BuyVM圣何塞迁移通告

其实这不算一个通告，主要是目前为止我发现使用BuyVM的VPS的朋友还是比较多的，官方发布迁移通告最早在一个月之前，目前已经确定了具体的时间了，所以好VPS在这里再发下，相当于给大家提个醒，美国当地时间1月19日晚上9点半开始，BuyVM的圣何塞将搬往拉斯维加斯了。

邮件原文：

Hello zhao!
This is just a friendly reminder about our upcoming move to the Fiberhub datacentre in Las Vegas this weekend.  To refresh:  all San José services will be powered down on 19 January at 2130h PST (9:30 PM GMT-8).  IMPORTANT:  IF YOU HAVE A KVM SERVICE, PLEASE INITIATE A MANUAL SHUTDOWN BEFORE 2100h.  WE CANNOT GUARANTEE A CLEAN SHUTDOWN FROM THE NODE. Once all of the gear is offline, we will be de-racking, loading into a cargo truck, and making the drive to Fiberhub.  Barring unforeseen traffic delays, we’re expecting to arrive in Las Vegas between 0500h-0600h Sunday morning.  By 0900h-1000h that morning, we should have over 80% of our nodes back online and functional.
There will be an additional network outage on the 20th when we cut over to our new networking setup – we’re estimating a max downtime of 10 minutes;  in reality, it should be no more than 1-2 minutes to bring the router online and get the cables swapped over.
To help clear up any potential questions, here’s a short FAQ for the affected users.  If you have a question not answered here, please feel free to open a Billing ticket, and we’ll clear up any confusion.

1. Will my IPv4 address change?
   Nope!  We own all of our currently-assigned IPv4 space;  when we bring the gear back online, your VPS will still have the same IP as it did before the move.
2. Will my IPv6 address change?
   Unfortunately, yes.  While we do have a generous amount of IPv6 space in our name, the majority of the current assignments are actually SWIP’d to us from HE.  Please understand that we will not be able to take requests for specific renumbering simply due to the sheer number of clients and assigned IPv6 addresses.
3. Is there any risk of data loss?
   Barring a catastrophic incident such as the cargo truck wrecking and catching fire (we promise that we won’t let Francisco drive), there is absolutely no risk of data loss for all non-KVM services.  There is only minimal risk of data loss for KVM services – clients with a KVM service are heavily encouraged to issue a full shutdown to their VPS no later than 9PM in order to eliminate this risk (we cannot guarantee a clean shutdown if we have to issue the halt from node-side).
4. Will I receive SLA/credit for this downtime?
   Short answer: No.  Long Answer:  This is not unexpected maintenance, routing issues, or other types of failure that are eligible for SLA.  This is a planned downtime announced well in advance (our first notice being more than a month prior to this one).
5. Will I be able to access the Billing Panel or StallionVM during the move?
   After some consideration, we have decided against setting up a temporary mirror for our billing and management panels.  The StallionVM panel would be next to useless anyways – with over 100 nodes offline, the panel would very likely lock up or cause damaging errors if clients ignoring these emails and unaware of the move repeatedly attempted to boot their VMs.  The same logic applies for our billing panel – we predict that a good number of people have chosen to remain unaware of this move, and we will already be busy enough testing and moving hardware without a flood of questions in our Support Ticket system.  Both panels will be temporarily replaced by a static page explaining the move for the duration of the downtime in order to prevent confusion.
6. With the panels down, how will we get updates during the move?
   We strongly recommend simply hanging around our IRC channel (irc.frantech.ca / #frantech), as several staff members who will not be involved in the physical move will still be around to answer questions and provide updates.  You are also welcome to follow Francisco (@FrantechCA) and myself (@Aldryic) on Twitter, as we will post updates via this medium as well.
7. Will New York services be affected?
   All NY services will remain online for the duration of the move, and will not be affected.  With this said, however, please be aware that we do not yet know if our NY deployment of mirrors.buyvm.net will be repaired before the move – as both locations are currently utilizing the San Jose mirror, package managers (Apt, Yum, Emerge, etc) on services in NY using our mirrors may not function correctly until we bring the gear back online in Las Vegas.
8. Why was the current stock removed?
   This will likely be our last mass-email before the move takes place.  I’ve pre-emptively removed all current stock to ensure that no new clients sign up after this email is sent and are unaware of the move, and we will not be reloading until 21 January at the very earliest.

Hopefully this covers all major questions that may come up.  Again, if there’s anything further we can clear up for you, please feel free to open a support ticket to our Billing department, and we’ll get any questions answered for you.  From all of us at Frantech/BuyVM, thank you for your continued support, and for being the type of amazing clients that make us truly love our jobs!

这份邮件，主要说了几个问题。
第一个是迁移的时间的问题，他表示在1月19日晚间21:30(GMT-8)开始关机，装车，拉到拉斯维加斯，这个开始的时间到北京时间好像是20号的中午左右。
他同时提醒KVM客户，提前自己关机下。车子把服务器拉到拉斯维加斯，然后上架，大约20号上午9-10点会重新联机(北京时间就是20日深夜到21日凌晨)。
通常情况下，是不会有任何的数据丢失，除非拉服务器的货车着火了，那就悲剧了。另外，所有用户使用的IP地址也不会变化，因为他自己的IP，可以广播到任何地方的，但是IPv6可能就没那么快给大家了。
纽约的服务器不会受到影响，反正大家21号早上尝试看看自己的小鸡上线了没有，原来的圣何塞用户将全部搬迁到拉斯维加斯Fiberhub机房。