问题

ESXi补丁和文件

  • 2020年4月27日
  • 2回答
  • 1796的浏览量
D
  • DarrylO
  • “航行者”号
  • 2回答

vCenter / ESXi6.7的3个节点集群

运行文件。

需要打补丁的特定主机上的Files VM应该关闭或移动吗?

我知道应该关闭Controller VM。

此外，我想我已经看到提到，我可以加载ESXi补丁的某处/以某种方式到Prism(Element, Starter, 5.10.9)和Prism将管理移动和关闭?

谢谢

3 node cluster running vCenter \/ ESXi6.7<\/p>

Also running Files.<\/p>

Should the Files VM on the\u00a0particular host\u00a0to be patched be shut down or moved?<\/p>

I know the Controller VM should just be shut down.<\/p>

Also, I think I\u2019ve seen mention that I could load the ESXi patches somewhere\/somehow into Prism(Element, Starter, 5.10.9) and Prism would manage the moves and shutdown?<\/p>

Thanks<\/p>

\u00a0<\/p>","quoteUsername":"DarrylO","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">

2回答

V
徽章 ＋1

嘿,DarrylO,

据我所知，fsvm可以在集群中自由移动(这就是DRS可以在集群上做的事情)，从这个意义上讲，它们可以被视为uvm。很高兴被专家纠正(本周晚些时候将开始修补)。

有了补丁，这已经在人们的愿望清单上很多年了。问题是，虽然你确实可以上传一个二进制文件到PRISM并通过一键更新推送，但你不能同时上传多个文件。一次只做一个。因此，当您部署ESXi Update(累积)或跨主机升级ESXi时，它可能会有所帮助。但是对于普通的补丁，你可以有多个文件，一键式是没有用的。我的意思是，你可以一个一个地推送补丁，在每个补丁之后等待滚动重启。但从经验来看，一个接一个地使用VUM和给主机打补丁会更快(老实说，也更方便)，但是手动处理CVM的关闭和启动，并在每次重启后等待PRISM中的Data Resilience返回OK。仍然非常痛苦。(附带说明，根据最近6.0->6.7升级的经验，我当然也会避免滚动ESXi版本的推送——当主机启动时有太多的情况，但在转移到下一个之前，你必须先参加一些事情。)综上所述，一键式管理程序显然适用于AHV，但ESXi仍然需要以自己的方式维护，处理cvm的开销和逐个运行。

希望这个有帮助。

Hey DarrylO,<\/p>

\u00a0<\/p>

From what I know, the FSVM-s\u00a0can be freely vMotion-ed across the cluster (this is what DRS can be doing on your cluster anyway), and they can be treated as UVMs in that sense. Happy to be corrected by experts (as going to start patching later this week).<\/p>

\u00a0<\/p>

With patching, that has been in a wish-list for years. The thing is, although you indeed can upload a binary in to PRISM and push via One-Click Update, you cannot upload multiple files at once. Only one at a time. So it may help in case when you deploy an ESXi Update (cumulative), or upgrade ESXi across the hosts. But with normal patching, where you can have multiple files, One-Click is of no use. I mean, well, you can push the patches one by one, waiting for rolled reboots after each patch. But from the experience, it\u2019s quicker (and let\u2019s be honest, more convenient) to use VUM and patch hosts one by one, but handle CVM shutdowns and starts manually, and wait for Data Resilience to return to OK in PRISM after each reboot. Still very painful. (A side note,\u00a0with recent experience of 6.0->6.7 upgrade, I\u2019d certainly refrain from rolled ESXi version push as well - too many cases when the host is up, but before moving on to the next one, you have to attend a number of things first.) In summary, One-Click for hypervisor obviously works fine for AHV, but ESXi still needs to maintained in its own way with the overhead of handling CVMs and going one-by-one.<\/p>

Hope this helps.<\/p>","quoteUsername":"VictorKiwi","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">

D

感谢对esx的“一次一个文件”信息的澄清。

Thanks for the clarification on the \u201cone file at a time\u201d info for esx.<\/p>","quoteUsername":"DarrylO","translations":{"Common":{"like":"Like","unlike":"Unlike"},"Forum":{"Quote":"Quote","Share":"Share"}}}">

回复


Baidu