FromDual Ops Center User Guide

We follow the Debian and Ubuntu distribution maintenance policies for Long Term Support (LTS) Releases.

Thus we support the following distributions for Ops Center itself (v1.2.0 and newer):

We had some problems with Redhat based distributions and O/S interactions. If you really need these distributions, please get in contact with us!

Ops Center supports the following distributions as targets (database server):

If you prefer to use our FromDual Repository you can skip the following steps and continue installing Ops Center with our Ubuntu and Debian repository.

For your distribution continue with:

You possible have to adapt the iptables firewall rules as well.

You possible have to adapt the iptables firewall rules as well.

Due to missing demand on market we did not test yet.

Check if web server is answering (replace IP address with your Ops Center machines IP address): http://127.0.0.1

Check if PHP is working:

and then go to: http://127.0.0.1/hello.php

Web-server is answering with SSL: https://127.0.0.1/hello.php

Continue with Installation of Ops Center.

Add the FromDual APT repository as described FromDual APT Repository.

Then install Ops Center with the following command:

Continue with Securing Ops Center under Debian and Ubuntu.

To install Ops Center manually you can download it from our download site and store it under /tmp.

Continue with Securing Ops Center under Debian and Ubuntu.

To install Ops Center manually you can download it from our download site and store it under /tmp. Please check Prerequisites first before continuing!

Continue with Securing Ops Center under Debian and Ubuntu.

If you are not allowed to modify your web server configuration files please consider to use our .htaccess template (tpl/htaccess.template).

centos.png redhat.png

Add the FromDual RPM repository as described here.

Then install Ops Center with the following command:

Continue with Securing Ops Center under Redhat 7.

To install Ops Center for MariaDB and MySQL manually you can download it from our download site and store it under /tmp.

Continue with Securing Ops Center under Redhat 7.

To install Ops Center for MariaDB and MySQL manually you can download it from our download site and store it under /tmp.

Continue with Securing Ops Center under Redhat 7.

Redhat, Oracle Linux, Rocky Linux, AlmaLinux and CentOS are currently not supported any more because of problems with O/S interaction. Please let us know if you need these distributions…​

Set firewall rules and load some default SElinux policies:

Important: Please note that Ops Center does currently not work correctly with SElinux enabled. To make Ops Center working correctly set SElinux to permissive:

If you are not allowed to modify your web-server configuration files please consider to use our .htaccess template (tpl/htaccess.template).

Before you start with the upgrade you should do a backup of your focmm database schema first. A simple command like this should do the job:

Your current used version of Ops Center you can find under: Configuration → Repository → FromDual Ops Center version.

Upgrading from Ops Center v0.3 to v0.9 and newer happens automatically. Please do a backup of your Ops Center Instance before you upgrade!

Copy the .htaccess file from the old to the new version.

Upgrading from Ops Center v0.9 to v1.2.1 and newer happens automatically. Please do a backup of your Ops Center Instance before you upgrade!

Copy the .htaccess file from the old to the new version.

In the case you hit a serious bug after upgrade you have to downgrade Ops Center again. To do a downgrade is quite simple: Install your old Ops Center version and restore your backup done above:

Ops Center relies on ssh to securely access your servers. Thus it is necessary for Ops Center to have its own ssh keys to avoid password prompts:

Typically Ops Center cannot create itself the necessary files and needs your help:

Create the required files as follows and try again:

If everything was successful it should look as follows:

Ops Center is free of costs for non-commercial use. For commercial use you need a license of FromDual. To confirm that you are using Ops Center correctly you have to add the License Key.

To request a new License Key click on one of the links:

When you have added the license key you get an overview over the systems you are eligible to use with Ops Center:

Ops Center needs the systems crontab to run some periodical jobs. For this you have to add the jobs to the crontab:

If you have created the crontab it should look like this:

If you want to remove the crontab jobs again click on the Delete Crontab button.

The menu item Repository provides mainly information about Ops Center versions and where the configuration file is stored. This is primarily needed for bug reports.

If you hit a bug with FromDual tools or if you have a feature request you can add them to our Bug database.

