mejv

Member
  • Content Count

    259
  • Joined

  • Last visited

Everything posted by mejv

  1. If people start buying drive like they buy cars ( they buy SUVs to commute alone on dry asphalt 95% of the time, never using 4x4 but plenty of gas, just because they might need the space if they'd go once evey 5 years with the whole family in anywhere!). Why would one buy a 250 GB drive if he's going to use 20 GB to surf the net? Cleaning the garbage collected over the years can help save space, and a 40GB would be plenty for that use. If you are in need of space, newer drive can add more capacity (and speed) for less $/GB than before. So buy what you need today, not too much, that you can get more later, cheaper... and backups are more and more needed as we use more and more important data on our systems... MEJV
  2. The deal with SAS (serial Attached SCSI) is that it was too exensive to get the 16 lines to work inparallele at 480 or 640 MB/s. Because of the skew of each line, and the hardware required to compensate it and get all the bits aligned, it was easier to go serial... Fibre Channel and Ethernet proved that a serial link could be as fast/faster without the complication of the parallelism of the bits... No skew to deal with, each bit arrive one after another in the proper order, it is easy to realigned them in bytes... That is why the fastest buses go serial (PCI-E) I thought you'd like to know... MEJV
  3. SO you have an ADAPTEC SCSI RAID Ultra2 (80MB/s) with 3 Seagate SCSI drives in a RAID 0 config. One drive is flacky, the whole thing is not accessible. Have you tried hooking the drives to another power supply, just in case the voltage is marginal... If you know which drive is the faulty one (ID 1 seems to be) mirroring it to another SCSI drive (same or bigger capacity) should do the trick... dd under linux, or ghost ... should do... Having a backup of whatever is stored should be helpful... But by now standard, this SCSI RAID 0 won't beat many single modern drive... plus the 27 GB is 'small' ... The ADAPTEC Bios verify is what it is, a verify command. The verifay command instruct the drive to READ the data to check every physical sector on the drive without transfering the data to the HBA... to make sure every sector can be read with ECC on... If an READ error occured, the 131U2 bios will prompt you to re-assign the faulty sector(s)... Good luck... MEJV
  4. I would think that your bios is responsible for this... What kind of machine is this ? How old is it ? what is the bios version ? If it is the bios, check for a bios update... MEJV
  5. As said earlier here, you can replace the faulty drive by any drive of any speed of the same or greater capacity. It does not have to be faster to work, but it would affect the spped of the array since they all need to complete the data request to allow the RAID to send the data to the host... It is a matter of speed vs functionality... A faster drive would not speed up the array, but a slower drive would... it is as fast as the weakest link... Good luck. MEJV
  6. mejv

    Sata Question

    I think you are confused between the speed of the SATA link itself and the sustain rate of a data stream from a drive. The speed of the link is the rate the data flows on the link. SATA 1 is defined to be at 1.5Gbits/second or ~150MBytes/s. This does not mean the drives attached to it will be able to get the data off its platters at that speed. The drive just get the data off and to the drive as fast as it can, it just get or send the data in a buffer memory temporarly. The data from the link would come in burst of a few kByte at a time, so a fraction of a seconde... Same for sending the data, only a few kbytes at one time is sent, as many times as necessary to transfer all the data asked for. Hope this helps... MEJV
  7. Have you tried Spy Sweeper : http://www.webroot.com/wb/downloads/index.php you can download a trial version that update once its spyware/adware list... MEJV
  8. It would be very wise to keep an eye on the re-allocated sector counter, to make sure there are no rapid growth... These you see might be boderline unreported defects that could have been easily corrected by ECC ... over time, it could have degraded slidely enough to make the drive decide to be safe rather than sorry... Growing defects are 'normal' over time... but any mechanical device has a lifespan... they don't work forever! but a few years should not be a problem !!! MEJV
  9. You are correct when assuming that a drive has defects from manufacturing... It is a big step in the manufacturing of a drive to check the platters for defects. No surface is perfectly flat (very close to for the vast majority). The surface scan is done and analysed to check for scratches and defects. From there, the factory defect list is done. After that, a grown list of defect is maintained, this is what the user has access (ATA drives). For SCSI drives, both list are accessible, and it is possible to reformat (low level) the drive ignoring the grown list of defects to check if the format would discover them again in case of real defects. Hope this helps, MEJV
  10. Did you load the latest driver/firmware ? I just surfed the ADAPTEC web site, checked the 2410SA firmware/driver page and there was an update made early May 2004 for both firmware and driver with comments about the performance boost... Good luck. MEJV
  11. mejv

