Sync #1
Open
Sync #1
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Sync |
We currently have two sets of configuration items for PG: One is common to all DBs, another is specific to the engine. Unite them, by moving the engine-specific set to the common one. Cases where this is relevant: 1. Right now, 'engine-backup --mode=restore --provision-all-databases' does not configure the engine-specific items. This causes a next run of 'engine-setup' (which is mandatory) to prompt, asking whether to configure them. 2. Trying to pre-create DBs using provisiondb, like I am now trying [1], leaves postgresql.conf without the engine-specific set. This is worse than in (1.), as the code prompting/asking works only with non-new databases. It's not a real issue for [1] specifically, because I can easily do this there, but decided it's better to patch engine-setup directly. This has no impact on the common case, where all DBs are local, as this configuration is PG-cluster-wide, not DB-specific. The main impact is on DWH running on a separate machine, with the database local to that machine. If the DWH DB is remote, our documentation already instructs to configure all items, not separating to two sets. [1] oVirt/ovirt-system-tests#241 Change-Id: Ief318aa5551cec2d778cf791926518effe404da2 Signed-off-by: Yedidyah Bar David <[email protected]>
Change-Id: I005a79fa4df940e6475bae4559bff61e1086c11a Signed-off-by: Yedidyah Bar David <[email protected]>
Change-Id: Ice14b93cd93f19acdf913fb84c98f83e461d9320 Signed-off-by: Yedidyah Bar David <[email protected]>
If the current version does not have '-SNAPSHOT', do not create a release commit, just a post-release one. Change-Id: I64dbdd09f837de197b7ee364244e6e86d7c4d6ef Signed-off-by: Yedidyah Bar David <[email protected]>
Change-Id: Iaaf47b701408da9585cacdfc0e8cc59f21845ffe Signed-off-by: Yedidyah Bar David <[email protected]>
When you build ovirt-engine on a system where a language other than English is setted, Maven Surefire uses it a default language and select MessageBundles according to this, making some tests fail. This commit set English and US explicitly as default language and country in the POM, allowing Surefire to pick the right properties files during tests. Bug-URL: https://bugzilla.redhat.com/show_bug.cgi?id=2120066
Add ovirt-engine-4.5* string to branches specification to remove the need to allow running CI on every new 4.5 branch in future Signed-off-by: Martin Perina <[email protected]>
Change-Id: Ib8e8deeca6ddbda02e3ba7dafef5b872d70d18b1 Signed-off-by: Yedidyah Bar David <[email protected]>
When we create the memory or metadata disks they are a result of other command. These disks are not act as virtual disks. In some cases the user can have permission to execute these command, but he may not have the disk profile permission to add those disks. This patch will skip setting the profile in those cases, allowing a user to create attach disks for disks that are not a real virtual disks. Change-Id: Ib04048d652c8d98df84ffccc4babd71730b218dc Bug-Url: https://bugzilla.redhat.com/1565183 Signed-off-by: Liran Rotenberg <[email protected]>
When running engine-setup with an answer file generated by a previous interactive engine-setup, without this patch, we log some secrets unfiltered. Fix this: - Require a new otopi that provides LOG_FILTER_QUESTIONS. - Conflict with older dwh/keycloak that are incompatible with the changes in getCredentials (see below). - Add an attribute 'asked_on' for constants. If a constant is_secret, require setting asked_on, to a list of question names that might change/set it. - Add a field CREDS_Q_NAME_FUNC to the various *DB_ENV_KEYS. This field should point at a function that should return the question name for a particular field. - Change getCredentials to not get a parameter queryprefix for constructing the question names, instead relying on CREDS_Q_NAME_FUNC. - Add functions *question_name for both passing as CREDS_Q_NAME_FUNC and for asked_on. - And finally: Patch filter_secrets.py to also loop over all the constants that set is_secret, and add their asked_on to env[LOG_FILTER_QUESTIONS]. This makes otopi filter out all the answers provided for these questions in answer files. Change-Id: Ibaca2a03f2020750f96ae30a3448ea2ad17fe43c Signed-off-by: Yedidyah Bar David <[email protected]>
