mirror of
https://github.com/kubesphere/website.git
synced 2025-12-26 08:32:55 +00:00
add upgrade tutorials
Signed-off-by: huanggze <loganhuang@yunify.com>
This commit is contained in:
huanggze
parent
4071aaf085
commit
30ccff625c
|
|
@ -11,12 +11,4 @@ icon: "/images/docs/docs.svg"
|
|||
|
||||
---
|
||||
|
||||
## Installing KubeSphere and Kubernetes on Linux
|
||||
|
||||
In this chapter, we will demonstrate how to use KubeKey to provision a new Kubernetes and KubeSphere cluster based on different infrastructures. Kubekey can help you to quickly build a production-ready cluster architecture on a set of machines from zero to one. It also helps you to easily scale the cluster and install pluggable components on existing architecture.
|
||||
|
||||
## Most Popular Pages
|
||||
|
||||
Below you will find some of the most common and helpful pages from this chapter. We highly recommend you to review them at first.
|
||||
|
||||
{{< popularPage icon="/images/docs/bitmap.jpg" title="Install KubeSphere on AWS EC2" description="Provisioning a new Kubernetes and KubeSphere cluster based on AWS" link="" >}}
|
||||
This section helps cluster operators to upgrade existing KubeSphere to v3.0.0.
|
||||
|
|
@ -1,155 +0,0 @@
|
|||
---
|
||||
title: "Upgrade KubeSphere Only"
|
||||
keywords: "kubernetes, docker, kubesphere, jenkins, istio, prometheus"
|
||||
description: "Upgrade KubeSphere without Kubernetes"
|
||||
|
||||
linkTitle: "Upgrade KubeSphere Only"
|
||||
weight: 200
|
||||
---
|
||||
|
||||
KubeSphere 2.1.0 was released on Nov 11th, 2019, which fixes known bugs, adds some new features and brings some enhancement. If you have installed versions of 2.0.x, please upgrade it and enjoy the better user experience of v2.1.0.
|
||||
|
||||
## Installer Enhancement
|
||||
|
||||
- Decouple some components and make components including DevOps, service mesh, app store, logging, alerting and notification optional and pluggable
|
||||
- Add Grafana (v5.2.4) as the optional component
|
||||
- Upgrade Kubernetes to 1.15.5. It is also compatible with 1.14.x and 1.13.x
|
||||
- Upgrade [OpenPitrix](https://openpitrix.io/) to v0.4.5
|
||||
- Upgrade the log forwarder Fluent Bit to v1.3.2
|
||||
- Upgrade Jenkins to v2.176.2
|
||||
- Upgrade Istio to 1.3.3
|
||||
- Optimize the high availability for core components
|
||||
|
||||
## App Store
|
||||
|
||||
### Features
|
||||
|
||||
Support upload / test / review / deploy / publish/ classify / upgrade / deploy and delete apps, and provide nine built-in applications
|
||||
|
||||
### Upgrade & Enhancement
|
||||
|
||||
- The application repository configuration is moved from global to each workspace
|
||||
- Support adding application repository to share applications in a workspace
|
||||
|
||||
## Storage
|
||||
|
||||
### Features
|
||||
|
||||
- Support Local Volume with dynamic provisioning
|
||||
- Provide the real-time monitoring feature for QingCloud block storage
|
||||
|
||||
### Upgrade & Enhancement
|
||||
|
||||
QingCloud CSI is adapted to CSI 1.1.0, supports upgrade, topology, create or delete a snapshot. It also supports creating PVC based on a snapshot
|
||||
|
||||
### BUG Fixes
|
||||
|
||||
Fix the StorageClass list display problem
|
||||
|
||||
## Observability
|
||||
|
||||
### Features
|
||||
|
||||
- Support for collecting the file logs on the disk. It is used for the Pod which preserves the logs as the file on the disk
|
||||
- Support integrating with external ElasticSearch 7.x
|
||||
- Ability to search logs containinh Chinese words
|
||||
- Add initContainer log display
|
||||
- Ability to export logs
|
||||
- Support for canceling the notification from alerting
