Backup System FAQ
We've had a lot of questions lately about our backup system, and this usually means lots of other people are wondering but not actually asking. Thus, this week's FAQ.
How often do we back up our file storage system?
Every night, the backup system searches out all files that have been changed that day and saves them to tape. Once every month, the system will back up the contents of the entire file system as extra insurance. The latest backup tapes are always kept off campus in case of a major disaster such as a fire or earthquake.
What is backed up?
All files that are stored on the network are backed up. This includes your personal network drive, your email, all departmental files, the WOU website, our database system, and so forth.
Help! I deleted a file by mistake! Who do I contact to get it restored?
Contact the UCS Help Desk at extension 88925 or ucshelpdesk@wou.edu. You must know the exact name of your file and the folder it was stored in.
Please note that it may take a day or so to get your file back! Since the most recent tapes are always stored off campus, a restore is not just a matter of typing in a few commands. We need to retrieve the proper tape, load it into the tape array, and only then can we start restoring your file. In some cases it may be faster to just re-create a document from scratch, especially if the document was not terribly important to begin with.
A restore request should be considered an emergency measure, and not a routine operation; please be careful when you work with files, and think twice before deleting or overwriting anything!
I created a file this morning and accidentally deleted it an hour later.
Why did the help desk tell me that my file couldn't be restored?
Since the backup only runs in the evening, if you create and delete a file during the day, the backup system will never see it. We only run the backup in the evening because it takes several hours and slows down the system somewhat. It is also better to run the backup outside normal work hours, so fewer changes are being made to files as the backup system looks at them.
I created a file on my local drive and then deleted it by mistake.
How do I get it back?
You don't, I'm sorry to say. Well, maybe if your department is willing to pay for Norton Utilities or something like that, and it was already installed on your machine before you lost the file. That costs a lot of money, though; it's cheaper to just store your stuff on the network. If you save a file on your local machine, you are responsible to make your own backups.
But I want to store my stuff on my local drive,
just in case the network goes down and I can't get to my files. Can I do that?
Well, you can if you really want to; however, you do it at your own risk. Remember that, except for planned maintenance, our file system has never gone down since 2002. Occasionally a network problem might make files inaccessible for a bit, but these are rare. Even if you never delete things by mistake (and come on, admit it, you're human) it is still possible that your hard drive might fail or your computer might crash and cause file corruption. It is far less safe to store files on your local drive than it is to store them on the network.
If my hard drive might fail, what about the drives on the cluster?
They see much heavier use, after all.
It's true that the drives in the cluster's storage system see almost constant use. However, those hard drives are arranged in a system called RAID (Redundant Array of Inexpensive Drives) that spreads data out across several drives in such a way that if any one drive fails, the data can be reconstructed in real time from the contents of the other drives. This means that a single drive failure will slow down the system, because of the extra work that needs to be done to reconstruct the data on that drive, but the data will still be readable.
On top of that, our RAID arrays include "hot spare" hard drives that are set up to automatically take over the function of a failed drive; a period of slowness caused by a failed drive will last only as long as it takes the storage system to reconstruct the missing data onto the hot spare. This all happens automatically, and then the system informs UCS so we can replace the failed drive; this can be done without turning off the array or interrupting service. It's considered normal for server hard drives to fail every so often, but our systems are set up so that it would take several such failures at once to damage any data, and even then we could recover everything except the current day's work from the backup system.
Is there enough room to store whatever I want on the network?
As of November 2006, there are over 500 gigabytes available on our system. (A gigabyte is about a thousand megabytes.) This may seem like a lot until you divide it by six thousand users! Still, there is enough space for normal operations. You should be able to store any reasonable (and even a somewhat unreasonable) amount of documents, email messages, and so on.
Music and video files, on the other hand, take up lots of room, so hopefully you'll only store them on our network if you really need them for academic reasons. Keep in mind that the daily backups mean there is a permanent record of any illegal files that anybody stores on the network.
We do recommend regularly emptying the trash folder in your Messenger Express email account; you don't need to delete anything if you don't want to, but when you do delete messages, they keep taking up space and slowing down the backup system until you empty that trash folder. The trash or recycle bin on your computer's desktop, on the other hand, doesn't matter since it is always stored on your local drive.