I'm trying to setup a mysql cluster for a web application to avoid having a single point of failure. The Mysql documentation says:
MySQL Cluster does not provide any sort of automatic failover between
SQL nodes. Your application must be prepared to handle the loss of SQL
nodes and to fail over between them.
The docs say the data is still accessible via the NDB API, but if my app is configured to point to a mysql server, how can i setup some type of loadbalancing/failover for multiple SQL nodes.
The failover functionality is provided by client connector libraries. For example, if you're using Java, your JDBC connection string can provide a list of MySQL host names. Here's documentation. http://dev.mysql.com/doc/connector-j/5.1/en/connector-j-config-failover.html
With ado.net and python, you'll need a setup called MySQL Fabric. This explains the connection string monkey business you'll need. http://dev.mysql.com/doc/connector-net/en/connector-net-programming-fabric.html
Or, you can write specific application connection code that detects connection timeouts, and then tries another MySQL server.
Related
sorry if title is not so clear, probably I am not finding what I need due I do not know how to search
I have few MySQL servers is separated online servers (from different wordpress) and I want to load some of the data on those databases/tablets into a SQL database located on Azure.
inside Azure portal itself I do not see where to establish external connections, neither at server level, neither at database level
I download and install Microsoft SQL Server Management Studio, connect to the server, I can see my databse and the master one, Security with logins, and Integration Service Catalog, nothing else.
I was looking for something like:
https://www.jetbrains.com/help/go/db-tutorial-connecting-to-ms-sql-server.html#step-3-connect-to-microsoft-sql-server-with-datagrip
but nowhere ...
maybe something like this:
https://www.devart.com/odbc/mysql/docs/microsoft_sql_server_manager_s.htm
but no Servers objects option available on my SSMS
Can be this done?
Note: Azure database is a basic wfor now, if that is a limitation
Some choices.
In your SQL Server Management Studio create a linked server pointing to each MySQL instance. You found the instructions for that. https://www.devart.com/odbc/mysql/docs/microsoft_sql_server_manager_s.htm But it probably will not work in Azure SQL Server; you don't have access to the underlying Windows OS to install stuff like MySQL ODBC drivers, which you need. (You could ask Azure techsupport if they can help.)
In each MySQL instance, try creating a federated table connection to appropriate table in SQL Server. That cross-vendor federation stuff only works in MariaDB, however; MySQL's federation only goes MySQL <--> MySQL.
Write yourself a purpose-built extract / transform / load (ETL) program, and arrange to run it every so often. Program it to connect to all the servers involved, retrieve the data needing to be transferred from your MySQL servers, and update / insert that data on the SQL server.
(edit) You may be able to use command-line SQL client programs. mysqldump, with its --compatible option, may generate usable INSERT statements in a file. You then may be able to use sqlcmd to run those INSERTs on your Azure server. It's going to take some hacking, and may take using sed(1) or awk(1) to make the MySQL output compatible with SQL Server.
I believe the third option is the most robust one for production use.
I have installed ejabberd on an AWS EC2 instance and am using the smack library to connect to it with my android app. At the moment the ejabberd server is using mnesia as the database, however I want to perform some complex queries on some of the data (mainly the MUC room names), as SQL will predominantly the best solution for this, I was wondering if it was possible to replicate the required data to an external MySQL database that I could then query.
Is this possible or am i better looking at a different approach to this problem?
There is no module built into ejabberd to replicate data in Mnesia to MySQL. However, the usual approach is to use the backend you need for each feature. If you want mod_muc to store data into MySQL instead of Mnesia, you can just change the backend to odbc (which means it will store data for that module in a relational database).
You can refer to ejabberd documentation for MUC module: http://docs.ejabberd.im/admin/guide/configuration/#modmuc
Once your MySQL is configured and schema is loaded, you can set db_type to odbc on a case by case basis to choose MySQL for that module.
I've been wondering what are the benefits of using an ODBC driver to interface with a MySQL server? What advantage does that have over directly connecting to the server via TCP and firing off your SQL commands directly?
I'm working with a code base written in labVIEW that references a UDL file which references a data source and its associated ODBC driver to handle the connection to the server. I'm not yet convinced its necessary. I can see the benefit of using a driver if you are connecting to Microsoft Access or excel but if your recipient is an SQL server of some type why would you need a middle man to handle your SQL commands?