In some scenarios CreateSnapshotFromTemplateCommand fails after creation of the record in the images table. In this case the record should be removed together with other DB records related to the disk. Change-Id: I201b49b9acd052c733c4d418abbac3b16c9431d5 Bug-Url: https://bugzilla.redhat.com/2110351 Signed-off-by: Shmuel Melamud <[email protected]>
org.eclipse.m2e:lifecycle-mapping plugin is used only when the project is edited in Eclipse IDE, it's not relevant for other maven invocation. Unfortunately it emits bellow warning during maven execution: [WARNING] The POM for org.eclipse.m2e:lifecycle-mapping:jar:1.0.0 is missing, no dependency information available [WARNING] Failed to retrieve plugin descriptor for org.eclipse.m2e:lifecycle-mapping:1.0.0: Plugin org.eclipse.m2e:lifecycle-mapping:1.0.0 or one of its dependencies could not be resolved: Failed to read artifact descriptor for org.eclipse.m2e:lifecycle-mapping:jar:1.0.0 So this patches moves the plugin invocation into its own profile, which is automatically invoked only when a project is loaded into Eclipse IDE. Signed-off-by: Martin Perina <[email protected]>
Try to remove snapshot even in case of low Storage Domain disk space. Otherwise, when remove snapshot is performed at the end of Live Storage Migration operation, the "RemoveSnapshotCommand" fails on validation and leaves 3 chunks' snapshot (7.5 GiB) unreleased. Additionally, the LSM operation is reported as failed, though the "move" part actually succeeded. Due to extra 3 chunks that are temporary used during LSM flow, we might temporary fall below disk space threshold. Currently not only that the code leaves unreleased an almost 8 GiB "junk", but it also leaves the Storage Domain in unhealthy low-space state which blocks other operations and requires a manual intervention instead of cleanly recovering from this temporary low disk space situation by performing the proper cleanup. Before the fix: 2022-08-23 13:14:56,448+03 ERROR [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (EE-ManagedScheduledExecutorService-engineScheduledThreadPool-Thread-45) [] EVENT_ID: IRS_DISK_SPACE_LOW_ERROR(201), Critical, Low disk space. iSCSI_SD2 domain has 4 GB of free space. 2022-08-23 13:16:10,455+03 WARN [org.ovirt.engine.core.bll.snapshots.RemoveSnapshotCommand] (EE-ManagedScheduledExecutorService-engineScheduledThreadPool-Thread-16) [03ef74cf-ddbe-4027-805f-02631bf96929] Validation of action 'RemoveSnapshot' failed for user admin@internal-authz. Reasons: VAR__TYPE__SNAPSHOT,VAR__ACTION__REMOVE,ACTION_TYPE_FAILED_DISK_SPACE_LOW_ON_STORAGE_DOMAIN,$storageName iSCSI_SD2 2022-08-23 13:16:15,555+03 ERROR [org.ovirt.engine.core.bll.storage.lsm.LiveMigrateDiskCommand] (EE-ManagedScheduledExecutorService-engineScheduledThreadPool-Thread-17) [03ef74cf-ddbe-4027-805f-02631bf96929] Ending command 'org.ovirt.engine.core.bll.storage.lsm.LiveMigrateDiskCommand' with failure. 2022-08-23 13:16:15,581+03 WARN [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (EE-ManagedScheduledExecutorService-engineScheduledThreadPool-Thread-17) [03ef74cf-ddbe-4027-805f-02631bf96929] EVENT_ID: USER_MOVE_IMAGE_GROUP_FAILED_TO_DELETE_SRC_IMAGE(2,025), Possible failure while deleting iSCSI_VM1_Disk1 from the source Storage Domain iSCSI_SD2 during the move operation. The Storage Domain may be manually cleaned-up from possible leftovers (User:admin@internal-authz). At the end "iSCSI_SD2" has 4 GiB available, which is below the 5 GiB "Critical Space Action Blocker". After the fix "RemoveSnapshotCommand" executes successfully: 2022-08-24 11:00:43,164+03 ERROR [org.ovirt.engine.core.dal.dbbroker.auditloghandling.AuditLogDirector] (EE-ManagedScheduledExecutorService-engineScheduledThreadPool-Thread-48) [] EVENT_ID: IRS_DISK_SPACE_LOW_ERROR(201), Critical, Low disk space. iSCSI_SD2 domain has 4 GB of free space. We fall below threshold and then automatically recover, 7.5 GiB is returned to the Storage Domain. At the end "iSCSI_SD2" correctly has 12 GiB available, which is above the 5 GiB "Critical Space Action Blocker". Signed-off-by: Pavel Bar <[email protected]>
