How to debug Lock wait timeout exceeded on MySQL?

Question

In my production error logs I occasionally see:

SQLSTATE[HY000]: General error: 1205 Lock wait timeout exceeded; try restarting transaction

Answer 1

Here is what I ultimately had to do to figure out what "other query" caused the lock timeout problem. In the application code, we track all pending database calls on a separate thread dedicated to this task. If any DB call takes longer than N-seconds (for us it's 30 seconds) we log:

-- Pending InnoDB transactions
SELECT * FROM information_schema.innodb_trx ORDER BY trx_started; 

-- Optionally, log what transaction holds what locks
SELECT * FROM information_schema.innodb_locks;

With above, we were able to pinpoint concurrent queries that locked the rows causing the deadlock. In my case, they were statements like INSERT ... SELECT which unlike plain SELECTs lock the underlying rows. You can then reorganize the code or use a different transaction isolation like read uncommitted.

Good luck!

Answer 2

The big problem with this exception is that its usually not reproducible in a test environment and we are not around to run innodb engine status when it happens on prod. So in one of the projects I put the below code into a catch block for this exception. That helped me catch the engine status when the exception happened. That helped a lot.

Statement st = con.createStatement();
ResultSet rs =  st.executeQuery("SHOW ENGINE INNODB STATUS");
while(rs.next()){
    log.info(rs.getString(1));
    log.info(rs.getString(2));
    log.info(rs.getString(3));
}

Answer 3

You can use:

show full processlist

which will list all the connections in MySQL and the current state of connection as well as the query being executed. There's also a shorter variant show processlist; which displays the truncated query as well as the connection stats.

Answer 4

Due to the popularity of MySQL, there's no wonder Lock wait timeout exceeded; try restarting transaction exception gets so much attention on SO.

The more contention you have, the greater the chance of deadlocks, which a DB engine will resolve by time-outing one of the deadlocked transactions. Also, long-running transactions that have modified (e.g. UPDATE or DELETE) a large number of entries (which take locks to avoid dirty-write anomalies as explained in the High-Performance Java Persistence book) are more likely to generate conflicts with other transactions.

Although InnoDB MVCC, you can still request explicit locks using the FOR UPDATE clause. However, unlike other popular DBs (Oracle, MSSQL, PostgreSQL, DB2), MySQL uses REPEATABLE_READ as the default isolation level.

Now, the locks that you acquired (either by modifying rows or using explicit locking), are held for the duration of the currently running transaction. If you want a good explanation of the difference between REPEATABLE_READ and READ COMMITTED in regards to locking, please read this Percona article.

In REPEATABLE READ every lock acquired during a transaction is held for the duration of the transaction.

In READ COMMITTED the locks that did not match the scan are released after the STATEMENT completes.

...

This means that in READ COMMITTED other transactions are free to update rows that they would not have been able to update (in REPEATABLE READ) once the UPDATE statement completes.

Therefore: The more restrictive the isolation level (REPEATABLE_READ, SERIALIZABLE) the greater the chance of deadlock. This is not an issue "per se", it's a trade-off.

You can get very good results with READ_COMMITED, as you need application-level lost update prevention when using logical transactions that span over multiple HTTP requests. The optimistic locking approach targets lost updates that might happen even if you use the SERIALIZABLE isolation level while reducing the lock contention by allowing you to use READ_COMMITED.

Answer 5

Extrapolating from Rolando's answer above, it is these that are blocking your query:

---TRANSACTION 0 620783788, not started, process no 29956, OS thread id 1196472640
MySQL thread id 5341773, query id 189708353 10.64.89.143 viget

If you need to execute your query and can not wait for the others to run, kill them off using the MySQL thread id:

kill 5341773 <replace with your thread id>

^{(from within mysql, not the shell, obviously)}

You have to find the thread IDs from the:

show engine innodb status\G

command, and figure out which one is the one that is blocking the database.

Answer 6

What gives this away is the word transaction. It is evident by the statement that the query was attempting to change at least one row in one or more InnoDB tables.

Since you know the query, all the tables being accessed are candidates for being the culprit.

From there, you should be able to run SHOW ENGINE INNODB STATUS\G

You should be able to see the affected table(s)

You get all kinds of additional Locking and Mutex Information.

