Redis Enterprise for Kubernetes 发行说明 6.2.8-11（2022 年 1 月）

支持 Istio 作为入口控制器、K8s 1.22（AKS、kOps、GKE）、OpenShift 4.9

适用于 Kubernetes 的 Redis Enterprise

概述

Redis Enterprise K8s 6.2.8-11 版本为 Redis Enterprise Software 版本 6.2.8 提供支持，并包含多项增强功能和错误修复。

下面介绍了主要的新功能、错误修复和已知限制。

图像

此版本包括以下容器映像：

  • Redis 企业版：redislabs/redis:6.2.8-53或redislabs/redis:6.2.8-53.rhel7-openshift
  • 作员：redislabs/operator:6.2.8-11
  • 服务索具：redislabs/k8s-controller:6.2.8-11或redislabs/services-manager:6.2.8-11（在 Red Hat 注册表中）

功能改进

  • 支持将 Istio 网关/虚拟服务作为入口控制器。请注意，对于主动-主动数据库，作员不会创建入站规则，应手动配置这些规则。测试的 Istio 版本是 Istio 1.12.0 （RED-64020）
  • 支持 K8s 1.22（AKS、kOps、GKE）和 OpenShift 4.9 （RED-64016）
  • 支持 REC 中的 Pod 终止宽限期自定义（高级使用案例） （RED-67217）
  • 改进了文档中 SCC 配置步骤的安全粒度 （RED-67321）
  • 更改了在集群上创建两个同名数据库时的行为。作员避免为他们创建服务以防止可能的损坏 （RED-64535）
  • 改进了有关在使用 Hashicorp Vault 时更改集群凭证的文档 （RED-65304）

修复的 bug

  • 升级后的 Go 依赖项标记为易受攻击（RED-63858、RED-68651）
  • 避免了在 K8s 1.21 上使用弃用通知淹没作员日志 （RED-67544）
  • 修复了在 Microsoft Windows 上运行的日志收集器实用程序问题 （RED-67682）
  • 修复了对 TLS 证书使用 Windows 行尾时对 RS 集群的过度更新 （RED-67874）

已知限制

大型集群

在具有 9 个以上 REC 节点的集群上，在某些情况下，Kubernetes 升级可能会导致 Redis 集群无响应。6.4.2-5 版本中提供了修复。在升级 Kubernetes 集群之前，请将 Operator 版本升级到 6.4.2-5 或更高版本。（红色-93025）

使用 Bundle 进行升级kubectl apply -f失败 （RED-69515）

使用 Bundle 进行升级kubectl apply -f失败，并给出以下错误：

The CustomResourceDefinition "redisenterpriseclusters.app.redislabs.com" is invalid: spec.preserveUnknownFields: Invalid value: true: must be false in order to use defaults in the schema

解决办法：在升级之前，将 REC 自定义资源的“spec.preserveUnknownFields”值设置为 false。您可以使用以下命令：

kubectl patch crd redisenterpriseclusters.app.redislabs.com -p $'spec:\n preserveUnknownFields: false'

Long cluster names cause routes to be rejected (RED-25871)

A cluster name longer than 20 characters will result in a rejected route configuration because the host part of the domain name will exceed 63 characters. The workaround is to limit cluster name to 20 characters or less.

Cluster CR (REC) errors are not reported after invalid updates (RED-25542)

A cluster CR specification error is not reported if two or more invalid CR resources are updated in sequence.

An unreachable cluster has status running (RED-32805)

When a cluster is in an unreachable state, the state is still running instead of being reported as an error.

Readiness probe incorrect on failures (RED-39300)

STS Readiness probe does not mark a node as "not ready" when running rladmin status on node failure.

Role missing on replica sets (RED-39002)

The redis-enterprise-operator role is missing permission on replica sets.

Private registries are not supported on OpenShift 3.11 (RED-38579)

OpenShift 3.11 does not support DockerHub private registries. This is a known OpenShift issue.

Internal DNS and Kubernetes DNS may have conflicts (RED-37462)

DNS conflicts are possible between the cluster mdns_server and the K8s DNS. This only impacts DNS resolution from within cluster nodes for Kubernetes DNS names.

5.4.10 negatively impacts 5.4.6 (RED-37233)

Kubernetes-based 5.4.10 deployments seem to negatively impact existing 5.4.6 deployments that share a Kubernetes cluster.

Node CPU usage is reported instead of pod CPU usage (RED-36884)

In Kubernetes, the node CPU usage we report on is the usage of the Kubernetes worker node hosting the REC pod.

Clusters must be named "rec" in OLM-based deployments (RED-39825)

In OLM-deployed operators, the deployment of the cluster will fail if the name is not "rec". When the operator is deployed via the OLM, the security context constraints (scc) are bound to a specific service account name (i.e., "rec"). The workaround is to name the cluster "rec".

REC clusters fail to start on Kubernetes clusters with unsynchronized clocks (RED-47254)

When REC clusters are deployed on Kubernetes clusters with unsynchronized clocks, the REC cluster does not start correctly. The fix is to use NTP to synchronize the underlying K8s nodes.

Deleting an OpenShift project with an REC deployed may hang (RED-47192)

When a REC cluster is deployed in a project (namespace) and has REDB resources, the REDB resources must be deleted first before the REC can be deleted. As such, until the REDB resources are deleted, the project deletion will hang. The fix is to delete the REDB resources first and the REC second. Afterwards, you may delete the project.

REC extraLabels are not applied to PVCs on K8s versions 1.15 or older (RED-51921)

In K8s 1.15 or older, the PVC labels come from the match selectors and not the PVC templates. As such, these versions cannot support PVC labels. If this feature is required, the only fix is to upgrade the K8s cluster to a newer version.

Hashicorp Vault integration - no support for Gesher (RED-55080)

There is no workaround at this time

REC might report error states on initial startup (RED-61707)

There is no workaround at this time except to ignore the errors.

PVC size issues when using decimal value in spec (RED-62132)

The workaround for this issue is to make sure you use integer values for the PVC size.

Following old revision of quick start guide causes issues creating an REDB due to unrecognized memory field name

The workaround is to use the newer (current) revision of the quick start document available online.

Compatibility Notes

See Supported Kubernetes distributions for the full list of supported distributions.

Now supported

This release adds support for the following:

  • K8s 1.22 for GKE, AKS, and kOps
  • OpenShift 4.9 (uses K8s 1.22)

Deprecated

  • Rancher 2.5/K8s 1.17 support is deprecated

No longer supported

This release removes support for the following:

  • Rancher 2.4 (previously deprecated)
RATE THIS PAGE
Back to top ↑