hotkeys; button enabled state; some memory leaks; added spinner while check for
updates is in progress; added alternative text for when automatic mode is not available.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
Implemented pre-check to check whether there is enough free disk space
before starting updating a pool.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
more obvious to the user that they can edit it. To this purpose change the button
Properties to Edit (and do the same on the Networking page for consistence). Also
show the name of the disk (ellipsized if too long) on the wizard's finish page.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
- Add a new update to the list of uploaded updates only when the upload action has completed successfully.
- Do not perform the disk space check if there is nothing to upload (update has already been uploaded).
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
in the calculation of the never-expiring licence ("greater than 3653 days" in
LicenceAlert and LicenceStatus vs "greater than or equal to 3653 days" elsewhere).
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
Despite the update sequences are correct by host, when they are flattened, the original order (order in minimal patches) was not considered properly, so a missing patch was scheduled for later in the sequence than it should have been, thus the upgrade failed with a precheck-error.
This bug occurs when we are given
* a pool of more than one host,
* with (at least) patchA not installed on 2 or more hosts
* patchA requires patchR that is not installed on at least one of the above hosts (but not not on all)
* the host that is missing patchR is ahead of other hosts in pool.Hosts.Cache
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
The PatchPrechecksOnMultipleHostsInAPoolPlanAction did not filter the patchmappings by masterhost, so with some probability, it could choose the wrong mapping (pool_patch object for the same update, but in different pool). This bug is being fixed in this commit, other changes only renamed (for more consistent naming) a class and removed an unused field.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
- The "Alarm Settings" action on Alerts page should open host's properties dialog for Dom0 alerts
- Also added help id for the Control Domain Memory Usage alert
Even though the correct row have been selected by the time the grid becomes visible, it is not displayed properly. This change set fixes the problem by selecting the row again.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
Bugs fixed:
* OK button not re-enabling when switching back to using existing credentials in Health Check Enrollment dialog
* Using empty credentials to authenticate with Citrix Insight Services does not disable OK button in Health Check Enrollment dialog
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
a selected item collection - there was no need to make temporarily each one of them
the command selection just to call CanExecute on them. (2) Do not mix the methods
returning the can't-execute-reasons (which can be unknown) with the tooltip property
(which can be null).
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
to null, otherwise the wrond text is shown for items with null can't-execute-reason
that are selected after items with non-null can't execute reason.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
FolderBrowser dialogs. Fixed text which caused label to look unaligned. Prevent
crash when attempting to launch the DownloadAppliance dialog with "http://" as URI.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
In the New Network wizard, when the "Single-Server Private Network" type is selected, the Network settings page should not display the MTU hint message, because it is not relevant to any controls on the page
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
title so we can tell the one from the other. For consistency I used a format similar
to Import/export XVA. I also removed the gerund from the latter as it cannot be
used when the action is completed.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
XenCenter no longer ignores checking if the master host already has the specified patch applied; it now checks before trying to apply it, which means we get no error in the Patching Wizard's Patching Page.
Signed-off-by: Frezzle <frederico.mazzone@citrix.com>
* [CA-214953] Multiple connection dialogs while attempting to connect
ConnectingToServerDialog now keeps track of the dialogs for each connection and focuses on them if they already exist, instead of creating new ones.
Signed-off-by: Frezzle <frederico.mazzone@citrix.com>
* [CA-214953] Multiple connection dialogs while attempting to connect
Moved code to handle open connection dialogs to XenConnectionUI, instead of being done inside ConnectingToServerDialog.
Only one error message appears now upon connection failure.
If connection dialog was minimized, and user double-clicks again to connect, it now returns to normal state on the screen as well as taking focus.
Signed-off-by: Frezzle <frederico.mazzone@citrix.com>
Also fixed a similar issue, where the broken SR was selectable (Create new virtual disk from the SR Storage tab)
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
VMstorage page uses the default colours of the controls, hence we only need to set
them on the VNCTabView and the linklabel does not need to be public.
Where (in code removed by this commit) we iterated through the planActionsByHost dictionary, the hosts were iterated through in the order how the dictionary was initialized. Despite the actions were added in the right order, they then were screwed up, because the Cache.Hosts can have the hosts randomly while we want the master to be first.
To fix the bug, I changed the data structure from a dictionary to a list that has the actions exactly in the (good) order, the order they were added, eliminating the risk of the actions being retrieved in the wrong order.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
Refactored to hide/show error controls instead of adding/removing them. Removed duplicate null check.
Do not use a second list to track active VM consoles.
