May 24, 2012

Comment: ext4 vs xfs on SSD

to assert that XFS is the way to go for MySQL, but I’m not sure how frequently they benchmark XFS vs ext4, and howtime to do another benchmark. What I found with XFS is that, to my great surprise, changing the number of allocation groups, setting

Post: Troubleshooting MySQL Memory Usage

in most cases you do not know where exactly memory is allocated. This was not much of the problem in MySQLTIME: NULL UPDATE_TIME: NULL 3 rows in set (0.00 sec) Innodb Memory Usage Finally it is often helpful to check howLock

Post: How to Monitor MySQL with Percona's Nagios Plugins

to monitor in MySQL. Those are only suggestions; any given deployment might want to monitor only some of them, as appropriate, and might need toto start, impossible to dig outlocking inhowsetting is a bit like a lack of disk space: it is possible for it to be slowly approached over time, and then to

Post: Helgrinding MySQL with InnoDB for Synchronisation Errors, Fun and Profit

to get the most out of Memcheck. Not everything is perfect though: the manual is only partially correct in how to properly build the server toin the daily work is to have the ability to run a selected MTR testcase under Helgrind. (For the time being, running the whole MySQL

Post: SHOW OPEN TABLES - what is in your table cache

how many threads are currently using this table, meaning have it locked or waiting on the table lock for this table to lock it. Name_lockedlocked (or lock wait started) – MySQL anyway initializes the timer so it would not be much overhead to store that time together in

Post: How to convert MySQL's SHOW PROFILES into a real profile

how much time MySQL spends in various phases of query execution, but it isn’…35 rows in set (0.00 sec) It’s pretty hard to figure out what consumed the most time, because it’s sorted in…closing tables | 0.000014 | 0.01 | 2 | 0.0000070000 | | System lock | 0.000010 | 0.01 | 1 | 0.0000100000 | | query …

Post: How to use tcpdump on very busy hosts

…W time to respond to the first request. Back to the question of, “what is a very busy host?” In… (around 30% when filtering only 4 out of hundreds of active clients). While …lock file if [ -e "$TMP/$LOCKFILE" ]; then echo “$self: lock file $LOCKFILE already exists, aborting” exit 1 fi # set trap to

Post: How Percona does a MySQL Performance Audit

how to read these, there’s a full explanation in our book, High Performance MySQL 2nd Edition (it’s not inout that I don’t spend a lot of time on settings and status. It is not where the greatest gains are toin Locked status, which might be an indication that it’s time to convert MyISAM tables to

Post: Identifying the load with the help of pt-query-digest and Percona Server

to enable logging atomically, not just for new connections as in MySQLLock_time: 0.000000 Rows_sent: 1 Rows_examined: 655360 use test; SET timestamp=1325145746; select count(*) from auto_inc; Compare that to

Post: The two even more fundamental performance metrics

how to maintain cheaply in the previous post. Here are some things we can do: Plot the data as time-series or intime is under a millisecond — maybe it’s 50 microseconds, maybe a bit longer. This is common in MySQLto a lock or some other queue, followed by the logjam clearing out