If the command monitoring a VM backup is no longer present in the command_entities table, it will be impossible to finalize the backup and it will be stuck in the finalizing phase. This patch checks periodically for backups that are active but are no longer monitored and fails them, notifying the user a cleanup may be required. Bug-Url: https://bugzilla.redhat.com/2107590 Signed-off-by: Benny Zlotnik <[email protected]>
…gMessages.properties
Bug-Url: https://bugzilla.redhat.com/2120040 Signed-off-by: Arik Hadas <[email protected]>
When a user will try to upload ISO file the Incremental Backup will be set to false and uncheckable Bug-Url: https://bugzilla.redhat.com/2120228 Signed-off-by: Artiom Divak <[email protected]>
Changed the error message shown when using resize policy together with dedicated CPUs on the host. Change-Id: Ie6beeae598fb283fcf8c222d3ee28241986cdf1f Bug-Url: https://bugzilla.redhat.com/2078189 Signed-off-by: Shmuel Melamud <[email protected]>
When we create a new template, it is not possible to change almost any value defined in the original VM. Therefore, it does not make sense to validate on the UI, because the user cannot do anything if the validation fails. Moreover, if the validation fails, there is no way to display the results to the user - the dialog just stays open with no message displayed. When we remove the validation on the UI, we will depend on the backend validations and if something fails, the backend error messages would be displayed to the user. Note that the exception from this is the check for the duplicate name, that is performed separately and will be preserved as frontend check. In this case, the name field is marked as not valid with proper error message.
Some actions need to be executed in closeup stage before remote engine is cleaned up, so the clean up action need to have unique name. Bug-Url: https://bugzilla.redhat.com/2122174 Signed-off-by: Martin Perina <[email protected]>
Path to a virtio ISO image used in VM import was constructed assuming that the image is located in an ISO domain. To support ISO images located in a data domain we need to pass the storage domain ID to ConvertVmCommand and use prepareImage()/teardownImage() to obtain the path to the image on the host where VM conversion is running. Due to limitation of prepareImage()/teardownImage() on VDSM side, these calls cannot be nested. That's why the ISO image located in a block data domain can be used exclusively by a single instance of ConvertVmCommand. Change-Id: I835602a9019ecbcf4c63529a9d006bc43ee793be Bug-Url: https://bugzilla.redhat.com/1721455 Signed-off-by: Shmuel Melamud <[email protected]>
If a VM contains a single disk not marked as bootable, RunVmCommand allows to boot OS from this disk. But RunOnce ignores this disk when setting the boot order. This patch fixes this inconsistence. Change-Id: Ia5652bb1549bdd5205284be355e3a2903fb6e9f1 Signed-off-by: Shmuel Melamud <[email protected]>
As a part of #616 lifecycle-mapping has been move into eclipse specific profile in root pom.xml, but unfortunately other places were missed. Signed-off-by: Martin Perina <[email protected]>
1. Extract dependencies and plugins versions into properties section 2. Split oVirt/external dependencies/plugins versions into separate sections Signed-off-by: Martin Perina <[email protected]>
1. Moves cors-filter dependency definition into root pom.xml 2. Remove specific versions from dependencies, because they are defined in root pom.xml Signed-off-by: Martin Perina <[email protected]>
1. Moves ovirt-engine-extensions-api dependency definition into root pom.xml 2. Removes specific versions from dependencies, because they are defined in root pom.xml Signed-off-by: Martin Perina <[email protected]>
1. Moves jboss-as-domain-management dependency definition into root pom.xml 2. Removes specific versions from dependencies, because they are defined in root pom.xml Signed-off-by: Martin Perina <[email protected]>
1. Moves jboss-modules dependency definition into root pom.xml 2. Removes specific versions from dependencies, because they are defined in root pom.xml Signed-off-by: Martin Perina <[email protected]>
