Today , one of the databases , is Percona Server (Percona XtraDB (http://www.percona.com) 1.1.8-20.1 )
Backgroud
Our website has a high traffic at 10:00 AM, 11:00 AM , 12:00 AM
when was in 11:00 , we found that the MySQL instance was bad in respose time. There are a lot of queries are executing .
we use the tcprstat to measure the response time of MySQL, Found that it was not very well .
After the high traffic, I test the same SQL with mysqlslap , and save the "show engine innodb status \G " information as the attament , would you help me to check it out what is going on?
Here is some key information about the innodb status;
SEMAPHORES
OS WAIT ARRAY INFO: reservation count 1539700, signal count 133057301
–Thread 140593242461952 has waited at row0row.c line 792 for 0.0000 seconds the semaphore:
X-lock (wait_ex) on RW-latch at 0x7fdebbc79ab8 ‘&block->lock’
a writer (thread id 140593242461952) has reserved it in mode wait exclusive
number of readers 1, waiters flag 0, lock_word: ffffffffffffffff
Last time read locked in file row0sel.c line 3090
Last time write locked in file /home/jiyuan/rpmbuild/BUILD/tb-mysql-5.5.18/storage/innobase /row/row0ins.c line 2
019
–Thread 140580246472448 has waited at row0sel.c line 2918 for 0.0000 seconds the semaphore:
S-lock on RW-latch at 0x7fdeb9d76e78 ‘&block->lock’
number of readers 10, waiters flag 0, lock_word: ffff6
Last time read locked in file btr0sea.c line 942
Last time write locked in file /home/jiyuan/rpmbuild/BUILD/tb-mysql-5.5.18/storage/innobase /row/row0upd.c line 2
–Thread 140593210758912 has waited at row0sel.c line 2918 for 0.0000 seconds the semaphore:
S-lock on RW-latch at 0x7fdeb9d76e78 ‘&block->lock’
number of readers 10, waiters flag 0, lock_word: ffff6
Last time read locked in file btr0sea.c line 942
Last time write locked in file /home/jiyuan/rpmbuild/BUILD/tb-mysql-5.5.18/storage/innobase /row/row0upd.c line 2