This is also true for feature requests. You can report them to our Bug database or send us an <a href="mailto:feedback@fromdual.com?Subject=Feature request">email</a> or report the feature request here:

Currently there is only one user called admin available to operate Ops Center.

After logging in to Ops Center you should change this default password (admin) to something more secret. This admin user is a very powerful user which has full rights over all your databases. So take care of it!

On the Create Schema page you have to possiblity to create a new schema with a default Character Set:

In the Account Details you can see the Global Privileges,

the Schema Privileges and the Table and Column Privileges of an Account:

Here you can grant and revoke privileges to and from an Account. The password of an Account is not visible here because the password hashing algorithm of MariaDB/MySQL does not allow to revert the password from the hash.

An Account can be deleted in Account Details on the bottom.

Under the Menu item Instances → Configuration you can manage your actual MariaDB/MySQL configuration files for each database instance.

When you click on the menu Configuration it will take a while because Ops Center will directly fetch the recent database configuration file from the machine.

Converting a M/S Cluster into a M/M Cluster or vice versa is not allowed. Delete the Cluster from Ops Center and add a new one.

Choose a unique and descriptive Cluster Name and select the Cluster Type Master/Slave:

Clicking on the Create Cluster button creates a new Cluster inside Ops Center.

After creating the Cluster some configuration work has to be done: The server_id must be different on all nodes of a Replication Cluster. The variable log_slaves_update must be set to ON on all nodes which have the binary log enabled. And all Slave nodes must be set to read_only if Slave activity is checked. This can be achieved by changing the database configuration under Instance → Configuration.

Additionally you have to set the Instance to Read only under Instance → Settings → Edit.

In the Settings tab you get an overview over your Cluster configuration:

Here you can also Edit your Cluster or Delete it from Ops Center again. If you want to remove on single Node from the Replication Cluster click on the Delete Instance icon on the right.

In the Checks tab of the Replication Cluster you get all the important Checks of the Master and all its Slaves.

The following Checks are performed:

In the Operations tab you get a short overview over your Replication Cluster. Here you can Switchover from your current Master to the elected new Master and force a Failover. The difference between Switchover and a Failover Switchover is that the Switchover is nicely where as the Failover is with "brute force". So check the Failover flag only if the Master Machine has a sever problem.

In the Slave section run different actions with your Slave like:

Converting a Master/Slave Cluster into a Master/Master Cluster or vice versa is not allowed. Delete the Cluster from Ops Center and add a new one.

For Master/Master Replication Cluster only an active/passive set-up (one writer node) is supported.

A Master/Master Replication Cluster is only supported as a 2-node ring. 3-node ring is NOT supported by Ops Center.

Choose a unique and descriptive Cluster Name and select the Cluster Type Master/Master:

Clicking on the Create Cluster button creates a new Cluster inside Ops Center.

After creating the Cluster some configuration work has to be done: The server_id must be different on all nodes of a Replication Cluster. The variable log_slaves_update must be set to ON on all nodes which have the binary log enabled. And all Slave nodes must be set to read_only if Slave activity is checked. This can be achieved by changing the database configuration under Instance → Configuration.

Additionally you have to set the Instance to Read only under Instance → Settings → Edit.

In the Settings tab you get an overview over your Cluster configuration:

Here you can also Edit your Cluster or Delete it from Ops Center again. If you want to remove on single Node from the Replication Cluster click on the Delete Instance icon on the right.

In the Checks tab of the Replication Cluster you get all the important Checks of the Masters and all their Slaves.

The following Checks are performed:

In the Operations tab you get a short overview over your Replication Cluster. Here you can Switchover from your current Master to the elected new Master and force a Failover. The difference between Switchover and a Failover Switchover is that the Switchover is nicely where as the Failover is with "brute force". So check the Failover flag only if the Master Machine has a sever problem.

In the Slave section run different actions with your Slave like:

Choose a unique and descriptive Cluster Name, select the Cluster Type Galera Cluster and then select the nodes which should belong to this Cluster:

Then click on the Create Cluster button.

