Release Date | Version Number | Change Content | Limitations and Impact |
2026-09-08 | v1.1.21 | Optimized features related to memory compression in some scenarios. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2026-08-26 | v1.1.20 | umrd is upgraded to v1.8.7, with related features optimized. The offline-scheduling-agent has been upgraded to v1.8.9, with related features optimized. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2026-07-30 | v1.1.19 | The crane-agent has been upgraded to v1.7.5, with related logic optimized to reduce the load on the apiserver. The offline-scheduling-agent has been upgraded to v1.8.8, with support for configuring the pod memory compression watermark. umrd is upgraded to v1.8.6, with related features optimized. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2026-07-20 | v1.1.18 | Upgraded offline-scheduling-agent to v1.8.7 and fixed the occasional issue of dropped Node change events. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2026-06-10 | v1.1.17 | The offline-scheduling-agent has been upgraded to v1.8.6 to fix the issue of missing annotations on Nodes under extreme conditions. Memory compression for all Pods on a node can be enabled by using the gocrane.io/memory-compression-all-pods=enable Tag. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2026-05-18 | v1.1.16 | Fixed the slow memory leak issue in crane-agent. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2025-11-24 | v1.1.15 | The newly deployed qos agent component is deployed only on native nodes by default. Add metrics to help determine whether offline pods are in a suppressed state. After the gocrane.io/memory limit is configured for an offline pod, the qos agent proactively sets the cgroup. Fix the memory compression issue. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2025-06-19 | v1.1.14 | Monitoring metrics for native nodes in the offline and online co-location scenario are now productized. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2025-06-09 | v1.1.13 | Optimized the node high-load detection algorithm for qos agent. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2025-05-21 | v1.1.12 | qos agent supports node high-load checks. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2023-11-30 | v1.1.6 | Enable the qos initializer webhook. Optimize the memory consumption of the crane agent. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2023-11-23 | v1.1.5 | Memory compression is supported. Limit permissions to improve component stability. Fix the bug where water level suppression for offline pods fails. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2023-09-20 | v1.1.4 | Supports temporarily increasing the cpu limit during application startup to accelerate the startup process. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2023-08-14 | v1.1.3 | Fix the panic caused by a nil pointer in calculateCPUUsage. Upgrade the cAdvisor version to fix the high CPU load issue. Disable the bfq feature in DiskIO Manager, as it may cause kernel panic. Fix the issue where Pod probe execution may fail after DiskIO manager is opened and then closed. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2023-01-01 | v1.0.1 | Fixed issues with the default value settings of some parameters. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
2022-08-30 | v1.0.0 | Adds a chart to provide preliminary support for kernel-based QoS capabilities. | This upgrade does not affect existing services. During the upgrade process, the component may be unavailable. It is recommended to upgrade during off-peak hours. |
피드백