Here’s a new blog post based on the information provided:
VMware Update Manager: The Need for Local Repository Support
When it comes to updating virtual machines (VMs) in remote sites with low bandwidth and high latency, using VMware Update Manager (UM) can be a challenge. In my previous article, I discussed why dropping OS patching from UM was a good thing. However, there’s still one feature that I believe would greatly benefit users in ROBO (Remote Office Branch Office) environments: the ability to tell a central host to apply patches from a local repository.
Currently, UM updates VMs by downloading patches from a central repository and applying them to the hosts. This can be problematic in low-bandwidth, high-latency environments where downloading large files over the network can cause significant delays or even failures. In such cases, it’s more practical to download the patches manually to a local vMA installation and apply them locally, minimizing downtime and potential failure risks.
To address this issue, I propose that UM gain the ability to tell a remote host to apply patches from a local file repository. This could be achieved in several ways:
1. Local NAS storage: Remote sites can maintain their own local NAS storage containing the patches, and UM can be configured to retrieve the patches from there.
2. DNS-based patch repository: By using some DNS magic, remote vSphere hosts can be told to fetch their updates from a central location such as _\patchrepo\vmware_. This would still be a local repository, but with the convenience of being able to manage all patches from a single location.
3. Replication of patches: UM could handle the replication of patches from the central installation to remote sites, ensuring that all hosts have the latest updates without relying on network downloads.
The benefits of this feature are numerous:
1. Reduced network usage: By using local repositories, remote sites can reduce their reliance on network bandwidth and minimize downtime.
2. Improved failover: In case of network failures or other issues, hosts can still apply patches from the local repository, ensuring business continuity.
3. Enhanced manageability: With the ability to tell a central host to apply patches from a local repository, users can more easily manage updates across their remote sites.
4. Better performance: Local repositories can improve patching performance by reducing network usage and leveraging existing infrastructure.
I understand that adding such a feature may require significant changes to UM’s architecture and functionality. However, considering the benefits it would bring to ROBO environments, I believe it’s worth exploring. In fact, vNinja.net’s Christian Mohn and Stine Elise Larsen have already expressed their support for this idea.
In conclusion, while dropping OS patching from UM was a positive move, adding the ability to tell a central host to apply patches from a local repository would further enhance the product’s value for users in ROBO environments. By leveraging existing infrastructure and minimizing network usage, this feature would improve update management, reduce downtime, and enhance business continuity.