Fixed issues

This article lists issues present in previous releases that are fixed in this release.



  • Live migration, storage live migration, and VDI migration can fail for VMs that have no attached VIFs. After this failure, the VM hangs in shutdown mode. (CA-327906)

  • When intra-pool storage live migrations complete, this information is not noted. As a result, the operations are still listed as in progress. These ‘in progress’ operations can block the host from performing actions that require a host evacuation, such as: upgrading or updating the host. (CA-326621)

  • During toolstack restart, xcp-rrdd can use 100% of the CPU for a long time. (CA-322045)

  • If you add a large number of SRs, XAPI might run out of file descriptors. (CA-322044)

  • If the HA daemon has previously failed on startup, HA might not be able to start. The following error message is displayed: “HA could not be enabled on the Pool because a liveset could not be formed: check storage and network heartbeat paths.” (CA-315962)

  • The vCPU count for dom0 was incorrectly set to the number of CPUs on the host. (CA-312644)

  • Sometimes, the storage garbage collection process might run continuously, taking snapshots, coalescing them, deleting the snapshots, and then repeating on an almost permanent basis. This introduces high IOPS load for little benefit. (CA-328536)

  • When read caching is enabled it is significantly slower to read from the parent snapshot than from the leaf. This is now improved for sequential reads. (CA-327558)

  • If your Citrix Hypervisor server was originally installed with XenServer 6.5 or earlier and then upgraded to XenServer 7.6 (through an upgrade path of any of the intervening releases), the logging capability can fail to work. (CA-319463)

  • Functions available in xenstore utilities have been extended. (CA-327088)


  • If you forcibly shut down a VM that has an AMD MxGPU attached, or the guest OS shuts down abnormally, your Citrix Hypervisor server might experience memory corruption. Perform a co-operative shutdown to ensure the hardware is correctly cleaned up during the shutdown process. (CA-297891)

  • On rare occasions, VMs with attached NVIDIA vGPUs can become stuck in a yellow state at VM start. This is likely to be caused by one of the vGPUs having compatibility-metadata set when it should not. To remove the metadata, suspend and resume the VM. (CA-312226)


  • When starting multiple VMs as the same time after enabling the PVS Accelerator, a Target Device VM can become stuck at PVS Server login with the error “Login request timeout”. (CA-314317)

  • If you run your Windows VMs with the viridian flags enabled, the VM might crash. (CA-321550)


  • If the username or password contains url-encoded characters, it is not parsed correctly. This issue causes the Citrix Hypervisor installation to fail. (CA-329747)


  • A GFS2 SR can raise a false ‘Failed to attach storage on server start’ alert, due to asynchronous plug of clustered storage. Check the SR status in XenCenter to determine whether the alert was a false positive. (CA-311625)

  • The garbage collector process for the storage manager generates zombie processes. This error can cause Citrix Hypervisor to slow down and for processes to fail as process IDs become unavailable. (CA-327382)


  • In XenCenter 8.1.0 and 8.1.1, while applying a livepatch to a Citrix Hypervisor or XenServer pool, XenCenter can attempt to migrate a VM to the host it is already on. XenCenter 8.1.2 was released on 6 Mar 2020 and includes a fix for this issue. XenCenter 8.1.2 is available on the Citrix Hypervisor downloads page.

  • In the XenCenter 8.1.0 Alerts view, if you choose to dismiss an end-of-life alert or sort a list of alerts that contains an end-of-life-alert, XenCenter can crash. XenCenter 8.1.2 includes a fix for this issue. XenCenter 8.1.2 is available on the Citrix Hypervisor downloads page.

  • Sometimes, the XenCenter storage live migration or the Import OVF/OVA wizard does not allow you to proceed and the Next button remains disabled. When this occurs, restart XenCenter and attempt the process again. (CA-314346)

  • In XenCenter 7.6 and earlier, XenCenter did not prevent the creation of VDIs that are larger than the maximum supported size. (XSI-465)