    Hdds With Built In Flash Ram?

    rfarris : to answer your question about the 100000 cycles... from the original text. "Concernant la durée de vie des mémoires flash, on parle de 100 000 cycles de lecture/écriture." Translation : "Concerning the lifespan of the FlashRom memories, it is said that 100 000 cycles of read/write." I hope this satisfies your curiosity... MEJV
  12. mejv

    Hdds With Built In Flash Ram?

    After reading the article from the link, I have to add the folowing : Microsoft that for an estimated $6, an 128 MB flash ROM could be added to laptop disk drive 1.8" or 2.5". With normal usage, the lifespan of the Flash ROM would be 40 years, but if left on 24/7 with access every 10 minutes, the 100000 lifetime cycle would be exausted in ~2 years. The added security would be when a laptop runs out of battery, the data not written to the disk would still be in the flashrom... MEJV
  13. mejv

    Problem Booting Up

    How about the SMART values ? Maybe the drive is getting bad and needs to retry a lot to read the data ? Cache READ/WRITE turned off on the drive? DMA turned off (PIO mode only?) In the BIOS, on the PCI latency the value should be between 32 and 64 (if I remember corectly) MEJV
  14. mejv

    Problem Booting Up

    Maybe it is time to check for worm, adware and virus on your system. Unless your system has not much memory (RAM that is)... Also, make sure every device has its proper device driver installed. You should also check for a fragmented disk and defrag it. Last check in the event viewer for error messages, that might give you a clue! That's it for now ! Good luck... MEJV
  15. mejv

    Sca: Scsi, Sata, Ide

    ... and as parallele SCSI can take a maximum of 15 devices, you have to get 2 channels to be able to accomodate 17 devices... MEJV
  16. mejv

    Fibre Channel And Scsi

    Well, The SCSI Bus was the first one to be used for the SCSI protocol. Fibre Channel came along and it uses SCSI commands slidely modified for allowing more devices and such. Logically, the operating system talks to the FC drives using SCSI commands over the FC cable. On the connection level, an FC drive cannot be directly connected to a SCSI adapter, similar to an USB enclosure cannot use an adapter to connect it to a FireWire (IEEE1394) connector or vice-versa... There are bridges for SCSI devices to connect to a FC HBA... But I don't know of any Bridge to connect a FC drive to a SCSI HBA... (Even if it did exist, I doubt it would be cheap enough to be worth it). That is why a USB enclosure connects to a USB connector/HBA, a FireWire enclosure connects to a FireWire connector/HBA... Hope this helps ! MEJV
  17. mejv

    Is Dropped Hard Disk Still Good?

    Well, I have 2 laptop drives a 4GB and a 4.6 GB... I read somewhere that the noise when shaking a drive is normal and not a sign of problem... That is for that. As for the reliability of a droped HD within a casing (your laptop in that case, I would think it should be OK, but who knows how much shorter its life will be due to that 'incident'... As said, backups are always a good thing ! MEJV
  18. Hum... You might want to try these Maxtors on a different HBA, without the cage... just to check if they are the cause... You'll need an adapter SCA to 68 pin + power... But that's the only way to check them out ! MEJV
  19. mejv

