Showing entries 91 to 100 of 471
« 10 Newer Entries | 10 Older Entries »
Displaying posts with tag: Insight for DBAs (reset)
File carving methods for the MySQL DBA

This is a long overdue blog post from London’s 44con Cyber Security conference back in September. A lot of old memories were brought to the front as it were; the one I’m going to cover in this blog post is: file carving.

So what is file carving? despite the terminology it’s not going to be a full roast dinner; unless you have an appetite for data which as you’re here I’m assuming you have.

The TL;DR of “what is file carving” is taking a target blob of data (often a multi GB / TB file) and reducing it in to targeted pieces of data, this could be for instance grabbing all the jpeg images in a packet capture / mysqldump; or pulling that single table/schema out of a huge mysqldump with –all-databases (if you’re not using mydumper you really …

[Read more]
Making HAProxy 1.5 replication lag aware in MySQL

HAProxy is frequently used as a software load balancer in the MySQL world. Peter Boros, in a past post, explained how to set it up with Percona XtraDB Cluster (PXC) so that it only sends queries to available nodes. The same approach can be used in a regular master-slaves setup to spread the read load across multiple slaves. However with MySQL replication, another factor comes into play: replication lag. In this case the approach mentioned for Percona XtraDB Cluster does not work that well as the check we presented only returns ‘up’ or ‘down’. We would like to be able to tune the weight of a replica inside HAProxy depending on its replication lag. This is what we will do in this post using HAProxy 1.5.

Agent …

[Read more]
InnoDB’s multi-versioning handling can be Achilles’ heel

I believe InnoDB storage engine architecture is great for a lot of online workloads, however, there are no silver bullets in technology and all design choices have their trade offs. In this blog post I’m going to talk about one important InnoDB limitation that you should consider.

InnoDB is a multiversion concurrency control (MVCC) storage engine which means many versions of the single row can exist at the same time. In fact there can be a huge amount of such row versions. Depending on the isolation mode you have chosen, InnoDB might have to keep all row versions going back to the earliest active read view, but at the very least it will have to keep all versions going back to the start of SELECT query which is currently running.

In most cases this is not a big deal – if you have many short transactions happening you will have only a few row versions to deal with. If you just use the system for reporting queries but do not …

[Read more]
Recover MySQL root password without restarting MySQL (no downtime!)

Disclaimer: Do this at your own risk! It doesn’t apply if you’re using Pluggable authentication and certainly won’t be usable if/when MySQL system tables are stored on InnoDB

Recover your root password with care!

What is the situation?

The situation is the classic “need to recover MySQL root password” but you cannot restart MySQL (because it is the master production server, or any other reason), which makes the –skip-grant-tables solution as a no-no possibility.

 What can I do?

There is a workaround, which is the following:

  •  Launch another instance of mysqld, a small one (without innodb).
  •  Copy your user.[frm|MYD|MYI] files from the original datadir to the datadir of the new instance.
  • Modify them and then copy …
[Read more]
MySQL 5.6 Transportable Tablespaces best practices

In MySQL 5.6 Oracle introduced a Transportable Tablespace feature (copying tablespaces to another server) and Percona Server adopted it for partial backups which means you can now take individual database or table backups and your destination server can be a vanilla MySQL server. Moreover, since Percona Server 5.6, innodb_import_table_from_xtrabackup is obsolete as Percona Server also implemented Oracle MySQL’s transportable tablespaces feature which as I mentioned gives you the ability to copy tablespace (table.ibd) between servers. Let me demonstrate this through one example where I am going to take partial backup of selective tables instead of an entire MySQL server and restore it on a running MySQL …

[Read more]
What happens when your application cannot open yet another connection to MySQL

Have you ever experienced a situation where one moment you can connect to the MySQL database and the next moment  you cannot, only to be able to connect again a second later? As you may know one cannot open infinite connections with MySQL. There’s a practical limit and more often than not it is imposed by the underlying operating system. If you’re getting:

ERROR 2003 (HY000): Can't connect to MySQL server on '192.168.0.10' (99)

…there’s a good chance you’re hitting such limit. What might be misleading in the information above is whom (which side) is preventing the connection from being established.

Understanding the problem at hand

Whenever a client uses the network to connect to a service running on a given port of a server this connection is established through the creation of a socket:

[Read more]
Tips from the trenches for over-extended MySQL DBAs

This post is a follow-up to my November 19 webinar, “Tips from the Trenches: A Guide to Preventing Downtime for the Over-Extended DBA,” during which I described some of the most common reasons DBAs experience avoidable downtime. The session was aimed at the “over-stretched DBA,” identified as the MySQL DBA short of time or an engineer of another discipline without the depth of the MySQL system. The over-stretched DBA may be prone to making fundamental mistakes that cause downtime through poor response time, operations that cause blocking on important data or administrative mishaps through the lack of best practice monitoring and alerting. (You can download my slides and view the recorded webinar here.)

Monitor the things
One of the aides to keeping the system up and …

[Read more]
Faster restarts for MySQL and Percona Server 5.6.21+

By default in MySQL 5.6, each time MySQL is started (regular start or crash recovery), it iterates through all the binlog files when GTIDs are not enabled. This can take a very long time if you have a large number of binary log files. MySQL and Percona Server 5.6.21+ have a fix with the simplified-binlog-gtid-recovery option. Let’s explore this issue.

Understanding the issue

It was first reported by Yoshinori @ Facebook (bug #69097).

Let’s start by looking at a MySQL 5.6 instance where binary logging is enabled but GTIDs are disabled.

If we restart MySQL with strace, we’ll see:

# strace -e open service mysql start
[...]
open("/var/lib/mysql5621/mysql-bin.000004", O_RDONLY) = 13 …
[Read more]
Sys Schema for MySQL 5.6 and MySQL 5.7

Performance Schema (P_S) has been available since MySQL 5.5, more than 4 years ago. It is still difficult to see production servers with P_S enabled, even with MySQL 5.6 where the feature is enabled by default. There have been several complaints like the overhead, that the fix is a work in progress, and the ease of use. 52 tables to query and 31 configuration variables is enough to scare people.

There is a solution for the second problem, the usability. It’s name is “sys schema“. It is a collection of views, functions and procedures to make P_S more user friendly.

Installation

If you are a MySQL Workbench user the installation is pretty easy because sys schema is already included. You just need to install it. Click on “Performance – Performance Reports” and there you will find the “Install Helper” button that will install sys schema. …

[Read more]
How a set of queries can be killed in MySQL using Percona Toolkit’s pt-kill

You might have encountered situations where you had to kill some specific select queries that were running for long periods and choking the database. This post will go into more detail with an example of report query offloading.

Report query (select) offloading to a slave server is a common practice to reduce the workload of the master DB server. The long running selects will be executed in the slave for report generation. I have observed in many cases where the slave used to get delayed or the slave DB encounters a slowdown due to some heavy long-running orphaned selects from some wrong reports.

There are two main ways to kill queries in MySQL: 1. use custom scripts that match on a regular expression, or 2. use a tool written and supported by Percona that is designed to kill queries based on matching conditions. Below is one script that will help you to kill those queries. The script will take the process list from MySQL and …

[Read more]
Showing entries 91 to 100 of 471
« 10 Newer Entries | 10 Older Entries »