Back
system           

v23

Important notes

  1. The Oxide CLI, Go SDK, and Terraform Provider have been updated for API enhancements described under New features.

  2. The support bundle API endpoints have moved out of the experimental namespace. The /experimental/v1/system/support-bundles paths are replaced by /v1/system/support-bundles. See the Troubleshooting guide for more details on generating support bundles. (omicron#11097)

  3. The discoverable field has been removed from the silo_create API endpoint. Requests from older clients that include the field are accepted and the silo will be created as discoverable. (omicron#11296)

System requirements

Please refer to v1.0.0 release notes.

Installation

Oxide Computer Model 0 must be installed and configured under the guidance of Oxide technicians. The requirement may change in future releases.

Upgrade compatibility

Upgrade from version 22 is supported. Existing instances will continue to run until the sled they run on is restarted, at which point they will be transitioned to the failed state. Instances with an auto-restart policy set to never will remain in the failed state until started manually. Any instances that are sensitive to hard shutdowns should be gracefully shut down from within the guest operating system before the update commences.

All existing setup and data (e.g., projects, users, instances) remain intact after the software update. More details can be found on the system update page.

New features

IPv6-only rack networking

The rack can now be set up with IPv6-only external networking. Control plane services with external addresses (external DNS, boundary NTP, and the API and web console) are assigned addresses from an IPv6 service IP pool. See the Network Preparations guide for an example configuration. (omicron#11208, omicron#11207, omicron#11092, omicron#11236)

Alert listing

Alerts can be listed and viewed directly with the alert_list and alert_view API endpoints without first creating an alert receiver. See Viewing alerts in the Alerts guide for more details. (omicron#11072)

Optional default resources on project and VPC creation

The project_create and vpc_create API endpoints accept an optional defaults field, which controls whether the default VPC and default subnet are created. (omicron#9750)

Sled slot in hardware inventory

The sled_list and sled_view API responses now include the sled’s slot number in the rack. (omicron#11077)

Web console

  • There are three new pages for operators under System: audit log, alerting with webhooks, and support bundles.

  • Silo quotas can now be edited from the system utilization page.

  • Number fields show an error for out-of-range values rather than attempting to set them.

  • The console is more usable on small screens, and screen readers now announce page changes and read button tooltips.

Full console changelog

Other enhancements and bug fixes

  • Old blueprints are pruned automatically (omicron#10309)

  • Instances are not placed on sleds being evacuated for update (omicron#11173, omicron#11293)

  • System update no longer shows a spurious "contact support" warning before the first blueprint exists (omicron#11157)

  • Nexus no longer polls VMMs that are terminal or belong to deleted instances (omicron#11100)

  • SAML responses are checked against the XML signature rules in section 5.4 of the SAML core specification (omicron#11322)

  • SAML identity provider metadata is fetched with the same external HTTP client configuration as other outbound requests (omicron#11320)

  • Failed alert deliveries are resent even when the same alert was delivered successfully to another receiver (omicron#11318)

  • OPTE periodically sends unsolicited IPv6 router advertisements, so service zones on an IPv6-only rack recover their default route without intervention (opte#1054, omicron#11313) (dendrite#356)

  • Release artifact size is reduced by about 25% (omicron#10887)

Patches

Known behavior and limitations

End-user features

Feature AreaKnown Issue/LimitationIssue Number

Disk/image management

Disks in importing_from_bulk_writes state cannot be deleted directly. The procedure for unsticking a canceled disk import can be used as a workaround.

omicron#2987

Disk/image management

Disk rejected by guest OS due to duplicate nvme device names. The issue is caused by a 20-character limit in applying the disk name to the device serial number. See the Troubleshooting guide for more information.

-

Disk/image management

The ability to modify image metadata is not available at this time.

omicron#2800

Instance networking

Instances can become inaccessible on a sled that loses communication with switches; a reboot is currently required to restore access.

-

Instance orchestration

Unable to start an instance that has a disk replica on a sled being updated.

crucible#1690

Instance orchestration

Instance start API frequently times out when attached to local disks.

omicron#9953

Instance orchestration

New instances cannot be created when the total number of NAT entries (private-to-external IP mappings) in the system exceeds 1024.

omicron#6939

Instance performance

The tsc clocksource is treated as unreliable by guest, resulting in its fallback to use substantially slower timestamp syscalls. A workaround for this issue can be found in the Troubleshooting guide.

omicron#8001

Instance performance

Linux guests unable to capture hardware events using perf record. A workaround for this issue can be found in the Troubleshooting guide.

propolis#894

VPC internet gateway

Changing a silo’s default IP pool causes some instances to lose their outbound internet access. This is due to a mismatch between the pool containing the instances' external IP (which are allocated from the new default pool) and the pool attached to the system-created internet gateways (which are linked to the old pool during creation time). Please see the Troubleshooting guide for some possible options for restoring instance outbound connectivity.

omicron#7297

VPC internet gateway

An internet gateway cannot have both an IPv4 and an IPv6 address attached.

omicron#10943

VPC routing

Subnet update clears custom router ID when the field is left out of the request body.

omicron#6406

VPC routing

Network interface update clears transit ips when the field is left out of the request body.

-

Telemetry

VM instance memory utilization and VPC network/firewall metrics are unavailable at this time.

-

Operator features

Feature AreaKnown Issue/LimitationIssue Number

Rack setup

On an IPv6-only rack, boundary NTP zones can lose their default route after a full rack reboot. The route is restored automatically, but time synchronization can take up to 10 minutes.

omicron#11301

Reliability

Sled memory usage improvement required to prevent system services from restarting when under extreme memory pressure.

-

Silo management

The ability to modify silo and IDP metadata is not available at this time.

omicron#3400, omicron#3125

System management

Real-time availability status for sleds and physical storage is not yet shown in the inventory API or UI.

omicron#2035

System management

Operator-driven instance migration across sleds is currently unavailable.

-

System management

Some running instances transitioned to the "stopped" state after update.

omicron#9177

Related documentation