- With these changes, a stopped or suspended VM can be moved across pools; this is performed as a vm migrate operation.
- The intra pool move will continue to be a vdi.copy + destroy operation for stopped VMs, but for suspended VMs we need to do it a as vm migrate operation.
- We use the Cross Pool migrate wizard for both intra and cross pool move operations, but introduced a wizard mode parameter in order to adapt the wizard to the specific operation we want to perform (migrate, move or copy)
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
-Added missing parameter to get_required_space plugin call.
-The required space is not shown in the notification when we are unable to tell how much space is required .
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
- The problem was that we didn't create a vgpu if the vgpu_type was null. But for Linux HVM, this is gpu passthrough case.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
-Added support for CIFS (SR type, SR Wizard)
-Added some Debug.Asserts to the Create SR action to watch for clear text paswords
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
- When the install fails, log the full error description
- Added a specific exception for the supp pack installation failure
- Better action title (depending on the number of servers it applies to)
- Added a "More Info" button on the "Update" page, which displays a message with more information on the supp pack installation failure
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
- Added a "Integrated GPU passthrough" section to the PoolGpuEdit page.
- This section is only visible for a host that has GPU capability and the enabling/disabling of integrated GPU passthrough is not restricted.
- With this addition, the GPU page can now be displayed for a pooled host as well (previously only pool or standalone host)
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
- Disable the Refresh button while a refresh is in progress.
- Introduced ExecuteContainerPluginAction - an action that stores the container it was called for, so that on completion we can decide whether to update the view or not (if the container displayed has changed, then we shouldn't update the view with the results from a previously selected container).
These changes apply to container Processes and Details tabs.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
- Pool_patch.precheck can fail with the xapi error OUT_OF_SPACE when is trying to copy the patch from master to the slaves. This happens before the patch-precheck checks if enough space is available for installation. Therefore is different from the error we are currently processing, PATCH_PRECHECK_FAILED_OUT_OF_SPACE
- If the precheck fails with OUT_OF_SPACE then we try and retrieve the disk space requirements and report a HostOutOfSpaceProblem.
- Changed the PatchPrecheckCheck to try and find problem from a xapi failure, as well as from the result of the precheck call.
- Also removed ampersand from the the link labels (More Info and Clean up)
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
- The cloud config parameters page is only visible on a VM that already has a config drive.
- The cloud config parameters can only be changed when the VM is halted.
- In the New VM wizard, if a template is selected that already has a config drive, them the wizard will display the existing configuration, not the default one.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
-Added HostOutOfSpaceProblem to let the user to clean up disk space when getting PATCH_PRECHECK_FAILED_OUT_OF_SPACE error at precheck stage in the patch install wizard.
-A bit of refactoring of DiskSpaceRequirements class in order to be reused here
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
New Page: Enlightenment on VM Properties Dialog, visible only for:
- VMs on Cream Or Greater hosts
- that can be enlightened, regardless of their power state
A VM can be enlightened if the key "xscontainer-monitor" exists in other_config.
A VM is enlightened if other_config:xscontainer-monitor if true.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
- don't update the progress bar when checking disk space (only the description)
- the color used for "in progress" action description (above the progress bar) should be the system default font color, not black
- actions should throw ArgumentNullException, not NullReferenceException
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
- Before starting the upload to the master hosts, check if there is enough disk space (check only performed for Cream or greater hosts)
- If enough space available the upload starts automatically; otherwise an error is displayed
- If we can free up enough disk space then we offer the option to Clean up. Otherwise we provide the user with the information on required and available space and the user will have to manually free up required space.
- Also Disable the Upload page for oem updates
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
UploadSupplementalPackAction:
- use http-put with the default timeout
- throw current exception in RemoveVDI (instead of encapsulating it in a new one)
- add other_config key before uploading to make sure the vdi has this flag set from the beginning of its lifetime
- upload progress bar now shows the total progress per row
Also included a cosmetic rewrite of PatchingWizard_PatchingPage,GetUpdateName function
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
- add an entry in the VDI's other_config to identify it as supp pack iso, so that we can easily find these vdis and remove if needed
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
- Select a supp pack in the hotfix wizard
- Recognise that it's a supp pack not a hotfix
- Check if there's enough space on a SR to contain it: (1) default SR if shared or (2) any shared SR or (3) local SR on each host in pool
- Create a VDI to contain it
- Upload the supp pack to the VDI
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
New properties:
- Pool.HasGpu = Pool has at least one PGPU
- Pool.HasVGpu = Pool has at least one PGPU that HasVGpu
- PGPU.HasVGpu = PGPU has at least one supported_VGPU_type that is not pass-through
New or modified helper functions:
- Helpers.GpuCapability = GPU feature not restricted (by licensing) and Pool.HasGpu
- Helpers.VGpuCapability = vGPU feature not restricted (by licensing) and Pool.HasVGpu
- Helpers.ClearwaterSp1OrGreater = API version is 2.1 or greater
The GPU dialogs are displayed as follows:
- GPU page on VM properties dialog: Visible only if VM.CanHaveGpu and the GPU feature not restricted (by licensing)
- GPU page on New VM Wizard: Visible only if VM.CanHaveGpu and the pool has GPU capability (Helpers.GpuCapability)
- GPU page on Pool properties dialog: Visible only if the pool has vGPU capability (Helpers.VGpuCapability)
- GPU tab: Visible only if the pool has GPU capability (Helpers.GpuCapability) and is Clearwater SP1 or greater
- On the GPU tab, the "Placement policy" panel: Visible only if the pool has vGPU capability (Helpers.VGpuCapability)
- On the GPU tab, the "Edit" button on the "vGPU types" panel: Visible only if the PGPU.HasVGpu and vGPU feature not restricted (by licensing)
Also:
- VM.CanHaveVGpu function renamed to CanHaveGpu
- On the GPU tab, renamed "Allowed vGPU types" to "vGPU types
- Added new page to the patching wizard, called "Upload" which displays the upload actions for each server and the progress of these actions
- MultipleAction: added the functionality to optionally stop the action on first failure
- Fixed the error where an existing patch downloaded from another server and then uploaded to a new one was not deleted on cancelling the wizard [CA-156788]
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
The problem was that the License Manager was listening to PropertyChanged events on the master only, not on slaves.
When the master's properties change (e.g. edition) we update the row in the license manager; but in some cases a slave (or more) hasn't been updated yet (as it may be updated in another event.from) and we think that the pool is partially updated.
Our solution is to listen to Host BatchCollectionChanged event, which is triggered once per cache update for the host collection (it any property changed for any of the hosts).
The sleep in the ApplyLicenseEditionAction is not needed anymore, nor is the extra call to update the cell after the action is completed, because the cell is getting updated correctly on the BatchCollectionChanged event.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
This fixes the following issue: event.from fails with "The underlying connection was closed: A connection that was expected to be kept alive was closed by the server.".
It happens because we end the connections to the slave hosts only after all hosts have joined the pool and in the meantime event.from on a slave fails.
We fix this by ending the connection to each host immediately after the host has joined the pool.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