I have used the ODBC driver 3.51 for connecting to MySQL Community Server 5.1 for some time now. The ODBC driver allows you (on a Windows computer) to add your MySQL connection to the ODBC data sources list. Now you can reference this data source in many compatible applications. I have been able to use this ODBC connection in VB.Net applications as well as proprietary applications for printing and other functions with ODBC functionality.
I would say that when coding web or python applications i always use the built in MySQL packages, but if you need to access your data through VB.Net or another proprietary application (like a stand-alone windows app) you probably need to set up a system ODBC data source to access your database. To me the ODBC lets Windows access MySQL easily. Without Windows i don't think you need it.
Why ODBC and not JDBC? Some reasons come to mind in order of helpfulness when using JDBC:
Standard api to access the results from a query. To iterate over the results and to get the actual values with the appropriate type.
No need to know how the protocol works to connect to the server. Just use the host, user, password, send the SQL and iterate over the result.
Abstract out the connection to the database. Changing the database should be simpler.
Standard way to commit and rollback for transaction based changes.
In my experience, JDBC is very verbose. You need to write many lines to accomplish simple things. I've found the Spring JDBCTemplate wrapper to be an extremely easy to use alternative for simple stuff.
I have an existing web application using Castle Activerecord to interact with a single MySQL database. The Database has recently been reconfigured to replicate to a number of Slaves.
How do you configure Castle Activerecord to direct writes to the MySQL Master and reads to the MySQL Slaves or are there other ways of achieving Master/Slave operation transparently in this setup?
I have never done this, but I think this kind of things are better handled at the DB-driver level. Indeed, Connector/J has a specific ReplicationDriver for this purpose. The Connector/NET documentation has a rather vague reference to replication:
Host, Server, Data Source, DataSource, Address, Addr, Network Address:
The name or network address of the
instance of MySQL to which to connect.
Multiple hosts can be specified
separated by &. This can be useful
where multiple MySQL servers are
configured for replication and you are
not concerned about the precise server
you are connecting to. No attempt is
made by the provider to synchronize
writes to the database so care should
be taken when using this option. In
Unix environment with Mono, this can
be a fully qualified path to MySQL
socket file name. With this
configuration, the Unix socket will be
used instead of TCP/IP socket.
Currently only a single socket name
can be given so accessing MySQL in a
replicated environment using Unix
sockets is not currently supported.
Amazon officially states: "Amazon RDS gives you access to the full capabilities of a familiar MySQL database. This means the code, applications, and tools you already use today with your existing MySQL databases work seamlessly with Amazon RDS."
I don't get this. Amazon RDS is accessible via web services and there a client libraries (like the one for .Net).
So if I have an existing .Net application that uses a DAL which in turn queries MySQL, how can I make the same DAL talk to the Amazon RDS (via the web services). Or am I missing something here?
Amazon RDS is pure MySQL, accessible by your app the same way as any other MySQL database; the web services interface to RDS is purely for creation, deletion, and modification of the DB instances, not the DB data. From their FAQ:
Q: How do I access my running DB
Instance?
Once your DB Instance is available,
you can retrieve its
endpoint via the DescribeDBInstance
API. Using this endpoint you can
construct the connection string
required to connect directly with your
DB Instance using your favorite
database tool or programming language.
In order to allow network requests to
your running DB Instance, you will
need to authorize access. For a
detailed explanation of how to
construct your connection string and
get started, please refer to our
Getting Started Guide.
This is the part of the Getting Started Guide you need -- it explains how to get the hostname of your new instance so you can connect to it, authorize the instance for access from the client, and then connect using the MySQL command-line client (as an example):
$ rds-describe-db-instances --headers
$ rds-authorize-db-security-group-ingress default --cidr-ip 192.0.2.0/30 --headers
$ mysql -h myinstance.crwjauxgijdf.us-east-1.rds.amazonaws.com -P 3306 -u mymasteruser -p
Amazon RDS is just a normal server with normal MySQL access. There's only the webservice that handles instance creation etc., but everything mysql related is still the same.