从 AWS 迁移到 GCP 的 Velostrata 失败:[云实例启动失败]

Question

我正在使用velostrata将一个简单的WordPress网站从AWS迁移到GCP。

我已经在两个VPC之间设置了VPN，并成功配置了velostrata源云和目标云。

迁移过程似乎在一段时间内进展顺利。我可以看到在 GCP 控制台中创建的实例，但过了一会儿它就消失了。

当我查看velostrata日志时，错误是：

Finished job: RunInCloud on VM: i-0bc57fc47e893fac0, result: Failed, reason: [Cloud instance boot failed] (Instance boot was unsuccessful)
2019-06-20 15:00:34 DEBUG   VmWorkerImpl    
Exception:
com.velostrata.management.automation.exceptions.FailFlowException: [Cloud instance boot failed] (Instance boot was unsuccessful)
at com.velostrata.management.automation.flows.FlowBase.isDone(FlowBase.java:74)
at com.velostrata.management.automation.flows.RunInCloud.isDone(RunInCloud.java:113)
at com.velostrata.management.automation.components.VmWorkerImpl.monitor(VmWorkerImpl.java:126)
at com.velostrata.management.automation.components.VmWorkerImpl.runFlow(VmWorkerImpl.java:88)
at com.velostrata.management.automation.components.VmWorkerImpl.safeRunFlow(VmWorkerImpl.java:74)
at com.velostrata.management.automation.components.VmWorkerImpl.iterate(VmWorkerImpl.java:69)
at com.velostrata.management.automation.components.GroupWorkerImpl.lambda$run$3(GroupWorkerImpl.java:70)
at com.velostrata.automation.infra.utils.Wait.till(Wait.java:40)
at com.velostrata.automation.infra.utils.Wait.till(Wait.java:87)
at com.velostrata.management.automation.components.GroupWorkerImpl.run(GroupWorkerImpl.java:57)
at com.velostrata.management.automation.components.JobsRunnerImpl.runGroup(JobsRunnerImpl.java:88)
at com.velostrata.management.automation.components.JobsRunnerImpl.runPartOfJob(JobsRunnerImpl.java:80)
at com.velostrata.management.automation.components.JobsRunnerImpl.run(JobsRunnerImpl.java:72)
at com.velostrata.management.automation.components.JobsRunnerImpl.start(JobsRunnerImpl.java:48)
at com.velostrata.management.automation.components.RunsManagerImpl$RunFutureTask.lambda$new$0(RunsManagerImpl.java:246)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)

源 EC2 VM 只是一个简单的 Ubuntu 实例，在 docker 中运行 WordPress 并指向 RDS 实例。

在这一点上，我真的不知道我应该怎么做才能让它工作。 知道我应该从哪里开始寻找吗？

谢谢！

Answer 1

每当看到以下错误时

Finished job: RunInCloud on VM: whateverVM, result: Failed, reason: [Cloud instance boot failed](Instance boot was unsuccessful)

您还可以执行以下故障排除步骤

检查实例串行控制台日志中是否存在多个超时，例如：

[132.044894] dracut-initqueue[383]: calling: settle
[132.553566] dracut-initqueue[383]: Warning: dracut-initqueue timeout - starting timeout scripts

运行以下命令，它必须只返回一个内核(不是多个，也不是空行)：

sudo grubby --default-kernel

运行以下命令，必须返回所有包：

sudo rpm -q iscsi-initiator-utils-iscsiuio

iscsi-initiator-utils device-mapper-multipath-libs device-mapper-multipath velostrata-prep

运行以下命令，您希望 dracut 存在一个 velostrata、多路径、iscsi 模块，而不是文件列表：

sudo lsinitrd/boot/initramfs-uname -r.img | egrep "(velostrata|multipath|iscsi)" |头 -3

可能的解决方案：

1.卸载并重新安装准备包，提供用于调试目的的完整输出

2.如果 state 已安装，请尝试为当前内核重建 initrd

可能的根本原因：

脱机准备无法安装在云中运行所需的 MPIO 功能，从而使迁移任务失败。