Server Failure, Upgrade & Transfer Woes

[Joe]

I've been a happy owner for several years. Recently, my 3U (14-2TB drive) server died a sudden death, so I decided to invest further in the brand by upgrading to a new 3U server with (14-4TB) drives. My dealer sent my drives, which contain roughly 1500 movies and 1500 CDs of content, to Kaleidescape to be transferred to the new server. After waiting through delays apparently due to a switch in suppliers, I was just informed one of my drives failed and another is experiencing intermittent failures. I am awaiting further details today, but it sounds like the company won't spend any additional time dealing with this issue without more money - and there still would be no guarantee my content could be recovered for transfer. Even though I have a speed reader and a vault, I still spent countless hours fixing titles, bookmarking and babysitting robotic misfeeds and don't relish the thought of repeating the process.

So, I guess I wanted to hear some experiences from folks who may have gone through this process. Has anything like this happened to any of you or any Kaleidescape owners you know / support? A company representative mentioned something about not being able to boot the OS due to the drive failure. Can these drives be cloned using something like Acronis? I would think - at least in the case of the intermittently failing drive, they could potentially do a bit-image clone to another drive and then attempt an OS boot with the RAID set. Also, I know drives fail, but one reason why I felt so comfortable making this dollar and time investment was because of the exceptional rarity of losing two drives and thus the dual-parity RAID. Have any of you lost two drives and thus all of your content on these servers? I always envisioned the scenario of losing a drive, but having time to replace it and thus preserve the RAID via the company's active monitoring of our systems. I've heard in this scenario, the company would ship out a replacement, I would pop it in, and the RAID configuration would automatically heal itself, bringing the new drive into the collective. Is that what normally would happen - or was I dreaming that? Losing two drives immediately makes me wonder about shipping and handling issues as well. Was sufficient care taken with my content? I just don't know.

I know nothing's guaranteed, especially when it comes to machines, but this strikes at the heart of the Kaleidescape value proposition for me, as my confidence in the security of my content is now shaken. In industry, there are ways to archive to tape and other media, but this thing is locked down. Are there any other options - say replicating to another server - for additional protection? Again, just looking for shared experiences / ideas at this point.

 

[cinelife]

Joe, sorry to hear about your problem. In all of my many years of ownership, dozens of drives and numerous servers, I've never experienced even a single drive failure. Among my friends/clients (hundreds), that I'm aware of, I can only recall perhaps a dozen or so single drive failures, over many, many years, and no one that has lost their content. There have been a couple of clients that had some problem 1TB drives that experienced more than one drive failure in a short period of time, but no one lost their content, and K always responded quickly and professionally to resolve the problems.

There is always a danger when moving content, especially when the drives have been shutdown for awhile and then activated, but it is still unusual to lose all content. Has your dealer spoken to Support to find out what, if anything, can be done to save the content (I ask because you mention additional cost, so wondering if that additional cost would result in saving your content). I'm not personally aware of any way to move content from a failed drive, unfortunately.

With RAID working properly, and a Hot Spare available, you should be able to lose up to two drives simultaneously without losing content, but there are other factors that may affect this process.


Jim

 

[Joe]

Jim, thanks for the reply. I'm glad to hear this is exceedingly rare - I thought it was, but needed some confirmation from the community. I should be getting a call from Kaleidescape support this afternoon to discuss what happened and mull over any options, if any.

 

[cinelife]

Okay, good luck with that, if it can be done, Support will do it.


Jim

 

[HiFiGuy1]

I will echo that Support is awesome, and will do anything they can to help. Rusty really helped me through a situation where a client lost two drives (I think) in quick succession, a few days apart, and lost their content. We were unfortunately not able to recover the original data. My understanding at the time was that since the second drive failed before we could get the first one replaced, and the eight drives in the 3U were FULL of data, with no room to reallocate from a bad disc after the spare became active, that we lost the whole file system. I think part of it was also because the second drive failure was in the middle of a rebuild.

It has been nearly a year since it happened, so I don't remember all the details, but whatever the circumstances, I do remember Rusty saying it was an exceedingly rare thing, and he stayed with me on the phone for hours, until well into the night, to try and save it.

 

[bedbuged]

I have a 1u server. I once had 2 drives failures happen simultaneously when I only had two drives in my system in total! K had me send them not only the drives, but my server as well so they could get a better understanding of what had happened. My story did have a happy ending though. K was able to retrieve all of my content except for 1 title. It did take them a while to do this, and they gave me no assurances . I think 1 weeks time if I remember correctly. In the end the official diagnosis from support was a motherboard failure on my server took the drives down and corrupted them. All the years on this forum and I never heard of this happening to anyone but myself until now.

I felt the same way as you... In my mind the system was bulletproof, but in my case it wasn't just a simple disk failure. it was the compound effects of a motherboard affecting all the drives at once, leaving no array to rebuild. To their credit though K support pulled it off, and they sent my a refurbished server to replace my defective one. My drives were under warranty at the time and they shipped replacement drives to me which arrived with my content reloaded onto them. Now that's service! I use my system so much, that one week seemed more like a month. That was around 5-6 years ago and I have gone without incident since.