• CBT: the thread to centralize your feedback

    Pinned
    443
    1 Votes
    443 Posts
    398k Views
    R
    Hi all, We recently upgraded our production pools to the latest XCP-ng 8.3 release. After some struggles during the upgrade (mostly around the pool master), everything seems to be running fine now in general. However, since upgrading, we’re seeing much longer durations for certain XAPI-related tasks, especially: VDI.enable_cbt VDI.destroy VDI.list_changed_blocks (during backups) In some cases, these tasks take up to 25 minutes to complete on specific VMs. Meanwhile, similar operations on other VMs are done in just a few minutes. The behavior is inconsistent but reproducible. We’ve checked: Storage performance is normal (LVM over local SSD) No I/O bottlenecks on the hosts No VM performance impact during these tasks It seems to affect CBT-enabled VMs more strongly, but we’re only seeing this behavior since the upgrade to 8.3 — especially after upgrading the pool master. Has anyone else seen this since upgrading? Is there a known issue with CBT or coalesce interaction in 8.3? Would love to hear if others experience this or have suggestions for tuning.
  • Feedback on immutability

    Pinned
    51
    2 Votes
    51 Posts
    10k Views
    olivierlambertO
    Thanks for your feedback, we'll discuss internally if there's any other possible approach (and I'm not sure).
  • VDI missing after trying to fix backup issue

    Solved
    5
    0 Votes
    5 Posts
    26 Views
    A
    @olivierlambert xe vdi-forget uuid=c6853f48-4b06-4c34-9707-b68f9054e6fc xe sr-scan uuid=6b24cd1c-22ad-0994-5b6b-a75389a6ddba xe vdi-list uuid=c6853f48-4b06-4c34-9707-b68f9054e6fc params=sr-uuid sr-uuid ( RO) : 6b24cd1c-22ad-0994-5b6b-a75389a6ddba yup, that worked, thanks
  • VDI Chain on Deltas

    16
    0 Votes
    16 Posts
    364 Views
    nvossN
    @olivierlambert Unfortunately even when shutdown the migration from NAS to local storage fails with the same SR_BACKEND_FAILURE_109 error. At this point we're trying copying the VHD, destroy the VM, re-create with the copied VHD on at least one machine. But it definitely appears to be more than one VM with the problem, unless the one machine causing the problem is preventing coalescing on all the other machines too somehow. The previously identified machine is the one I'm starting with.
  • 0 Votes
    3 Posts
    25 Views
    F
    @florent Thank you.
  • Error: invalid HTTP header in response body

    19
    0 Votes
    19 Posts
    528 Views
    florentF
    @markxc thank you that is the part I wanted I think someone on the xapi side should look into it
  • XOA 5.107.2 Backup Failure via SMB and S3 (Backblaze)

    2
    1 Votes
    2 Posts
    50 Views
    planedropP
    Going to give it more time, but restarting all the backups seems to have fixed the issue. Unsure why they would fail once and then resume just fine though.
  • Backup Failure to BackBlaze

    6
    4
    0 Votes
    6 Posts
    122 Views
    A
    Thank you for the update. Last night before I saw your replys that is exactly what i did. I purged the backup's for the two VMs from BackBlaze. The backup job completed successfully while copying over the previous backup images. [image: 1749549311956-screenshot-2025-06-10-055256.png]
  • Mirror Incremental Backup Error: Lock file already held

    Solved
    4
    1
    0 Votes
    4 Posts
    105 Views
    E
    Thanks @olivierlambert and @florent, that seems to have been the trick and fixed the issue.
  • Continuous replication problem with multiple SR

    Solved
    8
    7
    0 Votes
    8 Posts
    223 Views
    florentF
    @Gheppy it's always easier to fix when we can build a minimal setup that reproduce the issue thanks for your report
  • XO - Files Restore

