Try to match the alerts by message type, not by title, because for pooled hosts the title contains the host name followed by the friendly name, while for standalone hosts the host name is not included
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
Improved the ReadHttpHeaders function, by making sure that a chunk is read fully from the stream (the number of bytes read by the Stream.Read function can be less than the number of bytes requested, so we need to continue reading until the required number of bytes is read)
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
* CP-19605: Improve the performance of the CreateNewRootNode function
Make GroupKey a class and override GetHashCode and Equals
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
* CP-19605: Remove debugging lines
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
Also changed ApplyPatchAction and ApplyUpdateAction to install a single update to a single host, so that the appropriate action title can be displayed if an action fails. And replaced 2 hardcoded strings with resource strings.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
raises XenCenter error on XS 6.1
When Text is null error is raised by split() function, we
need to handle that case.
Signed-Off-By: Sharath Babu <sharath.babu@citrix.com>
When working on this ticket (see previous commit), I noticed that the WebClient object used in DownloadAndUnzipXenServerPatchAction did not use XenCenter's proxy settings. This commit fixes this issue. By default (with no proxy being set) the proxy variable will be null as expected.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
Using standard .NET Framework functionality instead of the custom implementation of the HTTP stack. (To download a single file.)
Note that this .net class WebClient is already being used to download hotfixes. However, in that code the proxy is not set, I am fixing that in a new commit on the same ticket.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
A wrong api was being tested against the RBAC roles. Removed
vmss.protect_now and added vmss.snapshot_now.
Signed-Of-By: Sharath Babu <sharath.babu@citrix.com>
Disabled the ability to select unreachable hosts and in addition
added a try/catch block to hide error message resulting from trying
to clean up an unreachable host.
Signed-off-by: Letsibogo Ramadi <letsibogo.ramadi@citrix.com>
For ElyOrGreater hosts, we try move the existing VIFs to the desired networks. We will only destroy and create new ones for older hosts or when a corresponding VIF cannot be found in the network mapping.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
Added "VM.pool_migrate" to the list of tasks suitable for meddling actions.
Removed the usage of MeddlingActionTitle other_config key, because it wasn't working as intended for two reasons:
- If XenCenter starts an action that does multiple async api calls, so multiple tasks, the action's title is assigned to all tasks as MeddlingActionTitles, so the second XenCenter instance would create multiple meddling actions with the same title.
- When a second XenCenter instance tries to see if a task is suitable for a meddling action, the MeddlingActionTitle is not yet present in the task's other_config, so the task is ignored in most of the cases.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
When getting the control domain VM, we should use the control_domain only if it is not null. there is no need to check the version (which was, however, looking at the API version of the master)
Remove the VDI only on a best effort basis if the upload failed in the UploadSupplementalPackAction. This will make XenCenter to display the message from the original exception and not from the removal action if the latter had also failed.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
Fixed a typo that caused XenCenter to show a Trace error dialog (and to fail to do the clean-up) after applying an update in manual mode. (when RBAC is involved)
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
- Replaced the information tooltip with a warning icon and label at the bottom of the page.
- Added some null checks
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
have the wizard do it before we get there. Also, replaced the labels in such way
that, if there should be lag, the user will not see the draft message used only
on the designer.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
- On the PVS Cache Configuration dialog, disable the cache storage fields with an information tooltip stating that this PVS cache storage cannot be changed because it is in use.
- Also set the minimum width for the splitter panels, to avoid display issues when the splitter is moved and either of the panels becomes unusable.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
This change implements a lightweight retry logic for downloading updates to protect against temporary network issues.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
Added an extra parameter to the CreateVmFastAction constructor, which is used to specify whether to set the VM's property IsBeingCreated. In order to set this property, we need to wait for the newly created VM to appear in the cache. But we cannot do this in the action test, because the cache is not being updated. Therefore the action needs skip this step when running from the tests. The IsBeingCreated property is only used in the UI, to refresh the PVS tab after the VM is created, so the accuracy of the action test (CreateVMFastActionTest) is not affected by this change.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
* [CA-233454] PVS tab doesn't show a new VM when it's created
Refined the rules for not adding a VM to the table, if it is a template (thus not_a_real_vm), and it has the __gui__ prefix (thus hidden), we still add it, but hide it.
When its name is changed (to remove the __gui__ prefix), we update its name and re-calculate whether it should be visible (in the case of a new VM this will be true once the __gui__ prefix is gone). Also resort the table if a node changes from hidden to visible, because it appears as an addition to the table.
Signed-off-by: Callum McIntyre <callumiandavid.mcintyre@citrix.com>
* [CA-233454] Update criteria for VM visibility to include is_a_real_vm
is_a_template is changed before the name_label removes the __gui__ prefix, so this works with no other changes to the vm property changed event.
Signed-off-by: Callum McIntyre <callumiandavid.mcintyre@citrix.com>
* [CA-233454] Update when the VMs are shown in the PVS list
New observable property IsBeingCreated for VMs, set to true when they're made a hidden object, and false when they're removed from hidden objects (both in CreateVMAction). In the PVS Page when this is set to false, we re-evaluate whether a VM can be shown. This means that new VMs show here at the same time they're added to the tree (only different is tree refresh time), instead of far earlier (and before their networks were added).
