Announcement

Collapse
No announcement yet.

Corrupted parity in the Raid header?

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Corrupted parity in the Raid header?

    Has anyone here heard of having 'corrupted parity in the Raid header'?

    We regularly find that a Consistency Check on our Raid 1+0 array fails with inconsistency found.

    It has been suggested by Dell that this is caused by corrupted parity in the Raid header and the solution it to completely wipe the array and recreate it. As this array has the Windows Server system on it is going to involve backing-up, reinstalling Windows and restore from backup.

    We have never heard of this 'corrupted parity in the Raid header' error before and I just wondered if anyone else out there has ever encountered this and can tell us whether trashing and recreating the whole array is definitely going to fix the issue.

    Is this really the case that the whole thing has to be completely wiped and rebuilt or is there some other way to recreate this Raid parity header? If this is a regular problem I am surprised that LSI do not have a utility to recreate the Raid headers.

    Server is a Dell T-420 with a Perc H310 which is based on an LSI 2008, with 4-disks in Raid 1+0 .

  • #2
    Re: Corrupted parity in the Raid header?

    It sounds (IMHO) like a piece of sales Bulls**t
    Only the RAID 5/6 variants have "parity" - Raid 1+0 is a mirror of stripes (or is it a stripe of mirrors) so parity (check blocks) should not come into it.

    However there is clearly some problem!

    Advice is generally good though - any doubts about drive reliability normally means rebuild as soon as you can.
    Tom Jones
    MCT, MCSE (2000:Security & 2003), MCSA:Security & Messaging, MCDBA, MCDST, MCITP(EA, EMA, SA, EDA, ES, CS), MCTS, MCP, Sec+
    PhD, MSc, FIAP, MIITT
    IT Trainer / Consultant
    Ossian Ltd
    Scotland

    ** Remember to give credit where credit is due and leave reputation points where appropriate **

    Comment


    • #3
      Re: Corrupted parity in the Raid header?

      Originally posted by Ossian View Post
      It sounds (IMHO) like a piece of sales Bulls**t
      Only the RAID 5/6 variants have "parity" - Raid 1+0 is a mirror of stripes (or is it a stripe of mirrors) so parity (check blocks) should not come into it.

      However there is clearly some problem!

      Advice is generally good though - any doubts about drive reliability normally means rebuild as soon as you can.
      Thanks Ossian. My thoughts precisely, despite Dells insistence I cannot see what 'Parity' is involved in Raid 1+0 let alone in the Raid header.

      My concern is that I do not want to get the customer to agree to spend the costs to wipe this array and start again if that is not going to fix the problem. In fact even if it does they will still have a legitimate argument about how this could have occurred on their brand new Dell server.

      They have already replaced the Raid Controller and cables but not the backplane. We have also had one sudden disk failure on this server.

      The consistency checks only find inconsistencies say 50% of the time, the other 50% of the time Consistency Checker runs fine.

      So question is what to do next. There is clearly still a fault but what is it; is it one of the disk controller circuits, the backplane or something else? How can we tell? It certainly doesn't sound like rebuilding the array is going to make any difference.

      Comment


      • #4
        Re: Corrupted parity in the Raid header?

        Assuming the kit is under an on-site warranty, I would be pushing Dell to get an Engineer out and fix it - they are bound to have some portable imaging tool to image the array and restore it.

        If not, does your backup software (whatever it is) support bare metal restores, in which case the risk and downtime are greatly reduced.

        I certainly would deal with it sooner rather than later - the risk is that, while things are OK (most of the time) now, any drive issues can only get worse and everything could be lost
        Tom Jones
        MCT, MCSE (2000:Security & 2003), MCSA:Security & Messaging, MCDBA, MCDST, MCITP(EA, EMA, SA, EDA, ES, CS), MCTS, MCP, Sec+
        PhD, MSc, FIAP, MIITT
        IT Trainer / Consultant
        Ossian Ltd
        Scotland

        ** Remember to give credit where credit is due and leave reputation points where appropriate **

        Comment


        • #5
          Re: Corrupted parity in the Raid header?

          Originally posted by Ossian View Post
          Assuming the kit is under an on-site warranty, I would be pushing Dell to get an Engineer out and fix it - they are bound to have some portable imaging tool to image the array and restore it.

