TechSpot

RAID 1 as Backup?

By mspies
Aug 19, 2005
  1. I am building a new computer system for my business and need to develop a backup system. I am planning on installing a removable SATA drive as one drive in a RAID 1 array. I have odered a second tray and drive so I can switch one drive every week. Giving me a complete mirrored drive to store off site. If I understand Raid 1 correctly, I can cave the system rebuild the switched drive to match the stationary drive when I make the switch. If there is a drive failure I would have the system restore from the switched drive.

    This is in addition to daily incremental backups to CD.

    Will this work or is there a better way?

    Mike
     
  2. Nodsu

    Nodsu TS Rookie Posts: 5,837   +6

    No point in messing with RAID. Since the other disk will be offline then the RAID array will be in constant error state. Any disk cloning program will do the job much better and you don't have to worry about both disks having similar geometries etc..
     
  3. mspies

    mspies TS Rookie Topic Starter

    Clarification

    I am planning on having 3 HDD. One that stays in the machine and 2 that are alternated weekly. There should only be an error state when the weekly swap is made. Rebuilding the swap drive will give me a valid array.

    I think this should give me the data redundancy during the week and an off site backup at the same time.
     
  4. Nodsu

    Nodsu TS Rookie Posts: 5,837   +6

    Ah. I get it now.

    Is uptime important to you? You will have to shut down for the drive swap to have a consistent filesystem. Furthermore, depending on your RAID controller you may have to shut down the machine to do the rebuild after you swap drives (rebuild in the RAID BIOS setup thingie instead of an online rebuild) for several hours.

    IMHO this is still too elaborate. Simple things are robust and Just Work (TM). No need to make your stuff overcomplicated, especially if it is something you rely on.
     
  5. Otmakus

    Otmakus TS Rookie Posts: 51

    I too am building a new system for business and need a backup system. I am considering a RAID 1 so i don't have to worry about backup anymore. I plan to have 3 HD, in which the first one (maybe 40 gigs IDE) is where i'll install the OS and stuffs, and the last two HD (200 gigs SATA) are for data only, and i plan to RAID them. I won't have to swap the HD, since the data flow isn't that large, and it's enough to backup from it once a week or so to CDs.

    I need some advice considering on what RAID controller i should have. Some people said that it's enough to use software controller, but some others said it will greatly reduce read/write speed and stresses the CPU. As for hardware controller, some said that buying M/B with onboard SATA controller is better because it's relatively cheap and more reliable, but some others said that onboard SATA controller will again reduce performance because it will take some computing power from the CPU.

    And about SATA 1, it is said that if one HD fails, the system can still work perfectly, isn't that right? but then how can we know when one HD fails? will it give us an alert or something? if the HD isn't detected in BIOS maybe we'll know, but what if one of the HD starts to have trouble but is still detected in BIOS? And how long does it take to rebuild a 200 gigs SATA if one HD does fail? Since it's only for office data, the system only has to be up and running on work hours, i can do the maintenance after that.

    And about the OS, what OS should i use? Windows 2000, XP, or 2003 Server?

    As for the computer itself, i will only use it as a file server, so what's the rough spec i will need for it? processor and RAM? and any suggestion on what brand should i buy for the M/B?

    Any suggestions or opinions are greatly appreciated, thanks.
     
  6. i_need_answers

    i_need_answers TS Rookie Posts: 76

    Hey,
    There are External harddrives now for pretty cheap. You can't just get 1 on 2 of those?
     
  7. jobeard

    jobeard TS Ambassador Posts: 9,342   +622

    RAID is NOT a substitute for regular backups and a disaster plan that has been test!

    First recognize that the major use of a backup is to restore human and software errors:
    oops, I deleted the wrong file, we have to roll-back to the prior version.
    Raid duplication makes these errors just as redundant as the good data and leaves
    you with out a means to recover.

    Raid-1 is intended to provide uninterrupted operations due to MEDIA or
    mechanical failure. The controller will switch to the secondary and the
    SERVER will continue without going down. Notice server? These are the types
    of systems that raid was designed for.

    You might quip, 'But I intend to periodically hot-swap one drive offline as a backup'.
    Ok, how many CD-RW disks can one buy before you cover the price of all the HDs?
    In a business environment, backups typically have 'cycles' for media reuse
    and a plan for 'acceptable data loss'. This means daily, weekly and monthly
    archives are created, totalling 7 + 4 + 12 backups to span one year.
    That's a lot of media!

    'Failing to make a plan is just plain allowing the default to fail'.
     
  8. Otmakus

    Otmakus TS Rookie Posts: 51

    That's exactly what i want to do, i need SATA 1 just for avoiding loss of workhours and data due to HD problem. As for avoiding loss of data due to human errors, that's why i said i'll backup to CDs periodically.

