September 24, 2013 By Severalnines

For those of you who know Severalnines and maybe use some of our tools & products, you’ll know that we provide our users with a monthly summary of all the resources & tools that we’re publishing. Since this is publicly available material, we thought it’d be useful also for the broader open source database community.

In the past month, we’ve made the following resources & tools available: 

• Troubleshooting MySQL Cluster (free online MySQL Cluster Training)
• Zero Downtime Data Center Migration with Galera
• ClusterControl 1.2.3 Released
• High Availability OpenStack - Clustering the Database Backend
• NoSQL Battle of the East Coast - Benchmarking MongoDB vs TokuMX Cluster
• Full restore of a Galera Cluster from Backup

With …

For MySQL 5.1, 5.5 and 5.6 in the same repository, after repacking:

bzr: 269MB (217MB pack, 52MB indicies)

git: 177MB repo (152MB pack)

One thing I’ll say is that BZR is always more chatty over the network and is substantially slower than GIT in pulling a fresh copy.

Ok, so after all the things that have been announced @MySQLConnect, I’ve got to play around with them. First stop: space (no.. not ‘the final frontier’).

I need more space on my f/s to get installing. I was a bit of a scrooge when I created my Oracle Linux virtual machine, so now I’m paying the price.

Filesystem            Size  Used Avail Use% Mounted on
/dev/mapper/vg_ol63uek01-LogVol01
                      7.1G  5.7G  1.1G  85% /

As I’m using Virtual Box, I’ve added a new SATA Controller vmdk of 10G, SATA Port 1 and then start it up.

fdisk -l

Will be able to identify the new & unused partition:

[root@ol63uek01 ~]# fdisk -l
Disk /dev/sda: 8589 MB, 8589934592 bytes
 255 heads, 63 sectors/track, 1044 cylinders
 Units = cylinders of 16065 * 512 = 8225280 bytes
 Sector size (logical/physical): 512 bytes / 512 bytes
 I/O size (minimum/optimal): 512 bytes / 512 bytes
 Disk identifier: 0x00091104 …

Ok, so after all the things that have been announced @MySQLConnect, I’ve got to play around with them. First stop: space (no.. not ‘the final frontier’).

I need more space on my f/s to get installing. I was a bit of a scrooge when I created my Oracle Linux virtual machine, so now I’m paying the price.

Filesystem            Size  Used Avail Use% Mounted on
/dev/mapper/vg_ol63uek01-LogVol01
                      7.1G  5.7G  1.1G  85% /

As I’m using Virtual Box, I’ve added a new SATA Controller vmdk of 10G, SATA Port 1 and then start it up.

fdisk -l

Will be able to identify the new & unused partition:

[root@ol63uek01 ~]# fdisk -l
Disk /dev/sda: 8589 MB, 8589934592 bytes
 255 heads, 63 sectors/track, 1044 cylinders
 Units = cylinders of 16065 * 512 = 8225280 bytes
 Sector size (logical/physical): 512 bytes / 512 bytes
 I/O size (minimum/optimal): 512 bytes / 512 bytes
 Disk identifier: 0x00091104 …

In this post, I present a few examples on “how to” for multisource replication using  the MySQL Labs Linux packages for masters and the slave. We also use replication performance schema tables to find out status of replication per channel. Multisource replication with GTIDs

1. Let us set up two masters and a slave.  The following are the important options to be set in the respective cnf files.       Masters:
     gtid-mode=on      
     enforce-gtid-consistency 
     log-slave-updates

            Slave:
            master_info_repository=TABLE            relay_log_info_repository=TABLE
            gtid-mode=on
    …

For me, a Code of Conduct is not actually guidelines for how to act. For me, a Code of Conduct is what to do if there is a problem with someones conduct. When I get on a plane, I am told what to do in case of emergency here are the exits, heres how to use your life jacket and oxygen mask. I am not told every little thing that could be an emergency I believe that would be a waste of time and invariably something would be left out.

Similarly, for conferences, listing out all the behaviors that might be problematic is a waste of time, and invariably, behaviors are left out. In my opinion, that is a waste of time. What is NOT a waste of time is giving out the information of what to do in case of emergency.

At the MySQL Connect website, under the Plan tab, is a link to the Oracle …

I distinctly remember creating and committing this code... and it has largely not changed. Yet, it seems that my name was scrubbed from its history.

$ bzr blame -q include/mysql/plugin_audit.h | grep antony | wc -l

       0

Sad-face.

=-(

From time to time we have to work with live environments and production databases. For some of us this is day-to-day job. And most of the time cost of a mistake is way higher than expected improvement especially on the databases. Because issue on the database side will affect everything else.

I heard enough war stories about ruined productions and can imagine well enough speed of DROP DATABASE command replicating across the cluster. So I’m scared to make changes in production. The more loss expected if things go wrong the more I’m going to be scared planning every change. But I still love to make improvements so the only question is how to make them safer.

This post is not intended to be a guide or best practices on how to avoid issues at all, it’s more invitation to discussion that started between me and @randomsurfer in twitter on how to avoid production failures. …

Brian Aker gives the “zinger” lightning talk about the newly announced “Drizzle”. This short (under 8 minutes) video captures Aker’s highlights of why he started the Drizzle project and how Drizzle is different from MySQL — both in what has been removed from MySQL and what features Drizzle can accomodate.

Play the video directly in your browser at http://technocation.org/node/576/play or download the 116 Mb file at http://technocation.org/node/576/download.

Brian Aker gives the “zinger” lightning talk about the newly announced “Drizzle”. This short (under 8 minutes) video captures Aker’s highlights of why he started the Drizzle project and how Drizzle is different from MySQL — both in what has been removed from MySQL and what features Drizzle can accomodate.

…

Join 11,000 others and follow Sean Hull on twitter @hullsean.

Bring me clients, please… pretty please!

When you first start out as a freelancer, your network is small. Without a steady stream of projects, the tendency is to reach for whatever you can find. Head shops & agencies build a brand, and ongoing relationships with firms. However a project with a middleman is a relationship with him or her, not the client directly. You lose control of a few very key things, such as fees, testimonials and payment terms.

Read: NYC technology startups are hiring

It’s all about the relationship, Luke

With independent consulting, …