1. Moves dbunit dependency definition into root pom.xml 2. Removes specific versions from dependencies, because they are defined in root pom.xml Signed-off-by: Martin Perina <[email protected]>
… backup Fixes an issue that originates from ed023e5 The following error occured when trying to create an incremental backup: ------ ERROR [org.ovirt.engine.core.bll.storage.backup.HybridBackupCommand] (default task-1) [full_cold_vm_backup] Command 'org.ovirt.engine.core.bll.storage.backup.HybridBackupCommand' failed: CallableStatementCallback; SQL [{call insertvmbackup(?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)}]; ERROR: duplicate key value violates unique constraint "vm_backups_pkey" Detail: Key (backup_id)=(1733241d-4a0b-4205-9e52-5383eb7a82f2) already exists. ----- This was caused by using vmBackupDao.save() instead of vmBackupDao.update(), because the backup already existed when the setVmBackupSnapshot method is executed. Signed-off-by: Brooklyn Dewolf <[email protected]>
This patch adds support for DC/cluster compatibility level 4.8. Hosts need to be running on libvirt >= 9.5 and qemu-kvm 8.1. Signed-off-by: Brooklyn Dewolf <[email protected]>
The discard-no-unref was introduced to qemu to counteract fragmentation. Without this, qcow2 images would grow over 100% of their size. See https://patchew.org/QEMU/[email protected]/ I have added the option to ovirt-engine so that this flag is enabled by default on hosts with cluster level compatibility level 4.8. The EnableQemuDiscardNoUnref option can be enabled/disabled in the config. The Virtual Disk option 'Enable Discard' also needs to be active before this flag is enabled. Signed-off-by: Brooklyn Dewolf <[email protected]>
Hosts supporting level 4.8 need to be running on RHEL 9.4 or CentOS Stream/RHEL 9. Signed-off-by: Brooklyn Dewolf <[email protected]>
…sCommand There is some situations when storageDomainId is not come with parameters (image upload cancel and pause operations). For this operations checked CREATE_DISK permission on SYSTEM_OBJECT (i.e. system-wide). Problem starts when we give permissions for user only on concrete storage domain object (not system-wide). Then permission check failed for operations without storage domain id info in parameters. Here I just add check permission for disk before other objects. Signed-off-by: Stanislav Melnichuk <[email protected]>
Signed-off-by: ShubhaOracle <[email protected]>
… and included a guide DEVELOPMENT.adoc to setup dev environment easily All the instructions on how to setup the Dev Container environment can be found in DEVELOPMENT.adoc Co-authored-by: Jean-Louis Dupond <[email protected]> Signed-off-by: Brooklyn Dewolf <[email protected]>
Adressing errors reported by https://github.com/oVirt/ovirt-engine/blob/master/.github/CODEOWNERS Signed-off-by: Sandro Bonazzola <[email protected]>
Signed-off-by: Brooklyn Dewolf <[email protected]>
Signed-off-by: Andreas Paul <[email protected]>
Signed-off-by: Stepan Ermakov <[email protected]> This PR fixes the following issue: incorrect Data Center is displayed when uploading a disk image from "Storage Domain Disks" UI. Steps to reproduce: 1. Create a Storage Domain at non-Default Data Center. 2. Open the Storage -> Domains -> <The created Storage Domain> -> Disks page. 3. Select the Upload -> Start menu item. The "Upload image" dialog is displayed. Expected behavior: The Data Center filed is disabled and displays the name of the non-Default Data Center. Current behavior: The Data Center field is disabled and display the name of the Default Data Center. Fixed in the following way: when "Upload image" dialog is displayed then query the Storage Domain details with information about its Data Center and select the Storage Domain's Data Center in the "Data Center" field
…set with UEFI SecureBoot". Issue: When VM emulator is changed to Q35 Chipset with UEFI SecureBoot from Q35 Chipset with UEFI, the NVRAM file needs to be rewritten. Fix: On VM update remove the previous NVRAM file. Signed-off-by: Saksham Srivastava <[email protected]>
