    ) as batting_ratings_talent_gap,
if(TRUNCATE((batting_ratings_talent_power+40)/40,0)<1, 1,if(TRUNCATE((batting_ratings_talent_power+40)/40,
0)>5, running_ratings_baserunning FROM ( SELECT  player_id, 5,
TRUNCATE((batting_ratings_talent_power+40)/40,0))) as batting_ratings_talent_power,
if(TRUNCATE((batting_ratings_talent_eye+40)/40,0)<1,1, if(TRUNCATE((batting_ratings_talent_contact+40)/40,
        if(TRUNCATE((batting_ratings_talent_eye+40)/40,0)>5,5, TRUNCATE((batting_ratings_talent_eye+40)/40,
                                0))) as batting_ratings_talent_eye, if(TRUNCATE((batting_ratings_talent_strikeouts+40)/40,
                                        if(TRUNCATE((batting_ratings_talent_gap+40)/40,
                                0)<1,1,
                                                        0)>5,5,TRUNCATE((batting_ratings_talent_strikeouts+40)/40,
                                                0))) as batting_ratings_talent_strikeouts FROM  players_batting) as ps RIGHT JOIN ( SELECT  first_name,
last_name,position,role,age,
0)<1,1,
0))) as running_ratings_speed, if(TRUNCATE((batting_ratings_talent_strikeouts+40
  )/40,
if(TRUNCATE((running_ratings_stealing+40)/40,0)<1,1,if(TRUNCATE((running_ratings_stealing+40)/40,
        0)>5,5,TRUNCATE((running_ratings_stealing+40)/40,0))
  ) as running_ratings_stealing,
        if(TRUNCATE((running_ratings_baserunning+40)/40,0)<1,
if(TRUNCATE((running_ratings_speed+40)/40, 1,if(TRUNCATE((running_ratings_baserunning+40)/40,
                        0
                          )>5,
                5,
                                0))
  ) as running_ratings_baserunning, pp.* FROM  players as p,
                    ( SELECT  dp2.* FROM if(TRUNCATE((running_ratings_speed+40)/40,
                0
          )>5, 5,
        SELECT  dp1.player_id,
                porder) as porder, slot FROM TRUNCATE((running_ratings_speed+40)/40,
                    ( TRUNCATE((running_ratings_baserunning+40)/40, ( if(porder IS NULL,
                                                999999, SELECT  dp.player_id,slot FROM  ootp_sql_draft_pool as dp LEFT JOIN  amateur_draft as ad  ON dp.player_id=ad.player_id
                                            WHERE  dp.league_id=101
                                              AND  ad.player_id IS NULL) as dp1
                                    LEFT JOIN  ootp_sql_draft_list as dl  ON dp1.player_id=dl.player_id
                                    WHERE  human_manager_id IN (3,-999,NULL)
                                    ORDER BY  porder) as dp2
                            GROUP BY  player_id) as pp
                    WHERE  p.player_id=pp.player_id
                      AND  position!=1) as px  ON ps.player_id=px.player_id
                          ) as x
    WHERE  pf.player_id=x.player_id
    ORDER BY  batting_ratings_talent_contact DESC,batting_ratings_talent_power DESC,
        batting_ratings_talent_eye DESC,last_name,first_name
    LIMIT  0,100

Best Answer

I would recommend adding indexes on the columns participating in the JOIN conditions and the WHERE clauses.

Indexes are designed to help the query optimizer make better choices. Unless you enforce uniqueness on a column that isn't meant to be unique, your indexes will only help.

That said, you may want to enforce uniqueness in the index for columns that should be unique, which helps the optimizer even more.

Keep in mind that indexes (and the statistics using these indexes) need regular maintenance, and they will add additional writes when the table is updated (a small overhead). This is generally far less overhead than the time taken to run queries.

Related Solutions

Mysql – How to debug a db memory-leak causing thesql to go before it’s own limits

...even surpassing it's theorically maximum possible allocation.