Here is a sample from one of my clients:

mysql> show engine innodb status\G
*************************** 1. row ***************************
  Type: InnoDB
  Name:
Status:
=====================================
110514 19:44:14 INNODB MONITOR OUTPUT
=====================================
Per second averages calculated from the last 4 seconds
----------
SEMAPHORES
----------
OS WAIT ARRAY INFO: reservation count 9014315, signal count 7805377
Mutex spin waits 0, rounds 11487096053, OS waits 7756855
RW-shared spins 722142, OS waits 211221; RW-excl spins 787046, OS waits 39353
------------------------
LATEST FOREIGN KEY ERROR
------------------------
110507 21:41:35 Transaction:
TRANSACTION 0 606162814, ACTIVE 0 sec, process no 29956, OS thread id 1223895360 updating or deleting, thread declared inside InnoDB 499
mysql tables in use 1, locked 1
14 lock struct(s), heap size 3024, 8 row lock(s), undo log entries 1
MySQL thread id 3686635, query id 124164167 10.64.89.145 viget updating
DELETE FROM file WHERE file_id in ('6dbafa39-7f00-0001-51f2-412a450be5cc' )
Foreign key constraint fails for table `backoffice`.`attachment`:
,
  CONSTRAINT `attachment_ibfk_2` FOREIGN KEY (`file_id`) REFERENCES `file` (`file_id`)
Trying to delete or update in parent table, in index `PRIMARY` tuple:
DATA TUPLE: 17 fields;
 0: len 36; hex 36646261666133392d376630302d303030312d353166322d343132613435306265356363; asc 6dbafa39-7f00-0001-51f2-412a450be5cc;; 1: len 6; hex 000024214f7e; asc   $!O~;; 2: len 7; hex 000000400217bc; asc    @   ;; 3: len 2; hex 03e9; asc   ;; 4: len 2; hex 03e8; asc   ;; 5: len 36; hex 65666635323863622d376630302d303030312d336632662d353239626433653361333032; asc eff528cb-7f00-0001-3f2f-529bd3e3a302;; 6: len 40; hex 36646234376337652d376630302d303030312d353166322d3431326132346664656366352e6d7033; asc 6db47c7e-7f00-0001-51f2-412a24fdecf5.mp3;; 7: len 21; hex 416e67656c73204e6f7720436f6e666572656e6365; asc Angels Now Conference;; 8: len 34; hex 416e67656c73204e6f7720436f6e666572656e6365204a756c7920392c2032303131; asc Angels Now Conference July 9, 2011;; 9: len 1; hex 80; asc  ;; 10: len 8; hex 8000124a5262bdf4; asc    JRb  ;; 11: len 8; hex 8000124a57669dc3; asc    JWf  ;; 12: SQL NULL; 13: len 5; hex 8000012200; asc    " ;; 14: len 1; hex 80; asc  ;; 15: len 2; hex 83e8; asc   ;; 16: len 4; hex 8000000a; asc     ;;

But in child table `backoffice`.`attachment`, in index `PRIMARY`, there is a record:
PHYSICAL RECORD: n_fields 6; compact format; info bits 0
 0: len 30; hex 36646261666133392d376630302d303030312d353166322d343132613435; asc 6dbafa39-7f00-0001-51f2-412a45;...(truncated); 1: len 30; hex 38666164663561652d376630302d303030312d326436612d636164326361; asc 8fadf5ae-7f00-0001-2d6a-cad2ca;...(truncated); 2: len 6; hex 00002297b3ff; asc   "   ;; 3: len 7; hex 80000040070110; asc    @   ;; 4: len 2; hex 0000; asc   ;; 5: len 30; hex 416e67656c73204e6f7720436f6e666572656e636520446f63756d656e74; asc Angels Now Conference Document;;

