A principle behind the configuration representation used in the STR, is to support both manual configuration, i e editing configuration files in a text editor, and scripted configuration using the mzsh topo
command. Most changes to the files can be done in either way.
The figure below illustrates how you can edit a particular system property, either using a command or a text editor.
Updating a system property in STR
Manual Configuration
There are three methods of manually editing configurations in STR.
Method 1
- Open a configuration file under
MZ_HOME/common/config/cell/default/master
. - Edit the file and save.
Run the mzsh command
topo activate
. This is required in order the for changes to become effective.$ mzsh topo activate
Method 2
Open a configuration by running the
mzsh command
topo open. The configuration opens in vi or the editor specified by the environment variable EDITOR.Example - Opening a cell configuration
$ mzsh topo open cell:default
Example - Opening a container configuration
$ mzsh topo open main1
Example - Opening a pico configuration
$ mzsh topo open ec1
or
$ mzsh topo open container:main1/pico:ec1
If the pico name is not unique in the system, you will be prompted to specify the container.
Example - Multiple pico configuration sharing the same name
$ mzsh topo open ec2 (/home/main1/common/config/cell/default/master/containers/main1/picos/ec2.conf,ec2,topo://container:main1/pico:ec2) (/home/main1/common/config/cell/default/master/containers/exec1/picos/ec2.conf,ec2,topo://container:exec1/pico:ec2) Multiple entries, select one: (1) topo://container:main1/pico:ec2 (2) topo://container:exec1/pico:ec2 [1] :
- Edit the configuration and save. The mzsh command
topo activate
will be called with the--verbose
option and the saved changes are displayed in a scripted syntax.
Method 3
As an alternative to the command line tool, you can manage pico configurations and start/stop pico instances from the System Administration GUI. For further information, see Managing Picos in Desktop.
Activation and Validation
When you use the mzsh commands topo
set
or topo open
, changes are automatically validated before they are copied to the active registry. If the command and its arguments can be parsed but fails the validation, you can update the configuration or use a reset command to undo the changes.
You can disable the validation by using the option --no-activation
. Changes performed by the mzsh topo
will then remain in the master registry until you submit a separate activate command.
$ mzsh topo set --no-activation
The options Example - Output from activate with verbose option You can then restore the master registry with the the command Restart the picos to apply the changes Note! Changes to the STR are not applied on running pico instances or services. For instance, if you have updated the properties of the Platform and an EC, both must be restarted after activation. Example, after an mzsh topo activate of ec5, mzsh shutdown and startup needs to be done to apply the changes.--dry-run
and --verbose
are useful to learn the mzsh topo
syntax. When you have edited the configuration manually, use the following command, to view the corresponding edits in a scripted syntax:$ mzsh topo activate --dry-run --verbose
$ mzsh topo activate -v --dry-run
mzsh topo set topo://container:main1/pico:ec1/val:config.properties.ec.httpd.port 9096 # (was: 9092)
Dry-run: Validation successful
Dry-run: Stopping without performing activation
Dry-run: Active registry not changed
mzsh topo reset
.$ mzsh shutdown ec5
$ mzsh startup ec5
Note!
For further information about the various sub-commands that are available in the mzsh topo
command, see /wiki/spaces/MD82/pages/3786575 in /wiki/spaces/MD82/pages/3778935.
For further information about how to manage pico- and service configurations, see Managing Picos with Topo and Managing Service Configurations.