    Raid Queuing And Access Question

    Gilbo, I think you have to check your facts about command queing... you wrote : =================================================== Details on CQ, A queue (for any type of command queuing ATA, SATA or SCSI) is initiated when the drive sends a "queued" response back to the host instead of a "busy" response. A command queuing enabled host can then send more commands if it has any lying around. =================================================== I know for a fact (for SCSI at least) that the drive is not the one deciding if it will do queing or not... If the drive can do it and the host (HBA/driver/OS) supports it, then each command can have a 'tag', and the drive can have up to 256 active one at once, and sort them as it wishes... So the drive is not the one initiating command queing. This means that if the host side has more than one command to send to the same drive, it will assign a tag to each command that the drive has to reply with. As the drive of the HBA will create the tag, a memory location is the most likely 'tag' to be used : the memory location of the queue used by the driver (ever done link list ?). Like that, when the HBA/driver get a reply back, it knows where to look for storing info about this command... and it is unique. You wrote : ==================================================== As I noted in my last post the commands are queued by the drive in ATA, SATA and SCSI command queuing. This CQ is done in hardware (although in ATA and SATA the host is actually software, so the host side can be done in software). ==================================================== The CQ is done at a firmware level assisted by hardware to speed up things... I thought this was pointed out to you a while back in an other thread, but it looks like you are still confused about the drive side... MEJV (Au pays des aveugles les borgnes sont rois!)
  20. mejv

    Raid Queuing And Access Question

    Well, it is not said if this is a SCSI Raid of ATA Raid... Also, the OS is not mentioned... a Win 98/ME would not do any queing for sure... But Win NT/2k/XP could... and if the RAID interface to the computer is SCSI (or fibre channel) it would do it as well... MEJV
  21. mejv

    3ware News April Fool's?

    This is correct, it was in the press back in early March a few weeks ago... It has been finalised and is now official... no joke ! http://www.3ware.com/about/news/AMCC_3ware...s%20Release.pdf MEJV
  22. mejv

    Hard Drive Question Someone Help

    Because of the data caching at the OS level, and at the drive level as well, all the data modified might not be written onto the disk media if the OS is not told to shut down and make sure all the data that needs to be written to the disk. If powered off without proper shut down, the disk would contain a mix of new and old data. Usually, modem OSs have a way to detect that the system got shut down imroperly and would check the data on the disk for potentiel problems... If the system was inactive for a while before the power off without proper shut down, chances are the data is OK... If the system had activities going on when powered off, all bets are off ! Good luck... MEJV
  23. mejv

    22k Rpm Sound Good?

    I would not be surprised if the 22k is in fact really ~22500 rpm or exactly 50 % faster rotational speed than 15k rpm... Just a thought... MEJV
  24. mejv

    Scsi Help

    shep9999, Are you sure the converter don't have termination on them ? As for the SCSI IDs, SCSI needs to have each device on the bus with a unique ID. Since this is a narrow bus, the IDs can go from 0 to 7, 7 beeing usually the HBA. You might have to go to the website of each device manufacturor to figure out where the jumpers are for setting the ID. MEJV
  25. mejv

    What Is Sas?

    michaelsil1, 320 MB/s is the speed the interface is going... The drive is capable of sending data from its buffer/cache to the HBA at up to 320 MB/s during a short burst of data (much less than 1 second at a time... ) From the media, the max xfer rate : (http://www.seagate.com/cda/products/discsales/enterprise/tech/0,1084,552,00.html) ================================================== Transfer Rates Internal Transfer Rate (min) 632 Mbits/sec Internal Transfer Rate (max) 891 Mbits/sec Formatted Int Transfer Rate (min) 56 MBytes/sec Formatted Int Transfer Rate (max) 86 MBytes/sec External (I/O) Transfer Rate (max) 320 MBytes/sec ================================================== This mean that a max sustain transfer rate would be closer to 80 MB/s observed at the system level due to overhead... MEJV