In the Settings tab you get an overview of all the nodes belonging to a Cluster, you can remove nodes from the Cluster by click on the Remove Instance from Cluster icon on the right. You can edit a Cluster with the Edit Cluster and delete the whole Cluster from the Ops Center Repository with the Delete Cluster button:

In the Checks tab you can see the actuall state of your Cluster and all the Cluster nodes:

In the Operations tab you get some actuall state information of your Cluster and all the Cluster nodes:

Further you can deploy a new Galera Cluster configuration with the Deploy Configuration button if there is no configuration yet.

If you deploy the Galera configuration via Ops Center make sure the variables wsrep_on (MariaDB Galera Cluster only) and wsrep_provider are set correctly. This is still an issue with v1.2.1 and older.

For MariaDB Galera Cluster on Debian based distributions:

For Galera Cluster for MySQL on Debian based distributions:

For MariaDB Galera Cluster on Redhat based distributions:

For Galera Cluster for MySQL on Redhat based distributions:

Before you start you should install the HAproxy Load Balancer package on the machines where the HAproxy Load Balancer should reside:

The HAproxy Load Balancer configuration file is located under /etc/haproxy/haproxy.cfg. A useful configuration file for testing purposes to start with looks as follows:

Ideally you add a Machine in Ops Center where the HAproxy Load Balancer resides. If you omit this step you should copy over at least the ssh Public Key of the Web-Server user to the Machine the HAproxy Load Balancer resides. For a Debian based distribution this is done as follows:

To add a new HAproxy Load Balancer to the Ops Center click on the Load Balancer item in the menu on the left and then click the Add Load Balancer button:

Choose a unique and descriptive Name for the Load Balancer and choose the Type HAproxy:

Then click to the Create button to create the Load Balancer in the Ops Center repository.

If you follow the Continue button you will end up in the configuration overview (Settings tab) of the Load Balancer. If you follow the link you will reach the Operations tab.

In the Settings tab you can see the actual configuration of your HAproxy Load Balancer. With the Edit Load Balancer button you can change some of those settings. With the Delete Load Balancer button you can remove the Load Balancer from Ops Center again.

Here you can change all the settings you have entered during adding your Load Balancer:

If you have changed your Load Balancer configuration do not forget to restart the Load Balancer.

In the Checks tab you can see if the HAproxy Load Balancer is up and running and get some simple monitoring statistics.

This functionality is currently not supported yet.

Before you start you should install the Galera Load Balancer package on the machines where the Galera Load Balancer should reside. If you are using the FromDual Repositories, this is quite easy:

The Galera Load Balancer configuration file is located under /etc/default/glb (on Debian based distributions) and under /etc/sysconfig/glbd (on Redhat based distributions). A useful configuration file to start with looks as follows:

Ideally you add a Machine in Ops Center where the Galera Load Balancer resides. If you omit this step you should copy over at least the ssh Public Key of the Web-Server user to the Machine the Galera Load Balancer resides. For a Debian based distribution this is done as follows:

To add a new Load Balancer to the Ops Center click on the Load Balancer item in the menu on the left and then click the Add Load Balancer button:

Choose a unique and descriptive Name for the Load Balancer and choose the Type Galera Load Balancer:

Then click to the Create button to create the Load Balancer in the Ops Center repository.

If you follow the Continue button you will end up in the configuration overview (Settings tab) of the Load Balancer. If you follow the link you will reach the Operations tab.

Next you should start your Galera Load Balancer if it is not started yet.

Here you can stop (Stop Load Balancer), start (Start Load Balancer) and restart (Restart Load Balancer) the Galera Load Balancer and Persist configuration of your Galera Load Balancer you have changed in the Ops Center.

When you persist the configuration the old version is backed-up in the same directory.

By clicking on the + or - icon you can increase and decrease the weight of a back end node. By clicking on the Drain node or Undrain node icon on the very right you can drain or undrain a backend node. A weight of 0.000 has the same meaning an effect as draining a node. Undraining a node leads to a node weight of 1.000. So the old value is not remembered at the moment.

