Known Issues¶

A complete list of known issues for OpenNebula is maintained here.

Also relevant information about bugs affecting the Community Edition are those resolved issues in maintenance versions of the Enterprise Edition.

This page will be updated with relevant information about bugs affecting OpenNebula, as well as possible workarounds until a patch is officially published.

Security¶

By default the RAW/DATA attribute is not restricted for regular users, which may lead to private data breach from the server. To fix this please modify the value VM_RESTRICTED_ATTR = "RAW" to VM_RESTRICTED_ATTR = "RAW/DATA" in oned.conf

Drivers - Network¶

Edge Cluster Public IP: NIC_ALIAS on the public network can only be associated to a NIC on the same network.

Drivers - Storage¶

LXC, XFS formatted disk images are incompatible with the fs_lvm driver. The image fails to be mounted on the host.
OneStor: Live migration doesn’t work for VMs with OneStor recovery snapshots enabled in 6.8.0
OneStor: Incremental backups doesn’t work for VMs with OneStor recovery snapshots enabled in 6.6+. OneStor recovery snapshot feature is going to be decomissioned in 7.0.

Sunstone¶

Guacamole RDP as is currently shipped in OpenNebula does not support NLA authentication. You can follow these instructions in order to disable NLA in the Windows box to use Guacamole RDP within Sunstone.
Creating a VM with SPICE graphics, on Alma9, will cause the VM to stay on FAILED state.

FireEdge¶

Due to a configuration mismatch the disk snapshot buttons are currently not being rendered properly. In order to re-enable the disk snapshots the following values need to be added in the vm-tab.yaml configuration file, under the storage section:

disk-snapshot-create: true
disk-snapshot-rename: true
disk-snapshot-revert: true
disk-snapshot-delete: true

OneProvision¶

Until the 6.8 is released, OneProvision repositories definitions don’t work

Install Linux Graphical Desktop on KVM Virtual Machines¶

OpenNebula uses the cirrus graphical adapter for KVM Virtual Machines by default. It could happen that after installing a graphical desktop on a Linux VM, the Xorg window system does not load the appropriate video driver. You can force a VESA mode by configuring the kernel parameter vga=VESA_MODE in the GNU GRUB configuration file. Here you can find the VESA mode numbers. For example, adding vga=791 as kernel parameter will select the 16-bit 1024×768 resolution mode.

vCenter Image behavior¶

Images created or modified with the vcenter datastore driver may fail. This is due to a missing require statment in the script which checks file size. To resolve the issue, please insert the following code directly below the require 'vcenter_driver' line, which should be on line 56, into the file /var/lib/one/remotes/datastore/vcenter/stat:

require 'open3'

vCenter Snapshot behavior¶

VMs in vCenter 7.0 exhibit a new behavior regarding snapshots and disks attach/detach operations. When vCenter 7.0 detects any change in the number of disks attached to a VM, it automatically cleans all the VM snapshots. OpenNebula doesn’t take this into account yet, so the snapshots stated by OpenNebula, after a disk attach or disk detach, are pointing to a null vCenter reference, and as such, cannot be used. Please keep this in mind before a solution is implemented.
In both vCenter 7.0 and vCenter 8.0 when trying to create a vcenter datastore image with a URL for the source Path, a Timeout error is shown instead of the image ID. The Image is still created, but the output value from oneimage is incorrect.

Warning when Exporting an App from the Marketplace Using CLI¶

When exporting an application from the marketplace using the CLI the following warning can be seen:

/usr/lib/one/ruby/opennebula/xml_element.rb:124: warning: Passing a Node as the second parameter to Node.new is deprecated. Please pass a Document instead, or prefer an alternative constructor like Node#add_child. This will become an error in a future release of Nokogiri.

This is harmless and can be discarded, it will be addressed in future releases.

Timeout when creating image from URL source using the CLI¶

When running the CLI command oneimage create while using a URL for the source of the image, and the curl version is at least 7.17.0, the command may output a timeout message ( Net::ReadTimeout with #<TCPSocket:(closed)> ) but will still create the image. This can be resolved by using the -o / --noretry flag for oneimage create to prevent the internal curl command from retrying and triggering a timeout.

Contextualization¶

GROW_ROOTFS and GROW_FS will not extend btrfs filesystems
onesysprep does not support Debian 12 yet

Backups¶

OpenNebula stores the whole VM Template in a backup. When restoring it some attributes are wiped out as they are dynamic or they need to be re-generated (e.g. IP). However some attributes (e.g. DEV_PREFIX) would be better to keep them. It is recommended to review and adjust the resulting template for any missing (and required) attribute. The list of attributes removed can be checked here.
Incremental snapshot backups on Debian12 doesn’t work due to issue with AppArmor.

Market proxy settings¶

The option --proxy in the MARKET_MAD may not be working correctly. To solve it, execute systemctl edit opennebula and add the following entries:

[Service]
Environment="http_proxy=http://proxy_server"
Environment="https_proxy=http://proxy_server"
Environment="no_proxy=domain1,domain2"

Where proxy_server is the proxy server to be used and no_proxy is a list of the domains or IP ranges that must not be accessed via proxy by opennebula. After that, reload systemd service configuration with systemctl daemon-reload and restart opennebula with a systemctl restart opennebula