reefer

Member
  • Content Count

    237
  • Joined

  • Last visited

Community Reputation

0 Neutral

About reefer

  • Rank
    Member
  1. The "Dell Diagnostic Partition" is a handy set of tools. You can still restore it but it is not necessary, since the diagnostic tools can be started from the bootable cd you shoud have received with your Dell. Let me stress the point again: if you do care for your data, you should at least have a decent backup. A tested method - as sysadmin i regularly encounter problems with unreadable tapes and backups that have not run, etc etc. As for storing data on a single disk... my motto : Imagine, like a good friend of mine had to, your harddisk dies. All data gone. My buddy was sick to his stomach, literally. I managed to recover some data... but most of HIS data was recovered from an old backup of MINE.... from a time i worked on HIS system.... Point is.... you choose how important and recoverable your data is. I even backup my buddy his data.... Maybe you have a good backup system, or choose not to have a good backup. I cannot help stressing the point - an occupational idiocy i guess :-)) Have you checked the bios ? If your new HDD is not properly supported by your bios, data corruption can happen. If your bios does not 'see' the new HDD as 320 GB you probably have a problem, even though all seems fine. .
  2. hi, just a few remarks : You will replace the 40 GB disk by a large one, and in the process breaking a ATA barrier if you go larger than 137 GB. Have you checked that your BIOS supports LBA ? If you should decide not to clone your disk but start anew, check whether the old disk still has a fairly small ( approx. 25MB ) diagnostics partition on it ; this holds very handy tools and i would recommend to clone at least that partition from the old disk to the new one. It is even possible to re-create this partition on the new disk if needed without cloning it. You say you value your data - do you have a good backup system in place ? Also, you could consider using mirroring or raid5 as disk setup, instead of trusting your data to a single disk that will invariably fail. I am in a hurry now but if you need help or clarification on any of the remarks, just let me know. Cheers,
  3. You are correct. It will improve performance somewhat. My point was, that the improvement in normal use is not much. Maybe measurable, but hardly noticable. Unless working on diskintensive things like rendering movies, that will benefit from a separate drive. And if the gain is not that big, maybe other considerations come into play. Like the number of times the actuator needs to move for all those small files and everchanging tempfiles. Those reduce the lifespan of the Seagate drive. And if there is not a good backup system in place, lifespan of the huge Seagate drive grows in importance. But sec performancewise it is the best solution, i agree.
  4. hmm... my reply was not restored after the server problems so let me try again... not long ago i tested with a fairly modern and fast drive and found that putting the swapfile on the 1st partition of it did not make the huge difference in performance it did with the older drives. Performancewise i can imagine the swapfile on the fast Raptor will give way better performance than putting it on the slower Seagate. Backup images, for me personally, do not need to be on a 'safe' part of the drive : i am someone that backups regularly. No sarcasm intended here. Please do not fall for a misleading safe feeling - just 1 hardware error on the disk.... and there is no 'safe' location on the disk, just a dead disk. My personal opinion about harddisks is simple : imho there are only 2 kinds of harddisks : those that are broken and those that will break. I would definitely not want temp files on the Seagate. That will only use space where i would want no temp files, for me the (mirrored) system disks will have to be the replacable workhorse. Why not put this stuff on the Raptor, it certainly is fast enough.. If the OS images reside on the Seagate and are backupped too (like on tape as well as on a bootable media to restore it as fast as possible after hardware failure) imho they are safe. These are my personal views - and i realize they do not completely stroke with general consensus on SR, but hey - if anyone can point out a mistake or a better way of doing things .... i will have learned something. Which is one of the reasons i visit SR.
  5. reefer

    unmountable_boot_volume bluescreen

    I would definitely do nothing with the drive that makes it work...If the drive is failing on the actuator, every turn of the arm is one too many. Have the machine turned off as long as possible and start ghosting asap after boot-up. This will turn the heat up though in the disk...so i fit an extra fan on the disk to keep it cool. I usually boot to and ghost over the network to slow the speed of data reading - ghosting over a 100 Mbit lan is slower than fitting an extra harddisk and going full speed, so the drive gets less hot. And yes, i had this happen... drive would fail ghosting to a 2nd disk, fail when ghosting over 100 Mb lan... after fitting a 10 Mb hub it ran ghost succesfully. I retried these steps since the drive was broken already, same results. Hope this helps.
  6. hi, personally i would partition the Seagate 1 large partition. NTFS is built for that, and will handle that size well. Since you are talking about above-average sized files you wish to store here ( music / video collections and image files ) i would format the drive with quite large clusters. That will give you more data per read action and faster defragmentation. It also gives some extra read speed so better performance. The storage loss due to slack will probably be near to none, but you may find it interesting to try Karen's Disk Slack Checker together with some disk speed tester like HD Tach to see for yourself.
  7. reefer

    RAID 5 or 10 with 8 drives?

    well, if you've got them anyway.... i'd use both as hot spares, personally.
  8. reefer

    RAID 5 or 10 with 8 drives?

    Microsoft has a nice article that may be of interest for you.... http://www.microsoft.com/technet/prodtechn...4.mspx?mfr=true
  9. reefer

    Rackmount Question

    Usually the mounting brackets are part of the drive rails. But driverails are brand-, even model specific.Especcially in width and with all mounting holes for the servercase in the right place. Buying generic rails will probably work but it may take some drilling and adjusting.
  10. reefer

    Rackmount Question

    i would ask calpc. You bought a standard 19" rack and the add for the servercase lists "Will Accommodate Slide Rails (Optional)" What you need is a set of sliderails for a standard 19" rack. Depending on how those hook up to your standard 19", you may need some special nuts (that click in the square holes of the 19" rack) and bolts to install those, but usually those come with the sliderails.
  11. Nice analogy sdbardwick ! Drugo, if it is speed you are after you have a load of options. You could stripe your 2 harddisks - that will speed things up. Or buy a PCI-E sata2 controller and stripe your current harddisks on the controller. Buy faster harddisks maybe... like Raptor 150's. If you use a PCI-E controller and a handful striped Raptor 150's you'd get a mean machine. Striping = more risk of data loss. So if it is a gaming machine with unimportant data that will probably not be a problem - just thought it needs to be mentioned...
  12. the drives will only deliver data as fast as the interface allows. Since your mobo supports sata150 and not sata300 - there's no point in trying to get the drives into a sata300 mode. Even though i am certain it will not damage disks or mobo, and i am pretty sure no data will be lost... why would you ? You have a working system now that will not perform better or faster after changing those options - and there always is a risk of damage or data loss when tinkering with those settings.
  13. I love that case... i found no other case that could store so many harddisks for so low a price. And a beauty to see.... it's got it all imho. Wow, you created kinda raid1... now that's data security !
  14. Hey, another Stacker fan ? I own 2 of the original STC-T01 cases, properly filled to the brim You are brave, using 8-disks JBOD arrays.
  15. That cage is not cooled ?! I would not dream of using those. Imho this is the same thing as trusting an expensive server to a cheap PSU... feed those cages 5 expensive harddisks and prepare for early deaths due to heat.... The idea is good though ! But i would use heavily cooled drive cages.