Migration Destroyed Our Site

[wesocp]

Hi all, here's a tale of woe I need to get off my chest as I can't speak to anyone at eUKhost.

We had a site built on DNN on a Virtuzzo W-VPS.
It fell over a few times due to the old NPP problem, but as I didn't dev the site I was reticent to have it moved to HyperV in case it broke something I couldn't fix.

Last week the site fell over 3 times in a few hours. After 4 or 5 hours of me complaining (and the snottiest "help" i've ever had the displeasure of being given - "not our problem, we built it on an unstable platform but won't support it, so I'll ignore your questions & try to wind you up even more") I was eventually offered three choices - move to HyperV now as we can't keep that node up, transfer it to a more stable node on the same platform, or keep ringing us to have the main node restarted.
Obviously I went with the 2nd option.

Got a reply to my ticket, the new server was online and working fine...
Except it wasn't.
For some reason a load of files (mainly the ASP folders) were corrupt and 0kb
A senior admin tried to reinstall ASP, which didn't work.
He tried to uninstall ASP, which wouldn't work.
He tried to delete the files manually, which wouldn't work.

Eventually I was told that the server was corrupt and that I would have to move to HyperV right now, the data and SQL db would be moved over.
This was done but still the site wouldn't come back up, which we're still working on.

Naturally I asked for the backup of the old "pre-move" site to be restored over the corrupt one.
Apparently these were deleted when it was moved to a new node. Seriously? Migrating the server to a new node, and not waiting to see if the site worked before deleting any working backups?! Okay...

So here we are. Day 6 with no website. No working backups were kept before carrying out a major change, no-one tested the migration (apparently as no-one had an Admin password, but went ahead anyway), I've no way to properly dump the SQL db out of the old server as SQL Manager relies on ASP... which is corrupt, and we're looking at days more downtime while the expensive webdevs consider their options and probably end up rebuilding the site from scratch.
My offsite FTP dumps of the SQL db & site fell over weeks ago as did the service that sent me an email if the FTP backups failed, so I don't have much here that can help.
Stupidly I relied on the site not getting corrupt and my hosts properly backing things up before making any changes to my server, and not replacing it on the weekly Saturday general backup.

What do I have to show for it? I'm halfway through a term for a server which I'm told isn't really supported, we spent thousands with a webdev creating a multilingual DNN site that has been pretty much destroyed and will have to be paid for again, I've lost 4 days work (and my weekend), and all I seem to get from the chat window and ticket system is that it's my fault for requesting to be on a node that didn't fall over every 10 minutes.

Not to say I don't appreciate the work the senior admins are doing trying to help, but something should have been in place for exactly this type of thing. You knew full well that when I renewed earlier this year that the Virtuozzo was unstable, and still happily took money for it, but what procedures were in place for the inevitable migration? Hundreds of these must have been performed, am I to believe that I'm the first to get a corrupt one so nothing was in place to allow for such a thing?

If anyone from the company (other than the poor admins trying to fix this) would like to contact me with an explanation, I'd really appreciate it.

[eUK-Ray]

Hello Wes,

I have been dealing with this ticket with you. I have apologized for the inconvenience caused to due the OS crash which is no fault of ours. As I have explained in the ticket Your OS crashed. This has nothing to do with the back end or backups. As I have explained when you requested the move to the New NODE your backup configuration would change and it would run on the new node. Once you move off a node the backups will not take any new backups of the vps on the old NODE as nothing is there on the old node to backup.

Now your backups were reconfigured to be done on the new node. Unfortunately your OS crashed and as explained in the ticket we recovered all your data including your DB . I then have updated the ticket with steps your developer should take to restore the DNN. If you would like some one else other then me to respond to the ticket I will do so as per your wish. But I have been trying to help you in anyway possible to recover your site from a crashed server.

I have manged to recover data from a crashed system for you and place it in the new server we have given to you. All your web developers need to do is follow the steps I sent you in the ticket LJB-597-40790 and update the configuration of DNN.

For your reference Terms of Service - eUKhost Ltd

11. Backups

11.1 eUKhost will use reasonable efforts to protect and backup data for clients on a regular basis, however, eUKhost does not guarantee the existence, accuracy, or regularity of its backup services and, therefore, you are solely responsible for making back-up files in connection with your use of the Services.


Thanks,
Ray

[wesocp]

Raymond, I'm not going to walk on eggshells with you, but I wasn't calling you or your work poor in that sense, I meant you were hard done by!
I'm sure you can understand my frustration at having calls and chat diverted to varying people/wherever, and not being in direct contact with the admins looking after the servers.

I have no problem with what you've tried to do to help, in fact I am grateful for the time you've put into it.
I simply posted on the forum to get a response from the Company who allows procedures such as this to take place without something to fall back to should someone mess up a migration, and to highlight to others the potential problems and maybe even get some help from someone who'd been in a similar situation.


But back to the point in hand.
You've mentioned an OS crash a few times now. The OS was fine before the failed migration. I was in the server looking at Plesk when it was taken down, I'd had people visiting the site, so unless you have a pre-migration backup to show this was the case (which we already know we don't), I think stating conjecture as fact is a little disingenuous.
Looking at the transcripts and Support Ticket from the evening, I can't see any mention of backups. Unless you were also Bruce that I was on the phone to, but even then there were no mention of backups made to me at any point. This was simply not a problem, moving from one piece of hardware to another.

From my point of view, a working server went down - a broken one came back up. With seemingly no intention of checking and no chance of restoration should anything fail.
This isn't ideal and I would warn anyone thinking of migrating to ensure you know the possible pitfalls involved.

Whilst both me and the WebDevs have reconfigured the DNN site settings several times now, we are still unable to get the site back up, and quoting Terms & Conditions doesn't help the situation.

[eUK-Ray]

Hello Wes,

As I have confirmed in the ticket. The OS crashed you logged into the server and checked the file system and saw the corruption. This was not something that we planned or wanted for you by us.

I have already restored the data from the crashed OS. Kindly let me know via the ticket if you require any further assistance.

If you wish to have a management team check into this you are most welcome to send an email to feedback @ eukhost.com. Our management team would be able to assist you further.

Thanks,
Ray