of the hosts to be upgraded and not only on the first one in the list.
For this purpose, the test has been moved to the prechecks, and the
failures are now shown as a non-fixable problem.
Further, specification of the network location of the installer files
has moved closer to the upgrade mode selection.
Also, moved the hotfix and upgrade related checks immediately after the
host liveness check.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
* CP-30335: Fix RPU wizard tests
* CP-30335: Add abstract class WarningWithMoreInfo and use it for two of the existing Warnings
* CP-30335: Add a new precheck which produces a warning if the Dom0 memory after the upgrade will be increased.
- if the Dom0 memory after the upgrade is known, then a warning is shown if this is higher than the current memory, regardless of the version of the upgrade
- if the Dom0 memory after the upgrade is not known (e.g. because the rpu plugin does not have the function that returns this), then a warning is shown if upgrading from a pre-Naples version
* CP-30335: Add a more detailed error message if the host evacuation fails due to not enough memory in the RPU and Patching wizards
* CP-30335: Add branding variable for version 8.0
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
* Simplification: use auto-properties.
* Removed unnecessary null check. Converted private methods to linq method chain.
* Removed unused file. Added null check.
* Corrected property names as it was difficult to distinguish between the alert,
the patch and the filepath.
* Removed duplicate lock.
* CA-296490: Rewrote the logic uploading single updates so as to facilitate uploading to multiple pools in parallel.
* Corrections as per code review.
* Account for the case of uploading from disk legacy updates that do not correspond to an update alert.
Removed unfriendly error messages (the failure is written out in the logs anyway).
* Wizard test corrections and some modernising:
- Only call cancel on the last page if the test allows us to.
- Do finish the UpdatesWizard test.
- Finish the RunWizardKeyboardTests using key strokes instead of button clicks.
- Ensure the Cancel/Previous buttons do not have focus when sending an Enter stroke.
- Use optional parameters in the WizardTest constructor and string interpolation.
* Renamed badly named property.
* Improved code cloning VMs by assigning new opaque_refs to cloned VBDs.
Compacted test object provider: use optional parameters instead of method overloads.
* Modifications to prevent multiple update of enabled state and ensure the wizard
buttons are updated last.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
or Host accordingly. Extended comparison of Problems to also take account of
the XenObject the Check was performed on, otherwise the Problems get mixed
up when there are homonymous members across pools.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
... because pre-Dundee hosts dont't have this license flag, but they might have it after the upgrade.
Also added a bit more logging in the function that generates the update plan actions
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
* CP-28431: Block updates and upgrades from Kolkata onwards if clustering enabled
* CP-28431: Rename files and change from Pool Check to [Host] Check
* CP-28431: Change description message
Signed-off-by: Seren Corbett <seren.corbett@citrix.com>
the liveness warning is not issued if the pool is disconnected. HAOffCheck tidy up.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
is enforced on all host post-liveness checks.
Added missing description to SafeToUpgradeCheck.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
a message in this case as the SRs are reconnected after reboot and this may lead
to duplicate messages. Also, simplified check-and-plug methods for PBDs to avoid
repetition of iterations.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
1) Always pool level selection of hosts, except uploading sup pack file
from disk.
2) Revert homogeneity precheck change recently. It won't happen in new version
update any more.
3) User can never update slaves to a higher version than master, so delete
master_version_not_compatible_problem.
Signed-off-by: Ji Jiang <ji.jiang@citrix.com>
* CP-28286: Support resume from a partially upgraded pool via Update wizard
Signed-off-by: Ji Jiang <ji.jiang@citrix.com>
* CP-28286: Fix comments of the previous commit
Signed-off-by: Ji Jiang <ji.jiang@citrix.com>
Because we couldn't control user's selected combination from
a mixed pool, and the version compatibility precheck exists already,
here just simply skip the precheck and issue a warning.
Signed-off-by: Ji Jiang <ji.jiang@citrix.com>
* CA-274082: Check status of SR uploaded updates during update installation
* fix review comments
Signed-off-by: Jisheng Xing <jisheng.xing@citrix.com>
it's obvious what it is. The Session's uuid will be deprecated from the API bindings.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
- The new "server selection" precheck is added for Inverness or greater pools and checks if all the servers in the pool are selected (if update not already applied) when applying an update that has the "enforce-homogeneity" flag set to true
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
prevent them from being serialised alongside the API properties. This will also
be useful for moving the API bindings out of XenModel.
Signed-off-by: Konstantina Chremmou <konstantina.chremmou@citrix.com>
When the update precheck fails with UPDATE_PRECHECK_FAILED_UNKNOWN_ERROR, we try to find the actual error in the failure's parameters (in the form of an xml string containing the error code) and display that error instead of the "unknown error".
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
- Ensure that the disk-space plugin functions are not called on Ely or greater hosts, except for the 'get_avail_host_disk_space' function, which is still needed
Signed-off-by: Mihaela Stoica <mihaela.stoica@citrix.com>
Changed the pre-check error message to refer to the absence of the Tools
when that is suspected to be the case.
Signed-off-by: Gabor Apati-Nagy <gabor.apati-nagy@citrix.com>
A bit of refactoring as requested by the reviewers
I decided not remove the NullReferenceExceptions and throw the Failures
immediately in order to avoid the additional error handling that seems to be unnecessary there. However, the reason I've decided to keep it as is is that it would change what gets logged and the performance gain would not be noticeable anyway. Furthermore, the current implementation (try-catchall) catches any other exceptions providing fallback to the default Failure.
