Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Switch naming to v2.5.1 #597

Closed
wants to merge 90 commits into from
Closed

Conversation

liranr23
Copy link
Member

No description provided.

liranr23 and others added 30 commits July 31, 2023 17:19
LUNs disks are a special case. They do not have a task when migrating
and we do not need to update their progress. The disk are pre-made and
just being attached. Therefore, we need to skip them.

Signed-off-by: Liran Rotenberg <[email protected]>
Signed-off-by: Arik Hadas <[email protected]>
Signed-off-by: Bella Khizgiyaev <[email protected]>
In c33a65b, we dropped by mistake the libvirt-domain-xml and
vddk-vol-mount mounts from the virt-v2v container when migrating from
vSphere (and kept them only when migrating from OVA provider). This
broke warm migration from vSphere and decrease the speed of
cold-migrations significantly. Thus, adding back the missing mounts.

Signed-off-by: Arik Hadas <[email protected]>
When migrating a LUN disk from ovirt source in warm migration we need to
mark the CopyDisk as completed.

Signed-off-by: Liran Rotenberg <[email protected]>
Signed-off-by: Benny Zlotnik <[email protected]>
Signed-off-by: Benny Zlotnik <[email protected]>
Fix the shared disk validation when the attachments
array is empty.

Fixes kubev2v#442

Signed-off-by: Miguel Martín <[email protected]>
Attempt to properly support remote source by using destination and
source clients where appropriate

Signed-off-by: Benny Zlotnik <[email protected]>
Since we now support two flows in the virt-v2v image, it doesn't end
with the actual `virt-v2v` command. In that case the pod may end with
success although, it didn't. Now the script will exit with an error on
the first error occurrence.

Signed-off-by: Liran Rotenberg <[email protected]>
virt-v2v and ovirt-populator podman build is broken without it

Signed-off-by: Benny Zlotnik <[email protected]>
Since adding support for OCP to OCP volumes may not only be PVCs so a
nil check has to be made

Signed-off-by: Benny Zlotnik <[email protected]>
Signed-off-by: Benny Zlotnik <[email protected]>
fixing the error in the entrypoint script:
line 5: [: missing `]'

Signed-off-by: Liran Rotenberg <[email protected]>
This patch will bump the ubi9-minimal image and include the fix of the
SSL connection to the cold virt-v2v.

Signed-off-by: Liran Rotenberg <[email protected]>
When looking up VMs in plans we need to use the namespace from the Ref,
and not rely on splitting the name

Signed-off-by: Benny Zlotnik <[email protected]>
The format has to always be raw for OpenShift Virtualization

Signed-off-by: Benny Zlotnik <[email protected]>
bkhizgiy and others added 21 commits August 24, 2023 14:41
Skip connection test when updating secret of OVA provider (the URL
doesn't change and credentials are not specified) and reduce congnitive
complexity of the validateProviderSecret function.

Signed-off-by: Arik Hadas <[email protected]>
Both methods work the same way but the latter better aligns with how we
create other images, like must-gather.

Signed-off-by: Arik Hadas <[email protected]>
If a volume of a VM changes we need to update the VM's revision to make
sure it is revalidated

Signed-off-by: Benny Zlotnik <[email protected]>
To force revalidation if the volume changed

Signed-off-by: Benny Zlotnik <[email protected]>
The format of the image is stored as image.diskFormat rather than
image.disk_format - change the validation and its tests accordingly.

Signed-off-by: Arik Hadas <[email protected]>
Signed-off-by: Benny Zlotnik <[email protected]>
We cannot support this currently

Signed-off-by: Benny Zlotnik <[email protected]>
In order to run on restricted namespaces, the v2v/populator pods need to
be set with securityContext.seccompProfile.type that is either
RuntimeDefault or Localhost. Thus, we set it in the security context
constraints of the service account forklift-controller.

Signed-off-by: Liran Rotenberg <[email protected]>
Introduce a new service account for the forklift-populator-controller
pod. Currently, it uses the same RBAC as forklift-controller service
account. Later on, we will restrict the RBAC to what the
populator-controller needs.

Signed-off-by: Liran Rotenberg <[email protected]>
This change will make the target namespaces of the migrations to be
restricted.

Signed-off-by: Liran Rotenberg <[email protected]>
... rather than when the hooks are called to avoid "concurrent map read
and map write" error.

Signed-off-by: Arik Hadas <[email protected]>
Until now the target namespace for the migration in the tests were
generated by the plan without labeling them. This change will create the
target namespace as restricted.

Signed-off-by: Liran Rotenberg <[email protected]>
These changes disable EMS checks when FIPS is enabled by changing
configuration files in /etc.

The link /etc/pki/tls/fips_local.cnf that points to a configuration file
within the FIPS crypto policy is replaced with a regular file, and
Options=RHNoEnforceEMSinFIPS is added to the crypto_policy section in
/etc/pki/tls/openssl.cnf

Signed-off-by: Arik Hadas <[email protected]>
The namespace may be missing from the network name, in this case we need
to use the VM's namespace

Signed-off-by: Benny Zlotnik <[email protected]>
Signed-off-by: Liran Rotenberg <[email protected]>
@sonarcloud
Copy link

sonarcloud bot commented Sep 20, 2023

Kudos, SonarCloud Quality Gate passed!    Quality Gate passed

Bug A 0 Bugs
Vulnerability A 0 Vulnerabilities
Security Hotspot A 0 Security Hotspots
Code Smell A 0 Code Smells

No Coverage information No Coverage information
No Duplication information No Duplication information

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

6 participants