GCE实例执行google_optimize_local_ssd脚本失败求助

我通过Container Registry将应用部署到采用Container-Optimized OS的Google Compute Engine（GCE）实例中，多年来运行正常，重新部署到其他实例也很便捷。该实例由调度器管理，夜间启动、早晨停止。但从上个月开始，实例启动时出现错误：Failed to run "google_optimize_local_ssd" script: exec: "google_optimize_local_ssd":。

我原以为是存储问题，尝试重新部署并通过容器镜像创建新实例，但问题依旧，目前毫无头绪，搜索后也未找到类似问题。

实例详细信息 #

• 实例名称：instance-1
• 机器类型：E2 medium
• 区域：asia-east1
• 部署容器：无卷挂载的自定义Docker镜像
• 启动盘：Container Optimized OS 109 stable，平衡持久磁盘，100GB

日志中的错误信息 #

GCE Agent Started (version 20230726.00)
WARNING 2023-11-04T18:26:55.966772772Z [resource.labels.instanceId: instance-1] Failed to run "google_optimize_local_ssd" script: exec: "google_optimize_local_ssd": executable file not found in $PATH
WARNING 2023-11-04T18:26:55.966902882Z [resource.labels.instanceId: instance-1] Failed to run "google_set_multiqueue" script: exec: "google_set_multiqueue": executable file not found in $PATH
INFO 2023-11-04T18:26:55.988737050Z [resource.labels.instanceId: instance-1] Instance ID changed, running first-boot actions
INFO 2023-11-04T18:26:56.362940682Z [resource.labels.instanceId: instance-1] OSConfig Agent (version 20230706.02) started.
DEFAULT 2023-11-04T18:26:56.363042052Z [resource.labels.instanceId: instance-1] {"logging.googleapis.com/diagnostic":{…}}

测试用极简Dockerfile #

我还尝试使用极简Dockerfile验证问题，发现问题仍然存在：

FROM ubuntu:20.04
RUN apt update && apt upgrade

内容的提问来源于stack exchange，提问作者hiroyuki tanaka