[OK] Maximum possible memory usage: 7.3G (46% of installed RAM)

There is not actually a way to calculate maximum possible memory usage for MySQL, because there is no cap on the memory it can request from the system.

The calculation done by mysqltuner.pl is only an estimate, based on a formula that doesn't take into account all possible variables, because if all possible variables were taken into account, the answer would always be "infinite." It's unfortunate that it's labeled this way.

Here is my theory on what's contributing to your excessive memory usage:

thread_cache_size       = 128

Given that your max_connections is set to 200, the value of 128 for thread_cache_size seems far too high. Here's what makes me think this might be contributing to your problem:

When a thread is no longer needed, the memory allocated to it is released and returned to the system unless the thread goes back into the thread cache. In that case, the memory remains allocated.

^{http://dev.mysql.com/doc/refman/5.6/en/memory-use.html}

If your workload causes even an occasional client thread to require a large amount of memory, those threads may be holding onto that memory, then going back to the pool and sitting around, continuing to hold on to memory they don't technically "need" any more, on the premise that holding on to the memory is less costly than releasing it if you're likely to need it again.

I think it's worth a try to do the following, after first making a note of how much memory MySQL is using at the moment.

Note how many threads are currently cached:

mysql> show status like 'Threads_cached';
+----------------+-------+
| Variable_name  | Value |
+----------------+-------+
| Threads_cached | 9     |
+----------------+-------+
1 row in set (0.00 sec)

Next, disable the thread cache.

mysql> SET GLOBAL thread_cache_size = 0;

This disables the thread cache, but the cached threads will stay in the pool until they're used one more time. Disconnect from the server, then reconnect and repeat.

mysql> show status like 'Threads_cached';

Continue disconnecting, reconnecting, and checking until the counter reaches 0.

Then, see how much memory MySQL is holding.

You may see a decrease, possibly significant, and then again you may not. I tested this on one of my systems, which had 9 threads in the cache. Once those threads had all been cleared out of the cache, the total memory held by MySQL did decrease... not by much, but it does illustrate that threads in the cache do release at least some memory when they are destroyed.

If you see a significant decrease, you may have found your problem. If you don't, then there's one more thing that needs to happen, and how quickly it can happen depends on your environment.

If the theory holds that the other threads -- the ones currently servicing active client connections -- have significant memory allocated to them, either because of recent work in their current client session or because of work requiring a lot of memory that was done by another connection prior to them languishing in the pool, then you won't see all of the potential reduction in memory consumption until those threads are allowed to die and be destroyed. Presumably your application doesn't hold them forever, but how long it will take to know for sure whether there's a difference will depend on whether you have the option of cycling your application (dropping and reconnecting the client threads) or if you'll have to just wait for them to be dropped and reconnected over time on their own.

But... it seems like a worthwhile test. You should not see a substantial performance penalty by setting thread_cache_size to 0. Fortunately, thread_cache_size is a dynamic variable, so you can freely change it with the server running.

Thesql works, thesqldump doesn’t

When reading the configuration file, the mysql command line client only reads configuration directives found in the [client] and [mysql] sections, while mysqldump only uses configuration directives found in the [client] and [mysqldump] sections.

If your socket directive in the config file is in the [mysql] section but not the [mysqldump] section then you should move it to the [client] section, and this should solve your problem.

So where does /var/run/mysqld/mysqld.sock come from? I remember very distinctly, once upon a time, beating my head against the wall repeatedly over that one.

It turns out, Ubuntu compiles their MySQL components from source, and when they do that, they compile them to use /var/run/mysqld/mysqld.sock unless the configuration file specifies otherwise or the socket is overridden on the command line... so if your socket directive isn't in [client] that would be one explanation of this behavior.

Best Answer

Related Solutions

Mysql – How to debug a db memory-leak causing thesql to go before it’s own limits

Thesql works, thesqldump doesn’t

Related Question