    Just a few weeks ago, the HD of our main designer suddenly starts clicking and stops working. He's working on an important project for a few days nonstop and the deadline is near (and no, he didn't make any backups). Everyone is freaked out and i had to try many things to revive the HD, including wrapped it in a vaccuum plastic and stored it in the refrigerator for a few hours, before i managed to make it work for half an hour or so, and copied all the important data from it to another HD.

    I know for such an important projects, backups should be made at least daily, but those designers are too lazy to do it, and sometimes i just don't have the time (and maybe too lazy too) to backup all their works myself.

    After that incident, I decided that i can live more stress free having a file server as a backup to avoid the same thing happening again. It's not 100% fault proof, but at least it's better than nothing.
     
  9. jobeard

    jobeard TS Ambassador Posts: 9,342   +622

    Sounds like you've got your head on straight! good job. May I suggest that you
    automate a backup while everyone is home sleeping? By carefulling planning
    the placement of 'critical data', you can avoid taking a total HD backup
    (redundant data, extra time & media) and if it's still greater than a CD-R,
    perhaps divide and conquer by running 1/2 from one system and the other 1/2
    from another in parallel. At least it would be unattended and uninterrupted so
    the delay in completing would not matter.

    Under pressures of tight or slipping schedules, sometimes we try to cut some
    corners to save a few hours here or there -- and invariablly it comes back to
    bite us! Any other time, any other day we get away with it, but just not when
    it counts the most.

    I recall once I was tempted to not take a backup prior to a massive system
    change - - it just took too long and we needed the change asap. The change
    was such that if it failed, we had NO WAY to get back to a runable system.
    I asked myself, which is worse, lost time in a needless backup or lost time
    rebuilding the system? Murphy stepped into the room and sure enough, the
    update crashed! The two hours used for creating the backup was well worth
    the effort!
     
  10. rumpo

    rumpo TS Rookie

    Config an offsite backup server

    I'm in the process of configuring an offsite backup solution for our biz, and thought I'd pick some more learned brains before jumping in too far and fast.

    We currently use Retrospect 7, and store the data from our 2 servers (2 locations) on individual external drives.

    I am contemplating a piece of replication software called Replistor which would mirror the data through VPN for each of the servers and add another layer of redundancy.

    On top of this I am looking towards a rack mounted server with hotswap drives at an offsite colocation facility. I am being told to start with a RAID-5 for scalability, but our dataset after compression and encryption is 300Mb per night. Should I be more focussed on the processor or some other factor based on the number of drives that are being used? Does anyone have a entry level config they'd recommend for this scenario?

    Thanks
     
  11. jobeard

    jobeard TS Ambassador Posts: 9,342   +622

    Use caution on the RAID-5
    see RAID topics
    MS Planning and another set of Considerations

    excerpt from Neopapsis
    =====
    If you had four drives, I'd suggest doing a RAID 0+1 (mirror +
    stripe) logical volume, which would get you maximum speed and decent
    reliability. You want to avoid RAID 5 if at all possible, unless you
    can be absolutely certain that the resulting device is still faster
    than your machine, because RAID 5 typically sucks for writes.

    Note, if you do RAID 0+1, you want to stripe the mirrors and not
    mirror the stripes. If you mirror the stripes and you lose one disk,
    then that stripe is unavailable and that causes the mirror to be
    degraded. If you stripe the mirrors and you lose one disk, that
    mirror will be degraded but the stripe should be fine.
    =====

    Back to the backup/replication question.
    Backups (and disaster planning) is a CYA activity; You're likely to never need it,
    but if you do, boy oh boy, it sure pays for itself. If this premise is true, then
    all the effort is actually WAISTED (keep reading). To lessen the overhead losses,
    you want to move as much CYA costs off the OLAP (online application processing) systems.
    If you fully implement the co-location site (a full duplication of the OLAP,
    capable of resuming business at a moments notice), then
    REPLICATION is the fundamental tool, it's overhead is the OLAP impact,
    and you can defer the cost of BACKUP to the co-location system :)

    Replication: this needs to be frequent and only recently modified data
    (similar to an incremental backup). This will reduce the amount of data being
    replicated and thus the impact of doing so. IF the size of the replication
    becomes so large that as soon as one cycle completes another starts, you
    then change the frequency by 1/2 of whatever it currently is which will
    idle the replication system by the 1/2 not used.

    Systems day-to-day operations:
    Biggest usage of a backup image is to reclaim user errors, eg:
    a) I erased the wrong file, b) we need to rollback to last weeks A/P, ...
    This makes two points:
    1) application backup is <> (not the same as) systems backup
    2) you will need to access the CYA data more than originally considered when
    creating the disaster plan. No one would do a full systems restore for either
    of these case, would they? {hint: no system availability at all until complete}
    Instead, the file (a) would be pulled from the image and then appropriately
    relocated. The A/P (accounting en total ) system would have its own image
    and that would be used {impacting accounting but no one else}

    Systems and Techniques:
    You made no metion of the OS, and while I could deduce like candidates from
    the Retrospect and Replistor references, I'm going to keep this as general as possible.
    In the dark ages of 8086, 80030, and even mainframes, Backup was brute force
    copy record 1, track 1, cylindar 1, record 2, track 1, cylindar 1 ... ad naseum(sp?)
    Every backup of the same size HD took exactly the same time, and faithfully
    copied even EMPTY blocks to the backup media (real dump hey).
    This was a physical backup. Today, GOOD systems use a logical backup;
    copy file data, not HD blocks. An HD with only a few files will backup at light speed :)
    The side effect is also a backup that is more portable, ie can be restored to
    a different sized logical disk.

