We are pleased to announce the availability of Lightbits v3.19.1 (STS). This release introduces built-in log streaming for integration with external observability and SIEM platforms, validates early interoperability with the new Windows Server NVMe-oF Initiator Preview, adds a containerized deployment option for the discovery client in OpenStack environments, and delivers a range of monitoring, resiliency, and security improvements across the stack.
Open Observability: Log Streaming to Loki, Rsyslog, and Your SIEM
Lightbits v3.19.1 introduces built-in log streaming, enabling storage cluster logs to flow automatically to external observability and security platforms, with no proprietary agents, no manual configuration files, and no service restarts required. Using Grafana Alloy as the open-source collector on each node, logs are tailed, parsed, and forwarded in real time to supported targets, including Grafana Loki, Rsyslog over mTLS, and any compatible SIEM. Configuration is fully API-driven via lbcli or REST and propagates cluster-wide within about a minute, making it straightforward to integrate Lightbits into existing observability stacks. This capability enables operations teams to correlate storage events with performance issues in real time, while satisfying audit and compliance requirements through tamper-evident, TLS-secured log routing, all built on open standards with no vendor lock-in. This capability is currently available as a Tech Preview and is planned to mature toward General Availability in the near-term release cycle.
Learn more about Open Observability for NVMe/TCP Storage, read the blog.
Containerized Discovery Client: A New Deployment Option for OpenStack Environments
As part of the v3.19.1 release cycle, a containerized deployment option for the discovery client has been added, complementing the existing .deb / .rpm package installation. Storage administrators can now choose whichever method best fits their host management model: package-based for environments that prefer native system services, or container-based for environments where Docker, for example, is the standard operational primitive. This addition is particularly valuable for OpenStack deployments managed with Kolla-Ansible, where all services are container-native and introducing a package-installed system service sits outside the standard operational model. With the containerized discovery client, every node in the OpenStack cluster that needs to talk to Lightbits storage: controller nodes, Cinder-volume nodes, and compute nodes, can be configured consistently using the same container lifecycle tooling already in use. This results in a more uniform deployment, easier rollout across nodes, and a lower operational burden for teams managing large OpenStack environments
More information is available on Discovery Client Container Deployments in Lightbits’ product documentation.
Windows Server NVMe-oF Initiator: Early Interoperability with Lightbits
Lightbits v3.19.1 has been tested against the new Windows NVMe-oF Initiator Preview, available in Windows Server Insider builds. This marks an important milestone in bringing native NVMe over TCP block storage connectivity to Windows Server, over standard Ethernet, with no specialized hardware required. In this release, storage administrators can provision a Lightbits volume and connect it to a Windows Server Insider host using nvmeofutil.exe in just a few steps. As this is a very early pre-release, we welcome feedback from customers and partners exploring this capability.
See this in action: Watch the demo video on YouTube
Learn more about the Windows NVMe-oF Initiator Preview on the Microsoft website.
Encrypted Clones from Unencrypted Base Snapshots
Lightbits v3.19.1 introduces the ability to create encrypted thin clones from unencrypted base snapshots. This capability is particularly valuable in multi-tenant environments where a provider maintains shared, unencrypted golden images, such as OS templates or base disk images, and needs to provision tenant-isolated volumes, each protected with its own unique encryption key. Tenants get strong data isolation and encryption without requiring the base image itself to be encrypted, simplifying image management while meeting per-tenant security requirements.
Richer Monitoring and Alerting
Lightbits v3.19.1 significantly expands observability coverage with new alerts and metrics across storage devices and cluster resources.
Drive and SSD Health Alerts – New Prometheus alerts have been added for disk health and disks approaching end-of-life, alongside enhanced SMART metrics with per-device dynamic temperature thresholds and richer device identification, including serial number, node UUID, and server UUID. These improvements give operations teams an earlier warning of hardware degradation before it impacts availability.
NVMe Error Log Alerts – A new collector surfaces NVMe error log data as metrics and alerts, with configurable DNR (Do Not Retry) filtering, enabling teams to detect and act on device-level errors with greater precision.
Cluster Resource Limit Alerts – New Prometheus alerts have been added for key cluster resource limits: volumes, snapshots, connected hosts, and SSDs, with a warning threshold at 70% and a critical threshold at 100%. This makes it easier to proactively manage capacity and avoid hitting hard limits in production.
Kubernetes v1.35.1 Support
The Lightbits CSI plugin has been updated to support Kubernetes v1.35.1, ensuring continued compatibility with the latest Kubernetes releases for customers running Lightbits in container-native environments.
Resiliency Hardening
This release includes a broad set of resiliency improvements that strengthen cluster behavior under adverse conditions. For example, journal device failure handling – on dual-node servers, a journaling device failure now triggers a restart of only the affected node instance rather than the entire server, reducing the blast radius of device failures and improving overall service continuity.
Continuing the Journey
Looking ahead, the Lightbits product roadmap continues to advance cluster manageability, federation, and platform breadth as customers modernize and scale their infrastructure. Every release reflects our commitment to delivering modern, efficient, and secure storage for performance-sensitive workloads across cloud, virtualization, AI infrastructure, and enterprise platforms. For more details on this release, please refer to the full release notes in the product documentation and book time to talk with an expert if you would like guidance on adopting any of the new capabilities.
Join us on the Lightbits User Community Hub to share your experiences and connect with our team.