Today while editing I heard the noise that puts the fear of God into any editor…the click…click…click of a hard drive failing.
It’s only through experience and your first failed hard drive that you get to recognise the deathly clicking noise telling you that failure is only minutes away.
Some sage advice, take a screenshot every month, especially if you are using a few hard drives, the shot above is from my iMac which has two internal drives and one external drive so screenshots every month for this computer is a tad excessive.
My edit suite has twenty five hard drives, twenty one of them running from five Sonnet fusion towers so you can see the sence in taking a screen shot of this amount of drives.
The screenshot will allow you to determine which drive is lost as it will disappear once you locate it and remove it, more important if you have a few drives.
My important files are QT (Quick Time) files so everything that starts with QT is a complete program. All .mov or .mp4 are potentially important, the .mov files especially, clients footage must be stored onto two separate drives just in case the worst case senario happens.
We become all to set in our comfort zones and forget that one day a drive will fail, God forbid it’s a critical drive or one with a major edit on it, thats why it’s important while filming a job to keep all the rushes till the job has been Quick Timed onto two separate hard drives.
I am going to spend a day this week siphoning off all my .mov files to a self contained hard drive as a further back up measure, you can never be too careful with important QT footage and remember you can always update and edit with a QT master file, granted you don’t have the original footage but most of my clients want an update using fresh footage so it shouldn’t cause a problem.
A more simpler way to store files is via a cloud service, today I joined Bitcasa a service that is as easy as drag and drop with the advantage of 10GB of free space, mind you thats only JPEGs etc. video files are too big for cloud usage.
Welcome to the Bitcasa Infinite Drive from Bitcasa on Vimeo.
I just had an entire MacBook Pro die on me… But thanks to Time Capsule, I was back u and running with most of my software in just a few hours…
HDW : Thats fine if you have a “Time Capsule” big enough to store the data from all your drives but having 25 drives does not give me any option for a TC.
Great idea -I always take screenshots of the contents of my drives in my FCP room (about every month or so). I also backup all client media, FCP project data and other assets onto separate drives -worth the additional expense in my opinion.
You ever considered getting a NAS? Seems like something like an 8-bay unit would be what you’d need with a few 3TB drives in it running RAID-5. That should give you about 20TB of usable space, or about 18TB if you keep a hot spare in there in case of a drive failure. Seems a lot more practical than having a stack of drives all over the place. Admittedly RAID is not a replacement of backups, but the NAS could at least be the backup of your Sonnet fusion towers.
A NAS with TM support is a very simple and easy way to get a revision based backup. But be careful because large TM backups on some NAS tend to get corrupt once in a while which require you to erase and start over repeatedly.
Running OSX server on a macmini with an external raid and its built in time machine service is a far better for stability. Though it cost a bit more. Unless you have a mini and a few FW raid towers collecting dust.
The Click Click Click is often an indication of an area of bad media or bad blocks on the drive, which can’t be read by the machine, and which cause it to lock up when it tries to read them. In my experience with a number of hard drives in the past, this will likely only cause a few files on the drive to be irretrievable and the rest can be salvaged with software such as disk warrior. The software allows the machine to slowly scan the drive and get back most of the files without locking up and failing in the process.
Although obviously if you have backed up recently, then it’s nae bother and just restore the files that way instead…
We factor is the cost of archive now into every project. We use WD Green 3TB drives (run very cool) and use 2 drives, 3 on real mission critical stuff per client. We’re now setting up a LTO backup solution so when this is fully operation, we’ll have a 1 hard drive backup kept in the office and a LTO tape kept off site. Scott
Losing a critical drive generally is something you do once…stuff of nightmares!
I use a software that monitors the health of hard disks, using S.M.A.R.T.
Almost all hard disks have now S.M.A.R.T. built-in.
Before to fail, the hard disks start to show errors in S.M.A.R.T. (this is very useful to predict a failure).
Few errors may not be bad, but if the number of errors reported by S.M.A.R.T. increases … this is a sign for a imminent failure.
Acronis Drive Monitor is for Windows and it is free, and it estimates in percentages the health of the internal or the external hard disks, after it reads S.M.A.R.T.
I have 15 external hard disks, mostly WD My Book Studio Edition II, and if I choose RAID 1 the data is saved on both 2 hard disks from each WD My Book Studio Edition II.
To my surprise, Acronis Drive Monitor estimates at 95% and 92% the health of 2 hard disks from my laptop, and 90% for 1 of my external WD My Book Studio Edition II. All other external disks and the other computer have 100% health.
Now I move all my data from these 3 hard disks, then I’ll scan these for errors (in Windows: right-click on disk, Properties, Tools, Error checking) and maybe I’ll format, to see if the S.M.A.R.T. can show 100% health (if not I’ll replace these disks).
Other softwares that read S.M.A.R.T.:
http://en.wikipedia.org/wiki/Comparison_of_S.M.A.R.T._tools