    Backups today should be focused upon Logical Volumes or
    partitions(file systems), not physical disks. This will allow resizing the data store
    to make more free space, as well as defragmenting it when restored
    (another use of backups).

    {note to moderator: hum; kind of lengthy. Free free to off load as an attachment, but if left in-line, it should be searchable }
     
  12. Otmakus

    Otmakus TS Rookie Posts: 51

    Can u explain more about that auto backup system? Do u mean to automatically set the computer to copy certain folders (which contain critical data) to another computer in the network? Do i need a third party software or is windows scheduler enough for this task? The most important thing is it have to be fully automatic, since the users can't be relied on doing anything to backup their own works.

    My office's data that need to be backed up is comprised of tens of Gigs of data files, and the data is spread among 4 workstations, used by the designers. Since the data is so big compared to the system files, i considered backing up the entire HD, and human error is minimal because those designers know what they're doing (at least they should).

    Instead of RAID backup, i consider cross backing up between two comps (comp A's files are backed up to comp B, comp B's files are backed up to comp A). This way we don't need to buy any new hardwares. What does anyone think about this set up? Can anyone suggest how to do this completely automatically?

    Thanks
     
  13. jobeard

    jobeard TS Ambassador Posts: 9,342   +622

    (1)Automatic implies scheduled task running some script, eg
    ..at 00:10 xcopy $s:\critical\*.* $t:\critical\
    ..were $s is the source and $t is the target
    see also MS Backup support

    Any OEM backup software that only supports a GUI(Graphical User Interface)
    means a live person must be present to do the work. IMO, these are toys; keep looking for software that can be SCRIPTED.

    Here we differ. Taking a backup (by whatever means) of the entire HD is wasteful;
    It can ONLY be restored directly to the system whence it came, contains tons
    of 'i really dont care data that can be obtained by other means',
    and there will be multiple copys of it stored. Big waste of time and effort.

    IMO, only the folder that contains critical data is necessary.
    Where you have a structure like
    \Projects
    \Projects\archives
    \Projects\active
    \Projects\active\AAA
    \Projects\active\BBB​
    you can pick the whole tree, eg: $s\Projects\*.* $t\Projects\
    or when projects are more loosely structured like

    \ProjectsAAA\
    \ProjectsBBB\ ​
    you can pick them individually
    eg: $s\ProjectsAAA\*.* $t\Backups\ProjectAAA\
    $s\ProjectsBBB\*.* $t\Backups\ProjectBBB\

    This is a DUPLICATION technique and it can be effective. implement as in (1) above.

    Issues: If any file is open by some user/application, it will not be copied.
    You'll need to run them sequentially rather than concurrently as they will consume
    heavy resources and one will degrade the performance of the other.
    In addition, make sure it's done off shift; less you invoke the rather of the users
    for impacting their work.

    You might start with a backup over the NETWORK. Be prepared and carefully
    consider an externally attached device to each system, eg USB or Firewire HD
    of sufficient size. Much better performance. The sole detriment would be the
    absence of your backup(s) being offsite in the case of fire/flood et al.
    Business continuation planning is much more than just backups, so discuss this
    with management.
     
  14. PanicX

    PanicX TechSpot Ambassador Posts: 669

    In addition to the previous comments, using RAID for a backup is a terrible idea in the scenario that only a couple files may need to be restored, however, you'll revert the entire system to a state weeks earlier to restore those files. Then you'll need to copy them to a seperate location so that you can reinstate the current drive. All the while your system is down, and should you have not removed your network connection during this restore, users may have been actively modifying your backup data.

    And on top of all these problems, hard drives, IMO, are not ideal canidates for removable media. Drop this drive once and you might as well not have a backup at all. Where as tape and optical storage is much more durable.

    Then there's also the issue of data security. Having your data off site in an unencrypted format may not be acceptable for sensitive data. If you just do a Google search you'll see dozens of articles of identity theft perpetraded through stolen backup media. If you're going to be taking sensitive data offsite, you need to encrypt it.
     
  15. jobeard

    jobeard TS Ambassador Posts: 9,342   +622

    good point, easy solution; on the target backup system, use an EFS filesystem,
    or encrypted folder
     
Topic Status:
Not open for further replies.

Similar Topics

Add New Comment

You need to be a member to leave a comment. Join thousands of tech enthusiasts and participate.
TechSpot Account You may also...