------------
TRANSACTIONS
------------
Trx id counter 0 620783814
Purge done for trx's n:o < 0 620783800 undo n:o < 0 0
History list length 35
LIST OF TRANSACTIONS FOR EACH SESSION:
---TRANSACTION 0 0, not started, process no 29956, OS thread id 1192212800
MySQL thread id 5341758, query id 189708501 127.0.0.1 lwdba
show innodb status
---TRANSACTION 0 620783788, not started, process no 29956, OS thread id 1196472640
MySQL thread id 5341773, query id 189708353 10.64.89.143 viget
---TRANSACTION 0 0, not started, process no 29956, OS thread id 1223895360
MySQL thread id 5341667, query id 189706152 10.64.89.145 viget
---TRANSACTION 0 0, not started, process no 29956, OS thread id 1227888960
MySQL thread id 5341556, query id 189699857 172.16.135.63 lwdba
---TRANSACTION 0 620781112, not started, process no 29956, OS thread id 1222297920
MySQL thread id 5341511, query id 189696265 10.64.89.143 viget
---TRANSACTION 0 620783736, not started, process no 29956, OS thread id 1229752640
MySQL thread id 5339005, query id 189707998 10.64.89.144 viget
---TRANSACTION 0 620783785, not started, process no 29956, OS thread id 1198602560
MySQL thread id 5337583, query id 189708349 10.64.89.145 viget
---TRANSACTION 0 620783469, not started, process no 29956, OS thread id 1224161600
MySQL thread id 5333500, query id 189708478 10.64.89.144 viget
---TRANSACTION 0 620781240, not started, process no 29956, OS thread id 1198336320
MySQL thread id 5324256, query id 189708493 10.64.89.145 viget
---TRANSACTION 0 617458223, not started, process no 29956, OS thread id 1195141440
MySQL thread id 736, query id 175038790 Has read all relay log; waiting for the slave I/O thread to update it
--------
FILE I/O
--------
I/O thread 0 state: waiting for i/o request (insert buffer thread)
I/O thread 1 state: waiting for i/o request (log thread)
I/O thread 2 state: waiting for i/o request (read thread)
I/O thread 3 state: waiting for i/o request (write thread)
Pending normal aio reads: 0, aio writes: 0,
 ibuf aio reads: 0, log i/o's: 0, sync i/o's: 0
Pending flushes (fsync) log: 0; buffer pool: 0
519878 OS file reads, 18962880 OS file writes, 13349046 OS fsyncs
0.00 reads/s, 0 avg bytes/read, 6.25 writes/s, 4.50 fsyncs/s
-------------------------------------
INSERT BUFFER AND ADAPTIVE HASH INDEX
-------------------------------------
Ibuf: size 1, free list len 1190, seg size 1192,
174800 inserts, 174800 merged recs, 54439 merges
Hash table size 35401603, node heap has 35160 buffer(s)
0.50 hash searches/s, 11.75 non-hash searches/s
---
LOG
---
Log sequence number 28 1235093534
Log flushed up to   28 1235093534
Last checkpoint at  28 1235091275
0 pending log writes, 0 pending chkp writes
12262564 log i/o's done, 3.25 log i/o's/second
----------------------
BUFFER POOL AND MEMORY
----------------------
Total memory allocated 18909316674; in additional pool allocated 1048576
Dictionary memory allocated 2019632
Buffer pool size   1048576
Free buffers       175763
Database pages     837653
Modified db pages  6
Pending reads 0
Pending writes: LRU 0, flush list 0, single page 0
Pages read 770138, created 108485, written 7795318
0.00 reads/s, 0.00 creates/s, 4.25 writes/s
Buffer pool hit rate 1000 / 1000
--------------
ROW OPERATIONS
--------------
0 queries inside InnoDB, 0 queries in queue
1 read views open inside InnoDB
Main thread process no. 29956, id 1185823040, state: sleeping
Number of rows inserted 6453767, updated 4602534, deleted 3638793, read 388349505551
0.25 inserts/s, 1.25 updates/s, 0.00 deletes/s, 2.75 reads/s
----------------------------
END OF INNODB MONITOR OUTPUT
============================

1 row in set, 1 warning (0.00 sec)

You should consider increasing the lock wait timeout value for InnoDB by setting the innodb_lock_wait_timeout, default is 50 sec

mysql> show variables like 'innodb_lock_wait_timeout';
+--------------------------+-------+
| Variable_name            | Value |
+--------------------------+-------+
| innodb_lock_wait_timeout | 50    |
+--------------------------+-------+
1 row in set (0.01 sec)

You can set it to higher value in /etc/my.cnf permanently with this line

[mysqld]
innodb_lock_wait_timeout=120

and restart mysql. If you cannot restart mysql at this time, run this:

SET GLOBAL innodb_lock_wait_timeout = 120; 

You could also just set it for the duration of your session

SET innodb_lock_wait_timeout = 120; 

followed by your query