Visit Open-E website
Results 1 to 10 of 10

Thread: Areca Error

  1. #1

    Default Areca Error

    hi, can someone explain this error?

    error Nov 29 01:56:18 dss kernel: arcmsr0: abort device command of scsi id=0 lun=1
    error Nov 29 01:56:18 dss kernel: arcmsr0: abort device command of scsi id=0 lun=0

    The system details:

    Intel S5000 chipset
    Areca 1260 with Battery.
    10x Seagate 1TB disks.
    Redundant 2+1 650 Watt Power supply.

    The error comes when we transfer a lot of data. Open-E DSS abort the iscsi connection because my data is not reliable anymore.

    Regards,
    UPservers

  2. #2
    Join Date
    May 2008
    Location
    Hamburg, Germany
    Posts
    108

    Default

    Quote Originally Posted by UPservers
    hi, can someone explain this error?

    error Nov 29 01:56:18 dss kernel: arcmsr0: abort device command of scsi id=0 lun=1
    error Nov 29 01:56:18 dss kernel: arcmsr0: abort device command of scsi id=0 lun=0

    The system details:

    Intel S5000 chipset
    Areca 1260 with Battery.
    10x Seagate 1TB disks.
    Redundant 2+1 650 Watt Power supply.

    The error comes when we transfer a lot of data. Open-E DSS abort the iscsi connection because my data is not reliable anymore.

    Regards,
    UPservers
    UPservers,
    I spat the error at Google and was provided with the following link:

    http://lkml.org/lkml/2008/2/25/507

    > > > From our experience and some customers' feedback, your issue could
    be
    > caused
    > > > by power instability or vibration to your HDs.

    There seems to be no final confirmation that the issue was related to vibration, but otoh there was no other response, either.

    Probably some other tips mentioned in that thread (increasing time-outs etc) might help with your situation, too.

    Regards,
    Jens

  3. #3

    Default

    Which Seagate drives are those you have? Many of the high end Seagte drives have firmware problems with RAID cards, if the drives are ES.2 on SN04 or SN05 firmware I would get AN05 on there pronto. There are also updates for some of the 7200.11 drives as well but I'm not sure of specifics.

  4. #4
    Join Date
    Jan 2008
    Posts
    82

    Default

    I think if the drives SN ends with 0, then they need a firmware update and if they ends in 1 then they should be good.. but not sure as this was more than 18 months a go...

  5. #5

    Default

    Quote Originally Posted by nightmare99
    Which Seagate drives are those you have? Many of the high end Seagte drives have firmware problems with RAID cards, if the drives are ES.2 on SN04 or SN05 firmware I would get AN05 on there pronto. There are also updates for some of the 7200.11 drives as well but I'm not sure of specifics.
    No we used in this old machines AS drives.

  6. #6
    Join Date
    Jan 2008
    Posts
    86

    Default

    G'day,
    Don't use the Seagate AS drives, on a RAID card, they are not supported. (I know that hurts when you have 10x).
    Apparently the firmware is "Desktop" orientated so it does not mesh well with the RAID instruction set.

    We got a lot of cmd_abort (1143) until we updated the NS drives to the AN05 firmware.

    http://kb.open-e.com/entry/90/

    Rgds Ben

  7. #7
    Join Date
    May 2008
    Location
    Hamburg, Germany
    Posts
    108

    Default

    Hello Ben,

    could you give me a pointer to information regarding the "RAID instruction set" that is to be supported by the disks? I agree that not all drives are supported 24x7 (though many newer drivers can handle that quite well), but a distinction regarding the SCSI / ATA instruction set is new to me.

    Regards
    Jens

  8. #8

    Default

    Its not an instruction set issue its a broken firmware. Even the ES.2 version of the drives had raid droppouts but these got resolved with a firmware update. As seagate can't be arsed to fix the same issue on the desktop version of the drives they just say they won't work in a raid config. Its mainly Intel IOP based controllers that have problems but other controllers can also be effected. Check out the Seagate forms for lots of unhappy campers.

  9. #9
    Join Date
    May 2008
    Location
    Hamburg, Germany
    Posts
    108

    Default

    Quote Originally Posted by nightmare99
    Its not an instruction set issue its a broken firmware. [...]
    Ah, I got it. Thanks for the clarification!

    Jens

  10. #10
    Join Date
    Jan 2008
    Posts
    86

    Default

    G'day Jens,
    Sorry I mistyped when I wrote my reply. Nightmare is 100% correct, it's the firmware that is the problem.
    According to an Adaptec tech I had a very long discussion with, there is a discrepancy in the interpretation of the SATA/SAS standards between Seagate and Adaptec(and probably others).
    So that we have one vendor, in this case Seagate, interpreting the standard in one way, and Adaptec saying "No it means this.....". As Nightmare said, it seems to be Intel IOPS based cards, which I think drive about 80% of the RAID cards out there :-(
    So the compromise is a "RAID Specific" firmware controlled by Adaptec.
    At least they now have it available as a link, and you don't have to "Apply for it".
    But of course to improve the saleability of their Seagates NS (Nearline) drives to enterprise, it would make good marketing sense not to provide it for their AS (Desktop) versions....
    I have learnt the hard way that if it's not listed on the vendors HCL, it's a gamble.

    Rgds Ben

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •