Announcement

Collapse
No announcement yet.

Unable to delete checkpoint subtree - subtree contains recovery checkpoint

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

  • Unable to delete checkpoint subtree - subtree contains recovery checkpoint

    Hi guys,
    I'm in a bit of a pickle, our backups have created a large subtree chain within server 2012 hyper-v R2 I am unable to delete the subtree as it states that it contains a recovery checkpoint. If I browse to the folder where the files are stored I can see the

    HD1.vhdx
    HD1_*****-*****.avhdx

    The VM is booting and working I just need to clear out the checkpoints. If I try to create a checkpoint to export I get an error stating that its exceeded 50 checkpoints. Any ideas? Thanks

  • #2
    On a side note - can I export the VM without any checkpoints?

    Comment


    • #3
      This is why all the documentation states that checkpoints (snapshots for the old-fashioned) should not be a substitute for backups!
      Second question - export can be done at a particular checkpoint and will include all previous ones - right click the checkpoint and "export", not the VM
      First one - suggest you merge checkpoints - I would do it from most recent backwards
      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


      • #4
        Thanks for getting back to me, there is a backup solution in place the checkpoints from what I can see where created when the backup process ran as the sub tree shows all the checkpoints with -Backup on them which was what threw me as I know there were no checkpoints in place for this vm. If I try and export the checkpoint it states export failed but if I browse to where the disk files are for that checkpoint they are there.

        Just to confirm a merge of the checkpoints wont lose the data? Also where I am confused - is the VM referencing all the checkpoint\AVHD files as its running, as obviously changes at point A are available at point D - ie files on the system etc.

        And to confirm - best practice for manually merging, right click on the settings for that checkpoint and check the disk ID's and then merge those ones into parent ones (assuming I do it on the latest not the one labelled NOW)?

        Sorry for all the questions I just want to get it right




        Comment


        • #5
          Merge (deleting checkpoints) combines the AVHD with the source VHD so you will not lose anything (if paranoid, copy the VM first). Personally I shut down the VM before merging although in theory you can merge running VMs. Also I have occasionally had "broken chain of differencing disks" when merging multiple snapshots, so I tend to merge from newest back to oldest
          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


          • #6
            Thanks Ossian this is extremely helpful and yes I'm always paranoid when data is involved. Is it a correct process that when a merge is completed you have to modify the VM so it knows which disk to use? Or should it update to point to the correct disk all be it an avhd(x)

            Comment


            • #7
              It should update automatically - it always has for me
              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


              • #8
                Absolutely: shut down the VM, merge the latest checkpoint & wait for that completion, and the file pointers will be updated by the Hypervisor. After all, the Hypervisor adjusted all the pointers when the checkpoints were created in the first place.
                *RicklesP*
                MSCA (2003/XP), Security+, CCNA

                ** Remember: credit where credit is due, and reputation points as appropriate **

                Comment


                • #9
                  Thanks again guys this is really helpful info. Just to double triple check the NOW stage of the disk has an ID of _BE45676****** and the last checkpoint before that has an ID of _A0D145****** with a date stamp of 28/2/2017 when I go into the edit disk to merge do I select the current\now avhd(x) or the last checkpoint avhd(x) to merge? Thanks

                  Comment


                  • #10
                    Why are you doing it in "Edit Disk" - I would
                    a) Shut down the VM, then make an "oh shit" copy
                    b) Select the last checkpoint in HV Manager
                    c) Delete said checkpoint
                    d) Watch the most recent AVHD merge with its predecessor
                    e) On success, repeat until no more AVHDs
                    f) On failure, copy the "oh shit" back to the original location and fire up plan B

                    If you do it via Edit Disk, you may find HV does not pick up the merge
                    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


                    • #11
                      I was doing it like this because there is no option to delete the checkpoints due to them being created from a Veeam backup process (as said I didn't create these checkpoints all I want to do is get rid of them )

                      If I try and delete the checkpoints from within Powershell I get an error - delete failed. I referred to this guide https://blog.workinghardinit.work/20...rtual-machine/ which states if you have no other option you can manually merge the checkpoints - hence the pickle the VM is in. I think there is a known issue with backups leaving the VM in a locked state

                      https://support.microsoft.com/en-gb/...a-locked-state

                      I have a hunch that's what caused this.

                      Comment


                      • #12
                        Am I correct in thinking I am safe to do a https://technet.microsoft.com/en-us/.../ee656415.aspx Disk2VHD and then re-import those disks to the VM?

                        Comment


                        • #13
                          Ah, sorry, lost track of the thread somewhere there

                          OK, if using Edit Disk, you will probably need to tell HyperV manager that the VM has new disks attached

                          I think Disk2VHD is physical to virtual only
                          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


                          • #14
                            Thanks Ossian for confirming - I thought that was\is the case. I'm still not 100% sure when merging - am I merging the NOW disks or the last Checkpoint disks? Apologies if this has been answered I cant see it anywhere.

                            Comment


                            • #15
                              When I merge disks, it is always the more recent with the less recent:
                              VHD + AVHD => VHD
                              VHD + AVHD(1) + AVHD(2) => VHD + AVHD(1) => VHD (of that makes sense)

                              IIRC Edit Disk has an option to merge to a new disk (leave originals unchanged)

                              As I suggested, take the time to copy the files and work with the copy
                              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

                              Working...
                              X