          If not, does your backup software (whatever it is) support bare metal restores, in which case the risk and downtime are greatly reduced.

          I certainly would deal with it sooner rather than later - the risk is that, while things are OK (most of the time) now, any drive issues can only get worse and everything could be lost
          Have pushed Dell and all they do is repeat this lark about corrupted parity in Raid header, despite there being no parity involved in Raid 1+0. They apparently do not have any portable imaging tool available which can image and restore the array. Moreover they claim that this is not covered by their warranty (despite customer paying extra for their ProPlus warranty) and it is down to us to delete and recreate the array. According to them it is not a hardware problem.

          Backup software is standard Windows Server Backup as ships with Server 2012 R2. So no bare metal restore supported.

          Problem is we really don't know how to deal with this problem, no one else seems to have heard of the 'corrupted parity in Raid header' error before.

          Comment


          • #6
            Re: Corrupted parity in the Raid header?

            If it truly exists Dell should be able to provide documentation or references to this. Have you asked them to provide details so that you can determine exactly how this works and how you can prevent it from happening again?

            Have you also asked this Q on Dell's server forums?
            A recent poll suggests that 6 out of 7 dwarfs are not happy

            Comment


            • #7
              Re: Corrupted parity in the Raid header?

              Have you updated the firmware on the card?

              Comment


              • #8
                Re: Corrupted parity in the Raid header?

                Originally posted by wullieb1 View Post
                Have you updated the firmware on the card?
                Yep, all firmware and windows drivers have been updated to the latest.

                Comment


                • #9
                  Re: Corrupted parity in the Raid header?

                  get one disk replaced
                  then get the next disk replaced
                  then get the backplane replaced

                  keep pushing and yelling at them till that's done
                  then, and only then, consider their option to rebuild the entire array.

                  escalate it up the chain to the support manager if you have to.
                  keep escalating it.

                  keep repeating you were told by a "trained dell engineer" that your RAID1+0 array has a "parity" error despite it not having any parity.

                  Ask the engineer for proof of the "parity error"
                  keep pushing them.
                  Please do show your appreciation to those who assist you by leaving Rep Point https://www.petri.com/forums/core/im.../icon_beer.gif

                  Comment


                  • #10
                    Re: Corrupted parity in the Raid header?

                    Originally posted by tehcamel View Post
                    get one disk replaced
                    then get the next disk replaced
                    then get the backplane replaced

                    keep pushing and yelling at them till that's done
                    then, and only then, consider their option to rebuild the entire array.

                    escalate it up the chain to the support manager if you have to.
                    keep escalating it.

                    keep repeating you were told by a "trained dell engineer" that your RAID1+0 array has a "parity" error despite it not having any parity.

                    Ask the engineer for proof of the "parity error"
                    keep pushing them.
                    Thanks tehcamel, was just looking for a sanity check in case I was wrong before I start shouting at them.

                    Comment


                    • #11
                      Re: Corrupted parity in the Raid header?

                      Originally posted by Nick_C View Post
                      Thanks tehcamel, was just looking for a sanity check in case I was wrong before I start shouting at them.
                      I'm with the others here.

                      There is no parity on a 1+0 so get cracking them with the whip and make sure you escalate it to a level 3 tech.

                      Comment


                      • #12
                        Re: Corrupted parity in the Raid header?

                        You could tell them the parroty is demised, has climbed the immortal coil and ceased to be; it's effing snuffed it. (Sorry, extra silly mood tonight)
                        1 1 was a racehorse.
                        2 2 was 1 2.
                        1 1 1 1 race 1 day,
                        2 2 1 1 2

                        Comment


                        • #13
                          Re: Corrupted parity in the Raid header?

                          Latest from Dell is that some inconsistencies found in Consistency Check is normal. As long as Patrol Read does not find any errors then safe to ignore. Suggestion was that the inconsistencies may be coming from Windows writing to the disk at the same time.

                          So question is does anyone else run regular Consistency Checks on a Raid 1+0 array and if so do you get regular inconsistencies or do they all pass clean?

                          Comment

                          Working...
                          X