In the checks tab you can see if the Galera Load Balancer is up and running and get some simple monitoring statistics.

In the Settings tab you can see the actual configuration of your Galera Load Balancer. With the Edit Load Balancer button you can change some of those settings. With the Delete Load Balancer button you can remove the Load Balancer from Ops Center again.

Here you can change all the settings you have entered during adding your Load Balancer. There is one additional option: Balancing Policy.

If you have changed your Load Balancer configuration do not forget to persist the configuration and Restart the Load Balancer in the Operations tab…​

FromDual Ops Center only supports MariaDB MaxScale version 6 (v2.6) and newer.

Before you start you should install the MariaDB MaxScale Load Balancer package on the machines where the MaxScale Load Balancer should reside.

MariaDB MaxScale can be downloaded from here: Download MariaDB.

When you have downloaded the package you can install it as follows:

Ideally you add a add a Machine in Ops Center where the MaxScale Load Balancer resides. If you omit this step you should copy over at least the ssh Public Key of the Web-Server user to the Machine the MariaDB MaxScale Load Balancer resides. For a Debian based distribution this is done as follows:

To add a new Load Balancer to the Ops Center click on the Load Balancer item in the menu on the left and then click the Add Load Balancer button:

Choose a unique and descriptive Name for the Load Balancer and choose the Type [.buttonstyle]MaxScale#:

If MaxScale is added successfully to the Ops Center repository start the Load Balancer under the Operations tab…​

In the Settings tab you can see the actual configuration of your MariaDB MaxScale Load Balancer. With the Edit Load Balancer button you can change some of those settings. With the Delete Load Balancer button you can remove the Load Balancer from Ops Center again.

In the Checks tab you can see if the MariaDB MaxScale Load Balancer is up and running and get some simple monitoring statistics.

In the Operations tab you can do the following actions: Stop Load Balancer, Start Load Balancer, Restart Load Balancer and Drain node or Undrain node on the icons to the right.

At the bottom of the page you can see the last 20 lines of the MaxScale error log (from journalctl)

Ideally you add a add a Machine in Ops Center where the ProxySQL Load Balancer resides. If you omit this step you should copy over at least the ssh Public Key of the Web-Server user to the Machine the ProxySQL Load Balancer resides. For a Debian based distribution this is done as follows:

To add a new Load Balancer to the Ops Center click on the Load Balancer item in the menu on the left and then click the Add Load Balancer button:

Choose a unique and descriptive Name for the Load Balancer and choose the Type ProxySQL:

If ProxySQL is added successfully to the Ops Center repository start the Load Balancer under the Operations tab…​

In the Settings tab you can see the actual configuration of your ProxySQL Load Balancer. With the Edit Load Balancer button you can change some of those settings. With the Delete Load Balancer button you can remove the Load Balancer from Ops Center again.

In the Checks tab you can see if the ProxySQL Load Balancer is up and running and get some simple monitoring statistics.

Operations functionality is not supported yet.

In the structure compare you can see the tables which are missing and the differences of the tables:

In the content compare you can see if the tables have different or if they have equal content:

IMPORTANT: If you have traffic on those tables it might be that the content is reported differently even it is equal. For performance reasons big tables are not compared at the moment yet.

The following Ops Center job status do exist:

Every Ops Center job creates some files on the server the job is executed in the directory specified in Configuration.inc ($gJobTmpDir). The file consists of a name and a job number:

For uploading a file just click to the Browse…​ button, select a file from your local file system and then click to Upload File.

After a while, when the upload is completed, the file will appear in the Download section of your focmm machine.

Alternatively you can also upload files from your personal computer to the focmm machine via your favourite file transfer tools like scp or sftp.

If you want to download a file from your focmm machine to your personal computer just click on one of the files shown in the download section. A window will pop-up and let you choose the location to store the file to.

Alternatively you can also download files from your focmm machine to your personal computer via your favourite file transfer tools like scp or sftp.

To transfer files between your focmm machine and your database machines or between your different database machines you have to choose the machines first, then navigate through your file system tree on each server and then press the transfer link (arrow left to right or arrow right to left) to transfer the file.