Fixed display issue, now showing pools (and standalone hosts at the same level), not masters.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
- removed unnecessary casts
- no need to catch exception only to rethrow it; do not interrupt the call stack by using "throw ex"
- redundant check and value setting
- For Dundee and newer hosts: the feature is restricted only if the "restrict_ssl_legacy_switch" flag exists and it is set to true
- For pre-Dundee hosts: the feature is restricted if the "restrict_ssl_legacy_switch" is absent or it is present and set to true
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
VM with the hci-warn-before-shutdown flag. Minor changes to make the warnings for
rebooting and shutting down more consistent (preselect the No button for rebooting;
extended the Yes button text; modified warning text)
If the license does not allow VM migration, then cause the VMs to be shut down / suspended on the pre-checks page
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
The last pGPU's row was stretching to the bottom, causing the lowest select box to fill its space and go down too much.
Changed the select boxs' dock style and margin.
Signed-off-by: Frederico Mazzone <fredericom@citrite.net>
Changed heading from Recommended Updates to Required Updates following further discussions around wording
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
Showing recommended* hotfixes for servers when the data is available (Check For Updates xml has been downloaded and has minimalPatches node in it for the given server version). *:Not applied ones from the minimal set.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
* CA-77990: Report required actions for all VMs at once in the hotfix wizard / RPUW
* CA-77990: Resolve precheck problems in parallel in the Patching wizard
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
* CA-77990: Resolve precheck problems in parallel in the Patching wizard
Changes following code review: Made the ParallelAction work across connections and removed CrossConnectionParallelAction
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
* CA-77990: Modified ParallelAction to create queue on RunSubActions only, with the correct queue size
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
Added support for defining custom update.xml location in Registry. This
can be either a local file or an URL. This will be beneficial when testing XenCenter.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
- Changed the FolderAction class so that it can be used to delete or move multiple folders across connections, instead of having multiple actions (one for each holder) which can not run in parallel.
- The api calls that update the EMPTY_FOLDERS entry in pool's other_config are preformed last for all connections, avoiding cross-thread exceptions when two threads try to add/remove the same key.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
DownloadUpdatesXmlAction is called on PageLeave and not when the user selects Automatic Mode. This way we get the error message at the right time, also showing a progress dialog what makes the UX much better.
New tooltips for servers on the Select Server Page to to cover two more cases
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
All event handlers are now unsubscribed when AttachDiskDialog is closed, allowing its memory to be garbage collected.
Image for each VDI now references an existing image instead of creating a new object.
Signed-off-by: Frederico Mazzone <fredericom@citrite.net>
My commit fixes the error message so the dialog will show "Could not create SSL/TLS secure channel." instead of "An unknown error occurred". This text is localised.
Unfortunately we see only a WebException with Status=SecureChannelFailure with no other details - couldn't provide an improved error message
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
Added AD-related strings to japanese and chinese resource files; still need translation.
Changed F1/Help mapping of the AD upsell tab to go to the 'Managing Users' help page.
Signed-off-by: Frederico Mazzone <fredericom@citrite.net>
Implemented error handling.
If there has been an error in a pool, we stop executing actions in that pool immediately, and report the error once every other actions have finished. Error in one pool will not affect the upgrade process of other pools.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
Made textbox bigger in health check policy dialog.
Made textbox bigger in the 'Authentication with Citrix Insight Services'
section of the health check enrollment dialog.
Adjusted some control properties to correct the position of OK/Cancel
buttons.
Changed 'User name:' labels to 'Username:' in health check enrollment
dialog.
Changed Chinese and Japanese resource files to match English changes.
Signed-off-by: Frederico Mazzone <fredericom@citrite.net>
* CA-210732: XenCenter VM properties GPU always shows "None"
The error occurs when there is a gpu group with only one vgpu type: When setting the selected item in the vGPU combobox, XenCenter tries to select the group header instead of the vGPU type; because the group header is not selectable, we then default to "None".
With this fix, we ensure that two items are not equal if one is a group header and the other is a subitem.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
* CA-210732: XenCenter VM properties GPU always shows "None"
Changes following code review: Added the reverse case in the Equals function
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
* CA-210732: XenCenter VM properties GPU always shows "None"
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
Removed the first item instead of changing it - it is more logical to not to have it displayed here.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
- Re-position the version radio-buttons, so that is clear that the selected version applies to both Create and Reattach
- Pass the selected version to the attach command in the device_config parameter
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
Added a Precheck to the RPU wizard, to check if the hosts in the pool have different CPU feature sets (using the old feature check for pre-Dundee pools) and display an error, offering to shut down all VMs.
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
When an exception is reported during the RPU, XenCenter tries to re-enable the host. But when this fails, we should just ignore it and report the original exception.
Signed-off-by: Mihaela Stoica <Mihaela.Stoica@citrix.com>
