talos/pkg/cluster/check/default.go
Andrey Smirnov bc9e0c0dba fix: re-implement upgrade (install) with preserve
For 0.6 -> 0.7 upgrade, in any case config.yaml is preserved and moved
from `/boot` to `/system/state`.

For single node upgrade, `EPHEMERAL` partition is not touched and other
partitions are re-created as needed.

Bump provision tests to 0.6/0.7 upgrades as we get closer to the new
release.

Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
2020-10-28 07:25:26 -07:00

97 lines
4.1 KiB
Go

// This Source Code Form is subject to the terms of the Mozilla Public
// License, v. 2.0. If a copy of the MPL was not distributed with this
// file, You can obtain one at http://mozilla.org/MPL/2.0/.
package check
import (
"context"
"errors"
"time"
"github.com/talos-systems/talos/pkg/conditions"
"github.com/talos-systems/talos/pkg/machinery/config/types/v1alpha1/machine"
)
// DefaultClusterChecks returns a set of default Talos cluster readiness checks.
func DefaultClusterChecks() []ClusterCheck {
return []ClusterCheck{
// wait for etcd to be healthy on all control plane nodes
func(cluster ClusterInfo) conditions.Condition {
return conditions.PollingCondition("etcd to be healthy", func(ctx context.Context) error {
return ServiceHealthAssertion(ctx, cluster, "etcd", WithNodeTypes(machine.TypeInit, machine.TypeControlPlane))
}, 5*time.Minute, 5*time.Second)
},
// wait for bootkube to finish on init node
func(cluster ClusterInfo) conditions.Condition {
return conditions.PollingCondition("bootkube to finish", func(ctx context.Context) error {
err := ServiceStateAssertion(ctx, cluster, "bootkube", "Finished", "Skipped")
if err != nil {
if errors.Is(err, ErrServiceNotFound) {
return nil
}
return err
}
return nil
}, 5*time.Minute, 5*time.Second)
},
// wait for apid to be ready on all the nodes
func(cluster ClusterInfo) conditions.Condition {
return conditions.PollingCondition("apid to be ready", func(ctx context.Context) error {
return ApidReadyAssertion(ctx, cluster)
}, 2*time.Minute, 5*time.Second)
},
// wait for kubelet to be healthy on all
func(cluster ClusterInfo) conditions.Condition {
return conditions.PollingCondition("kubelet to be healthy", func(ctx context.Context) error {
return ServiceHealthAssertion(ctx, cluster, "kubelet", WithNodeTypes(machine.TypeInit, machine.TypeControlPlane))
}, 5*time.Minute, 5*time.Second)
},
// wait for all the nodes to report in at k8s level
func(cluster ClusterInfo) conditions.Condition {
return conditions.PollingCondition("all k8s nodes to report", func(ctx context.Context) error {
return K8sAllNodesReportedAssertion(ctx, cluster)
}, 5*time.Minute, 30*time.Second) // give more time per each attempt, as this check is going to build and cache kubeconfig
},
// wait for all the nodes to report ready at k8s level
func(cluster ClusterInfo) conditions.Condition {
return conditions.PollingCondition("all k8s nodes to report ready", func(ctx context.Context) error {
return K8sAllNodesReadyAssertion(ctx, cluster)
}, 10*time.Minute, 5*time.Second)
},
// wait for HA k8s control plane
func(cluster ClusterInfo) conditions.Condition {
return conditions.PollingCondition("all control plane components to be ready", func(ctx context.Context) error {
return K8sFullControlPlaneAssertion(ctx, cluster)
}, 5*time.Minute, 5*time.Second)
},
// wait for kube-proxy to report ready
func(cluster ClusterInfo) conditions.Condition {
return conditions.PollingCondition("kube-proxy to report ready", func(ctx context.Context) error {
return K8sPodReadyAssertion(ctx, cluster, "kube-system", "k8s-app=kube-proxy")
}, 3*time.Minute, 5*time.Second)
},
// wait for coredns to report ready
func(cluster ClusterInfo) conditions.Condition {
return conditions.PollingCondition("coredns to report ready", func(ctx context.Context) error {
return K8sPodReadyAssertion(ctx, cluster, "kube-system", "k8s-app=kube-dns")
}, 3*time.Minute, 5*time.Second)
},
// wait for all the nodes to be schedulable
func(cluster ClusterInfo) conditions.Condition {
return conditions.PollingCondition("all k8s nodes to report schedulable", func(ctx context.Context) error {
return K8sAllNodesSchedulableAssertion(ctx, cluster)
}, 2*time.Minute, 5*time.Second)
},
}
}
// ExtraClusterChecks returns a set of additional Talos cluster readiness checks which work only for newer versions of Talos.
//
// ExtraClusterChecks can't be used reliably in upgrade tests, as older versions might not pass the checks.
func ExtraClusterChecks() []ClusterCheck {
return []ClusterCheck{}
}