    24
    2
    0 Votes
    24 Posts
    841 Views
    P
    @lsouai-vates Is this a description of the cause of the problem ? From what you describe, if one selects a backup containing a LVM partition scheme, it should be able to be mounted at least the first time ? As in here: I verified that this machine is using LVM first, and when selecting it, I immediately selected the large LVM partition to try to restore a file from that (which failed). [image: 1748956125525-e7148e43-aa01-433d-8af4-a936e30550e6-image.png] The machine (Debian 12) running XO do not itself use LVM, so "ubuntu-vg" should be free for mounting this first time.
  • 0 Votes
    2 Posts
    122 Views
    C
    This happens to me on 8.2 with one specific VM. It doesn't always happen but about once a week (it feels like) the backup will get stuck in Starting and it's always, or almost always, the same VM it's stuck on. I usually reboot the VM running XO and that makes the backup task change to Interrupted and then the next scheduled backup usually works. Would be nice to know if there's something that's causing this that I can try to fix. Not a world-ending problem but it's a nuisance since it stops all future backups for that Job until it's resolved. I'm happy that all the other VMs in that job get backed up at least.
  • Our future backup code: test it!

    128
    5 Votes
    128 Posts
    8k Views
    Tristis OrisT
    I created a new CR job for another VM and it worked. However, it didn't work with the XO VM. So maybe the root cause of the problem is that the old CR copies have disappeared. Maybe they still exist, but I can't find them?
  • Backup success or failure ?

    1
    3
    0 Votes
    1 Posts
    48 Views
    No one has replied
  • Can't save backup task after removing proxy - BUG

    1
    5
    0 Votes
    1 Posts
    53 Views
    No one has replied
  • Suspicius presentation of time in backup

    12
    3
    0 Votes
    12 Posts
    360 Views
    lsouai-vatesL
    @ph7 I ask the XO team about it, thanks for the report.
  • Moving management network to another adapter and backups now fail

    15
    0 Votes
    15 Posts
    506 Views
    tjkreidlT
    @syscon-chrisl Note that when changing the management interface, it's highly recommended to reduce the PMI down to just one NIC on all your hosts before you make the change. That said, it's always a scary thing to do and as others have stated, best avoided if at all possible. Making sure all hosts are at the same hotfix levelasand that their NICs are all in the same order and speeds are essential requirements.
  • 0 Votes
    9 Posts
    333 Views
    tjkreidlT
    @yeopil21 Run top, xentop, and iostat to see if dom0 and.or the storage device might be a bottleneck. The configuration of your storage device can also be a big factor, and in some casaes, various performance tweaks are possible. The specific configuration would be helpful: connectivity (NFS, fibre channel, iSCSI), number and size and speeds of disks, RAID configuration, total number of VMs resident on the device, number of independent SRs, provisioning (thin or thick), network (if not fibre channel) speed and settings. STorage optimization is a bit of an art and in many cases, can be the limiting factor, but as stated, so can the lack of dom0 resources. Also, what clock speed and number of CPUs are on your hosts?
  • Restoring a file from a CR Delta

    6
    0 Votes
    6 Posts
    146 Views
    olivierlambertO
    That's great, those are companies with 100% backup dedicated products We cannot be everywhere and we have to prioritize on what most people expect. Recovery capabilities on a second site is like one of the most needed feature, that's why we did CR. From a technical stand point, when a VM is replicated, it's on another SR and XO doesn't have any direct access to the SR to mount it. IDK how VEEAM is doing, likely accessing the SR or similar (at least not on XCP-ng )
  • warm migration vs broken backup link

    5
    1
    0 Votes
    5 Posts
    148 Views
    lsouai-vatesL
    @kosa7 from what the XO devs told me : for now you have to handle it completly manually, since we have separated collection a not a relationnal db... You could maybe use a smart mode backup with a VM tag. We plan to implement a relational database in the future, which would at least offer the possibility to handle this. "Reasonnably : warm migration should be more a one off , so the user must handle the consequence. If he uses this regularly, he should add a tag to the VM " DBPRODBACKUPPLZ" and then configure his job to use this tag" FYI @olivierlambert