Agentless Application Migration Service Last Snapshot Job Failed

0

Hello, Setting up AMS with agentless discovery and replication onprem using vCenter 7.x works fine for the initial snapshot creation and replication but subsequently fails on any further snapshot creation/updates on the VM so the virtual machine shows as "stalled" in the AMS console. Checking all requirements, and the fact that it works initially, I am unable to decipher as to why subsequent snapshots are failing and there isn't much documentation on troubleshooting this so I was curious if someone had hit this before and might have some guidance?

Environment:

  • vCenter 7.x
  • Windows 2019 VM and Ubuntu 20 VM in testing - both show "stalled" after successful initial replication
  • MGN client plugin 1.1.6

Thanks for any help!

laughmk
已提问 2 年前701 查看次数
1 回答
1
已接受的回答

I have the same problem after the initial snapshot. I traced the error to this:

In the vcenter-client logs I see this error:

{
    "message": "VixDiskLibVim: Failed to verify SSL certificate: actual thumbprint=**VCENTER_THUMBPRINT** expected=**THUMBPRINT_FROM_CONFIG**\n",
    "log": {
        "type": "Remote",
        "name": "Agent",
        "level": "INFO",
        "level_number": 20
    },
    "source_code": {
        "function": "Log",
        "file": "com.amazonaws.cloudendure.agent.snapshotflavor.JVixLoggerImpl"
    },
    "runtime_thread": 24,
    "origin": {
        "hostname": "mymgn.local"
    },
    "@timestamp": "2022-08-31T18:39:18.128Z"
}

And then the snapshot transport fails, for what I see, in the vcenter-client.config file there is a thumbprint of the vCenter from the initial agent configuration, but when It fetches it again its different. I checked the thumbprint manually and I get the same as the agent when tries to transport the snapshot. So I don't know where it got the other one.

Getting the thumbprint with:

openssl s_client -connect vcenterhostname:port < /dev/null 2>/dev/null | openssl x509 -fingerprint -noout -in /dev/stdin

replace it in vcenter-client.config file, and running "vcenter_configuration_update" script seemed to work.

已回答 2 年前
  • Hi,

    You nailed it with this response. Thank you and great find! Ironically, I have gotten nothing from AWS support, but this seems to have fixed it for me. I still see some other errors in the logs so my hope is to use the support case to identify the answer to your question above about how that thumbprint is initially obtained. I will reply back here if/when I receive that answer. Great job and thank you for sharing!

    Thanks!

您未登录。 登录 发布回答。

一个好的回答可以清楚地解答问题和提供建设性反馈,并能促进提问者的职业发展。

回答问题的准则