This commit removes the obsolete version tag in the docker-compose file and upgrades the postgres database container to 13.16 to match the client used in ovirt-engine Signed-off-by: geoff o'callaghan <[email protected]>
Avoids having to upgrade on centos stream updates. Only needs to change if/when ovirt-engine requires postgres:14 or above. Signed-off-by: geoff o'callaghan <[email protected]>
engine: Add button to remove all direct user permissions This button useful for large instances, when we have many roles for each user. Some times we want to totally clean roles for user, and this can be hard to look at all roles and search each direct one. Signed-off-by: Stanislav Melnichuk <[email protected]>
…iders CVE-2024-7259 Issue: Password was visible for external providers after changing input type from password to text in browser developer tools (Inspect tools). Fix: Added the logic for sending the ******* as password text to UI and updating the password only if user makes any change in password. Signed-off-by: Saksham Srivastava <[email protected]>
…later move such disk to a block storage (FC/iSCSI) then it will be converted to Cow/Thin Provision. And this might corrupt the disk. To avoid this issue we suggested to always upload ISO images as Raw/Preallocated. In this case no conversion needed while moving to a block storage. And disk works fine in both types (block and non-block) storages. Signed-off-by: Stepan Ermakov <[email protected]>
When using an answer file that contains: QUESTION/1/OVESETUP_ENGINE_ENABLE=str:yes The CoreEnv.ENABLE was not set to True if fqdn's didn't match and it was not a new database. But as we explicitly want to enable the engine according to the answer file, we must enable it. Signed-off-by: Jean-Louis Dupond <[email protected]>
Altered documents that threw errors for the newly defined checkstyle configuration Signed-off-by: Jasper Berton <[email protected]>
Changed to latest version 10.20.0 due to support available for: - WhitespaceAfter for LITERAL_RETURN, LITERAL_TRY, LITERAL_SYNCHRONIZED, LITERAL_CATCH and LITERAL_FINALLY Signed-off-by: Jasper Berton <[email protected]>
Changed rules and suppressions used by checkstyle to enforce more strict whitespace rules. Signed-off-by: Jasper Berton <[email protected]>
When doing a LiveStorageMigration, the code checked if the disk/image would fit not only on the destination, but also on the source. This because theoretically the snapshot image can grow until the disksize. But this means that if we needed to live migrate a 1TB disk, we need at least 1TB of free space on the source domain also. This is quite useless, and causes problems to move disks from (for example) an already overloaded source domain. As this check is neither being done when creating a simple snapshot, just drop the check on the source domain for LSM. Signed-off-by: Jean-Louis Dupond <[email protected]>
…le disks When moving multiple disks, the dropdown with the storage domain for all disks only contained the name of the storage domain. Extended this so it also shows the free space of the target storage domain. (like it's the case with the individual dropdowns). Signed-off-by: Jasper Berton <[email protected]>
When a standalone disk is made, don't try to release locks on VM, because there are none Signed-off-by: Jasper Berton <[email protected]>
…Machine(s) pop-up (#925) Check for null values to avoid UI exceptions. Signed-off-by: ShubhaOracle <[email protected]>
Altered code base so no new conflicts arise for newer checkstyle rules Signed-off-by: Jasper Berton <[email protected]>
More strict rules around indentation and parenpad Signed-off-by: Jasper Berton <[email protected]>
Added ability to sort on following collumns on Storage tab: - Allocated space - Confirmed free space - Total space Signed-off-by: Jasper Berton <[email protected]>
Update jsass to bring some fixes and possibility for build under arm macos. Signed-off-by: Stanislav Melnichuk <[email protected]>
: #900) Since the SecretValue structure is used in GWT, it should implement Serializable interface. Also add a couple of null checks to avoid NPE in the scenario explained in #900. Signed-off-by: Stepan Ermakov <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.