Issues Fixed in Lightbits 3.0.1
| ID | Description |
|---|---|
| 36722 | Users can reference the NVMe device by its path name (e.g., /dev/nvme0n1) - as used during the initial system setup - to determine the storage SSD used by servers in the Lightbits storage cluster. However, this could lead to data loss since device names are not persistent across reboots. |
| 28728 | DCPMM devices might not be identified by profile-generator after power-up, leading to a non-optimized profile after a Lightbits upgrade of systems with DCPMM DIMMs installed. |
| 28582 | Discovery using nvme-cli (nvme discover, nvme connect-all) will not work with nvme-cli versions 2.0 and above. With nvme-cli below 2.3, this will result in nvme segfault. With nvme-cli 2.3 and above, it will fail with the following error: Failed to get discovery log: Invalid slot. To work around this issue, use Lightbits' discovery-client or downgrade your nvme-cli version. |
| 27046 | Lightbits events (lbcli list events) are capped at 64K events. For systems that have had more than 64k events, new events will be pushed to the end of the events list and may even be lost. Each event has a unique ID that is an incremental counter, This counter wraps around after 64K events. Events are sorted from the last event (largest ID) to the oldest. When there are more than 64K events, the ID wraps around and pushes the newest events to the end of the queue. The events cleaner mechanism (maximum events stored is 10k events) will at some point begin and remove events that are incorrectly assumed to be oldest. |
| 25573 | List events with the "--until" flag does not work. Specifying the until flag for list events could lead the specific API service to become unavailable for a short time after issuing the command. |
| 21202 | If volumes are created or deleted while a node in the cluster is in "Activating" state, other volumes might not complete the rebuild on the activating node and remain in "Degraded" protection state. If you encounter such a situation, contact Lightbits support. |
| 21159 | A volume update operation could fail to complete if an empty IP-ACL string is provided in the request. The volume could remain stuck in this state, blocking further API operations on this volume. Client I/O access to the volume will not be affected by this issue. If necessary, please contact Lightbits support for manual update of the volume state. |
| 20711 | An update of the default QoS policy will lead to a cluster that is unresponsive to any management operations. As such, do not invoke: api/v2/clusterConfig/DefaultQoSName or lbcli update cluster-config --parameter=DefaultQoSName The cluster can be recovered to a working state by deleting a few offending keys from etcd and restarting Lightbits services. |
| 20183 | If a project is updated to use a particular policy as its default QoS policy, subsequent update policy requests for this policy will fail. When a project is updated to use a particular policy as its default QoS policy, the project's name will be added to the policy's projects scope, even if the policy is global. Therefore, with the correct permissions, listing/getting policies will show a global policy with a non-empty projects scope. |
| 18523 | If the monitoring stack is installed on one of the Lighbits servers, the Lightbits services will not start on that server. |
| 16575 | Snapshots are deleted according to their retention time only if created by a snapshot policy. Snapshots created using the "create snapshot" API can only be deleted using the "delete snapshot" API. |