(Update as of June 3, 2025)
A temporary fix has been implemented for this error. For more details, please refer to the following:
Announcement of Resumption of mdx II System Operation and Commencement of GPU Node Services
—-
Currently, the following issues are occurring in the mdxⅡ system.
1. File system error due to volume corruption of instances
・Issue Description
In Red Hat OpenStack Platform on mdxⅡ, volume corruption may occur in some instances.
We have confirmed a problem where the OS of the instance encounters a file system error or becomes read-only either during or after the instance startup.
Below is an example output of the file system error on the OS when this issue occurs.
【Example output on Ubuntu Server 22.04】
|
#80887: comm rs:main Q:Reg: iget: checksum invalid [ 9470.149891] EXT4-fs (vda1): Remounting filesystem read-only [ 9470.151476] EXT4-fs error (device vda1): ext4_lookup:1858: inode |
・Current Status:
The cause of the issue is under investigation.
Based on the current verification results, the issue tends to occur less frequently on Rocky Linux instances.
However, we are not in a position to definitively state that the issue does not occur at all on Rocky Linux.
・Temporary Measure:
If a file system error occurs, we apologize for the inconvenience, but please contact us via the inquiry form.
Recovery will be performed using one of the following methods:
|
・Repairing the file system using fsck ・Restoring from a backup of the volume file ※ The system performs a daily backup of volume files at 1:00 AM. If recovery is performed from a backup, the data may revert to a state from as early as 1 day to as late as 2 weeks ago. |
2. Error when creating a volume from instance or volume snapshot
Due to the maintenance scheduled from May 19 to May 21, 2025, the issue with the snapshot feature is expected to be resolved, and it should be functioning normally after the maintenance is completed.
・Issue Description:
In Red Hat OpenStack Platform on mdxⅡ, we have confirmed a problem where creating a new volume from a snapshot taken from an instance or volume results in an error if it’s the second or later snapshot.
This issue occurs under the following procedures when creating a volume from a second or later snapshot of an instance or volume:
【Creating a volume from an instance snapshot】
|
(1) Log in to the user portal (2) Click [Project] menu > [Compute] panel > [Instances] tab (3) In the Actions column of the instance you want to snapshot, click [Create Snapshot] → This issue occurs when using snapshots taken from the second time onward (4) Click [Project] menu > [Compute] panel > [Images] tab (5) From the dropdown menu at the right end of the snapshot, click [Create Volume] to create a volume |
【Creating a volume from a volume snapshot】
|
(1) Log in to the user portal (2) Click [Project] menu > [Volumes] panel > [Volumes] tab (3) In the Actions column of the volume you want to snapshot, click [Create Snapshot] → This issue occurs when using snapshots taken from the second time onward (4) Click [Project] menu > [Volumes] panel > [Snapshots] tab (5) From the dropdown menu at the right end of the snapshot, click [Create Volume] to create a volume |
・Current Status:
This is a known issue with Red Hat OpenStack Platform and is currently under investigation.
・Temporary Measure:
To mitigate this issue, we have temporarily hidden the following actions and tabs on the user portal.
We apologize for the inconvenience, but until the issue is resolved, snapshot operations for instances and volumes will not be available.
|
+ The [Create Snapshot] action in step (3) of 【Creating a volume from an instance snapshot】 is hidden + The [Snapshots] tab in step (4) of 【Creating a volume from a volume snapshot】 is hidden |
We apologize for the inconvenience and appreciate your understanding.