Azure Stack Hub hotfix 1.2108.2.130
- Improved the exception handling in Cluster Manager to avoid falsely reporting memory exhaustion alerts.
- Added labels for on-demand log collection when "infrastructure service unavailable" alert is generated.
- Improvements to support tools.
- Fixed a bug in storage table re-encryption that was causing secret rotation to fail.
- Fixed an issue in which the BlobSasManager service might crash during VM deletion.
- Support for the
.defaultscope has been added to AD FS. Now the client libraries supporting the OAUTH 2.0 flow can be used against AD FS environments.
Fixes rolled up from previous hotfix releases
- Improved stability of SDN components.
- Improved the PnP device attached alert and moved it back to preview.
- Introduced health probe for SecretService to improve service resilience and availability.
- Fixed an issue in which scaling a VMSS in and out would eventually fill a subnet's IP address space.
- Optimized reading of disk IOPS values to support VMs with a large number of data disks.
- CRP now self-heals a VM with a SCSI disk that failed to attach, instead of requiring operator removal of the disk from the VM.
- Removed IIS default website to prevent server IP address leak vulnerability.
- Fixed an issue in the Virtual Machine Scale Set portal creation experience that caused the addition of an existing load balancer to fail.
- Removed unsupported Reapply feature in the virtual machine portal experience.
- Fixed an issue in which the SRP container portal cannot display more than 1 page of containers.
- Fixed an issue with searching for a container by prefix in the SRP container portal.
- Improved blob metadata backup stability by skipping unnecessary dependency.
- Added support for Azure Stack Hub root certificate rotation.
- Fixed an issue that prevented guest operating system activation of Windows Server 2022.
- Fixed a null reference issue when calling the Compute Resource Provider API to power off a virtual machine without doing a shutdown.
- Fixed an issue in which some StorageController requests might time out under high concurrency.
- Removed some of the network performance counters and reduced collection interval for other perf counters.
- Cleaned up unneeded networking traces from Baremetal, NC and XRP VMs.
- Fixed an issue deleting Microsoft.ContainerService/managedCluster resources that occurred when resources managed by the AKS resource provider were manually deleted beforehand.
- Fixed a regression in which VM status is reported as UNKNOWN in the portal.
- Fixed an issue that could impact updating from 2102 to 2108.
- Support for new Kubernetes versions in AKS.
- Fixed bugs in trace collector.
- Fixed a bug related to physical disk health when repairing a node with SED drives.
- Improvements to support tools.
- Fixed bugs in log collection.
- Fixed code defect leading to VM deployment failures.
- Improved the resolution of the Network Resource Provider.
- Fixed a bug in incremental disk snapshots in which a failed snapshot can block any future snapshot operations.
- Fixed a disk snapshot failure and VM IO hang when taking snapshots.
- Shortened the PEP tokens and made them human-readable.
- Fix to improve SLB throughput after enabling Simultaneous Multi-Threading (SMT).
- Fixed an issue in which the table service partition was offline when its underlying storage was out of space.
- Added retry logic around Get-Volume calls in Test-AzureStack InfraCapacity validation.
To apply this hotfix, you must have version 1.2108.2.65 or later.
As outlined in the release notes for the 2108 update, make sure that you refer to the update activity checklist on running Test-AzureStack (with specified parameters), and resolve any operational issues that are found, including all warnings and failures. Also, review active alerts and resolve any that require action.
Download the following files. Then, follow the instructions on the Apply updates in Azure Stack page on the Microsoft Docs website to apply this update to Azure Stack.