Signed-off-by: Callum McIntyre <callumiandavid.mcintyre@citrix.com>
* [CA-233454] Set IsbeingCreated in the CreateVMFastAction
* [CA-233454] Properly support the VM Fast Create action
Further changes to CreateVMFastAction, to ensure it works with PVS tab - use the __gui__ prefix when the VM is created and then change it back just before showing.
* [CA-233454] Small logic adjustments/tidying up
If batch updating fails in a pool, this code will do the clean up by
running the appropriate (next in the queue) RemoveUpdateFilesFromMaster action. In addition to Pool_update.clean(), Pool_update.destroy() will be called when needed.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
confirmations are displayed. Added checkboxes to the dialogs requesting confirmation
of dismissal of notifications so the user can opt to skip this step.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
Changes as discussed on the ticket: No longer disable the rows when they have a warning in the dialog, support warning about storage configuration as well as PVS configuration, and also include these warnings in the PVS page table.
The Pool_update does not always covers all the VDIs XenCenter creates. These orphaned ones have to be deleted as well.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
on a background thread and show a spinner while this process is going on. Also,
removed certain duplicate controls and made some minor layout changes (margins,
sizes, column widths).
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
The resume on server uses the CrossPoolMigrateCommand to implement the migration of the VM to the desired host. Before this change that command had no support for resuming a VM after migrating it, so the VM was not resumed. With this change, the CrossPoolMigrateWizard can take an optional (default false) parameter to restart the VM after migration. When this is true, the migrate action becomes a MultipleAction, first migrating and then restarting the VM.
Signed-off-by: Callum McIntyre <callumiandavid.mcintyre@citrix.com>
Also changed the page to use different resource strings for the Storage type in the right hand panel (no hotkeys should be shown there)
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
- show a different message when deleting a Cache Configuration (a PVS_site object) that is actively in use by a PVS site
- some more text changes on the "Configure PVS Cache" dialog
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
On the General tab XenCenter shows a list of updates that requires the host to be restarted. This is shown for applied updates for which the required guidance hasn't been done (eg. restartHost or restartAgent after-apply-guidances)
This commit fixes a regression that caused restartToolstack warnings to be not shown in the list of warnings.
The displayed messages are also improved, XenCenter will show Toolstack restart or host restart appropriately and not just restart only as it used to.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
We have some new RRDs - read, write, read_latency and write_latency. This adds support for them by adding a friendly name and using it to display them. It also adds some filtering (see GraphHelpers.cs) because for backward compatibility io_throughput_read and io_throughput_write (which differ from read and write only in scale) are still available, but shouldn't be displayed in XC. If the new read or write are available, we filter these old datasources out.
Signed-off-by: Callum McIntyre <callumiandavid.mcintyre@citrix.com>
Introduce fails with a Failure if the uploaded update has already been introduced, in that case XenCenter will use the existing Pool_update
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
- added a new precheck, called "Live patching status" which checks is a reboot is needed for each host. If reboot is needed, the check will display a warning, otherwise it will show a green tickbox, saying that the server does not need to be rebooted, and if live patching is used, this information will be included as well.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
- Renamed Add/Remove PVS Site buttons
- Added a note on the PVS Cache configuration page, regarding the recommended cache size and the allowed size for "memory only"
- Added a warning about the Site configuration not being complete until the PVS part is done as well (only visible if PVS_uuid is not set for a PVS_site)
- Added PVS Read Caching status on the VM's general tab
- The absolute maximum allowed for a cache size is 2TB
- Changed the rubric on the Enable PVS read caching dialog
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
2 new radio buttons in Connection Options page, for Basic and Digest; Digest is the default as it's the most secure.
New proxy authentication method setting, which is used to configure XenAPI's HTTP class and the .NET AuthenticationManager (which handles which authentication schemes can be used by the .NET web classes).
The new setting is also transferred and used by the Health Check service.
The bug noted in CA-214653 also occurs here, but the work-around on PR#1201 for that bug will also work for this.
Signed-off-by: Frezzle <frederico.mazzone@citrix.com>
- We shouldn't include the hidden SRs in the list of available SRs in the Migration wizard
- Updated the SR.SupportsVdiCreate function to return false for the Memory SR; this fixes other places where the Memory SR might be visible (e.g. all places where SrPicker is used)
- Also fixed the same issue in the Attach disk dialog
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
Normally, a heartbeat is done every 15 seconds by getting the server time, and starts as soon as connection to host is established, but if an error occurs XC retries once more in another 15 seconds before actually closing the connection to the host.
Now, if a 407 Proxy Authentication Required error occurs on the initial heartbeat then XC does not retry again in 15 seconds; it will close the connection immediately. The disconnection was instant to up to 2 seconds, depending on time taken to get server time (testing with FreeProxy was ~instant and CCProxy was ~2 seconds). This is only a work-around, as the exact reason for this behaviour is still unknown.
Signed-off-by: Frezzle <frederico.mazzone@citrix.com>
- check the current number of vCPUs (by getting the latest value from the cache) before trying to change it.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
