Skip to content

For the complete documentation index, see llms.txt. Markdown versions of all docs pages are available by appending .md to any docs URL.

1.31.1

Page as Markdown

Solo build of Istio version 1.31.1 patch release.

This release note describes what’s different between Solo builds of Istio versions 1.31.0 and 1.31.1.

Upgrade Notes

Kubernetes API requests are now bounded by client-side deadlines

The defaults suit most clusters. Environments with a slow or heavily loaded API server can raise them, or set any of them to 0 to restore the previous unbounded behavior, with these environment variables on the affected component.

VariableDefaultApplies to
ISTIO_KUBE_HEADER_TIMEOUT90sHow long any request may wait for its first response headers. Must stay above the API server’s --request-timeout (60s by default).
ISTIO_KUBE_REQUEST_TIMEOUT120sTotal duration of a normal request, including reading the response. Must exceed ISTIO_KUBE_HEADER_TIMEOUT.
ISTIO_KUBE_STREAM_TIMEOUT15mTotal duration of a watch or other stream. Must exceed the API server maximum watch duration.
ISTIO_INFORMER_REQUEST_TIMEOUT15mTotal duration of an informer list. Must accommodate a full list of the largest resource in the cluster.

A canceled request increments kube_client_deadline_fired_total and logs a warning naming the method and path.

General Changes

Solo Flavor Changes

  • Fixed listeners going dark during an EnterpriseListenerSet to ListenerSet migration: while the control plane converged on a newly created same-named ListenerSet, the superseded EnterpriseListenerSet could still be arbitrated against it, marking every listener on both objects conflicted for the convergence window. A superseded EnterpriseListenerSet is now excluded from listener conflict arbitration entirely.

  • Fixed Istio components hanging indefinitely when a Kubernetes API request gets no response, for example when a load balancer or proxy silently drops the return path. Previously such a request could stay open forever, stalling the controller that issued it until the process restarted. Requests are now bounded by client-side deadlines, so a stuck request is canceled and the component recovers on its own. Each cancellation is logged as a warning and counted by the kube_client_deadline_fired_total metric.

  • Fixed the documentation link in istioctl multicluster check --help, which pointed at a path that no longer exists. The link now resolves to the multicluster installation guide at docs.solo.io/istio/latest/multicluster/install/default/manual/.

  • Fixed an issue where the network gateway selected for a workload was picked in a random order when a network had more than one gateway entry, which caused unnecessary workload (WDS) pushes on every recompute and could alternate the gateway address a workload uses.

  • Fixed the peering controller re-validating every peer cluster, and with it every peered workload and service, after each of its own Segment status writes, which kept the peering leader busy for a long time after a leader change.

  • Fixed an issue where every peered workload was recomputed each time the peering leader updated the Segment status, which caused elevated istiod CPU usage and unnecessary xDS pushes during a peering resync.

  • Fixed an issue where Envoy proxies subscribed to Workload metadata discovery (HBONE sidecars and ambient gateways) received a full endpoint (EDS) rebuild on every workload or WorkloadEntry change.

FIPS Flavor Changes

No changes in this section.