Get greater infrastructure visibility with our new post-install monitor tool

5 min read

By: Adam Shaw

Get greater infrastructure visibility with our new post-install monitor tool

To maximize your user experience, we’ve upgraded our post-installation script system. This updated system changes how the post-installation scripts are executed on your devices during provisioning.

The new system is more robust and improves the reliability of post-provision executions. Our legacy system logged into the device during the final stages of provisioning (via SSH) and downloaded and executed any customer-supplied scripts. While this generally worked (with the exception of custom internal network configs that blocked remote connections through firewalls or Vyattas), we noticed an infrastructure issue that caused the SSH connection to terminate during the final stages. We worked on a new system that downloaded and executed the post-provision scripts locally on your devices (instead of logging in externally) and released this new system to all VSIs.

Starting in February, we’re switching bare metal post-provisions to the new system—with a migration path that includes testing current post-provision scripts using VSIs. By provisioning a single VSI and verifying that your script is still executing properly, you can ensure that your scripts will work when we switch over to the new system.

While most customers were not affected by this change, some did encounter issues—for instance, in some cases they relied on the availability of specific Linux subservices upon execution. Since the new system executes earlier in the OS boot process, those services were not yet running.

We discovered that an internal tool built by our IBM Cloud development team can work for our customers, too. We’re releasing our post-install monitor tool publicly, since it gives you better visibility into your provisions and reloads. While this tool doesn’t solve all infrastructure visibility needs, it is written to be extensible.

The tool runs on an on-premises system or on an internal-only VSI or bare metal server. The monitoring tool continually watches the client deployment and notices any changes in the VSIs and hardware deployed on the account. Once a device completes deploying or reloading, it will SSH into the device, download a provided script, and execute that script remotely (identical to the legacy execution system).

A configuration file is either supplied on startup or the tool will look for a “monitor.local.cfg” in the execution directory. (Have a look at a sample configuration file.)

Users specify when using the private network (best for monitoring VSIs running on our infrastructure). If this is not set to true, the public API endpoints are used and connect to the public NIC/VIP via SSH. This is especially important when monitoring internal-only device deployment.

This tool supports monitoring of hardware and VSIs distinctly and permits the specification of various SSH settings. It was written to help customers and internal groups with post-provision script issues and provides a system which monitors all active provisions/reloads.

Greater visibility of your infrastructure allows you to programmatically react to your environment. Start using the post-install monitor tool today.

Be the first to hear about news, product updates, and innovation from IBM Cloud