DCF can be installed on the platform of supported Linux hosts. This procedure specifically uses the Linux installation procedure as an example.
Fulfill the following prerequisites before starting the installation:
- Installation server must have bash installed.
- Installation server must have zip installed.
- Installation server must have installed curl utility.
- MongoDB must be already installed along with first MongoDB user.
Refer, #GUID-17DDA769-C791-44A4-9689-EE021FE0E426 for more information.
- When you install DCF on a Linux Platform, assign executable permissions to the installation script.
- Ensure that you have a login with root privileges to run the installer.
- Download the installation file from http://support.vmware.com and place it in a temporary location on the server.
- Login server as root.
- Type ./vsa_dcf-10.1.5.0.bin –i console to run the installer in the console-based installer mode, press Enter.
The introduction command prompt screen appears.
- Press Enter to continue.
License agreement appears.
- PRESS ANY KEY TO CONTINUE TO READ LICENSE AGREEMENT OR PRESS '0' to ADVANCE TO END. Press Enter.
- Type Y and press Enter to accept the license agreement.
- Specify the directory to install DCF when installer prompts, or else press Enter to accept the default folder.
Default Install Folder: /opt/DCFNote: If the target directory already contains an existing DCF installation, then it upgrades (patch or full), else fresh installation takes place.
- Press Enter to continue when pre-install summary displays.
Installation continues.Note: Installer triggers rollback action on any kind of failure. Except DCF_Logs, all files and folders gets uninstalled. Errors can be seen inside DCF_Logs/INSTALL_ERR.log
- Press Enter to exit the installer when Install completed message displays.
What to do next
After successful DCF installation, Install VMware Smarts Notification Collector to pull all the events from Smarts Presentation SAM and publish the same to KAFKA for consumption.
Note: If you shutdown or stop IP/ESM server which has Cisco ACI / VCD / Cloudify discovered and restart the server with the same repos / topology using --ignore-restore-errors or --clean-topology then no issues. However if you shutdown ESM server and restart with --norestore option then old monitoring collectors for Cisco ACI / VCD / Cloudify will be running and those need to be stopped manually.