|
||||
|
||||
### UPGRADE & ENHANCEMENT
|
||||
|
||||
- Improve the performance of log search
|
||||
- Refine the hints when the logging service is abnormal
|
||||
- Optimize the information when the monitoring metrics request is abnormal
|
||||
- Support pod anti-affinity rule for Prometheus
|
||||
|
||||
### BUG FIXES
|
||||
|
||||
- Fix the mistaken highlights in the logs search result
|
||||
- Fix log search not matching phrases correctly
|
||||
- Fix the issue that log could not be retrieved for a deleted workload when it is searched by workload name
|
||||
- Fix the issue where the results were truncated when the log is highlighted
|
||||
- Fix some metrics exceptions: node `inode`, maximum pod tolerance
|
||||
- Fix the issue with an incorrect number of alerting targets
|
||||
- Fix filter failure problem of multi-metric monitoring
|
||||
- Fix the problem of no logging and monitoring information on taint nodes (Adjust the toleration attributes of node-exporter and fluent-bit to deploy on all nodes by default, ignoring taints)
|
||||
|
||||
## DevOps
|
||||
|
||||
### Features
|
||||
|
||||
- Add support for branch exchange and git log export in S2I
|
||||
- Add B2I, ability to build Binary/WAR/JAR package and release to Kubernetes
|
||||
- Support dependency cache for the pipeline, S2I, and B2I
|
||||
- Support delete Kubernetes resource action in `kubernetesDeploy` step
|
||||
- Multi-branch pipeline supports trigger other pipelines when create or delete the branch
|
||||
|
||||
### Upgrades & Enhancement
|
||||
|
||||
- Support BitBucket in the pipeline
|
||||
- Support Cron script validation in the pipeline
|
||||
- Support Jenkinsfile syntax validation
|
||||
- Support custom the link in SonarQube
|
||||
- Support event trigger build in the pipeline
|
||||
- Optimize the agent node selection in the pipeline
|
||||
- Accelerate the start speed of the pipeline
|
||||
- Use dynamical volume as the work directory of the Agent in the pipeline, also contributes to Jenkins [#589](https://github.com/jenkinsci/kubernetes-plugin/pull/598)
|
||||
- Optimize the Jenkins kubernetesDeploy plugin, add more resources and versions (v1, app/v1, extensions/v1beta1、apps/v1beta2、apps/v1beta1、autoscaling/v1、autoscaling/v2beta1、autoscaling/v2beta2、networking.k8s.io/v1、batch/v1beta1、batch/v2alpha1), also contributes to Jenkins [#614](https://github.com/jenkinsci/kubernetes-plugin/pull/614)
|
||||
- Add support for PV, PVC, Network Policy in deploy step of the pipeline, also contributes to Jenkins [#87](https://github.com/jenkinsci/kubernetes-cd-plugin/pull/87)、[#88](https://github.com/jenkinsci/kubernetes-cd-plugin/pull/88)
|
||||
|
||||
### Bug Fixes
|
||||
|
||||
- Fix the issue that 400 bad request in GitHub Webhook
|
||||
- incompatible change: DevOps Webhook's URL prefix is changed from `/webhook/xxx` to `/devops_webhook/xxx`
|
||||
|
||||
## Authentication and authority
|
||||
|
||||
### Features
|
||||
|
||||
Support sync and authenticate with AD account
|
||||
|
||||
### Upgrades & Enhancement
|
||||
|
||||
- Reduce the LDAP component's RAM consumption
|
||||
- Add protection against brute force attacks
|
||||
|
||||
### Bug Fixes
|
||||
|
||||
- Fix LDAP connection pool leak
|
||||
- Fix the issue where users could not be added in the workspace
|
||||
- Fix sensitive data transmission leaks
|
||||
|
||||
## User Experience
|
||||
|
||||
### Features
|
||||
|
||||
Ability to wizard management of projects (namespace) that are not assigned to the workspace
|
||||
|
||||
### Upgrades & Enhancement
|
||||
|
||||
- Support bash-completion in web kubectl
|
||||
- Optimize the host information display
|
||||
- Add connection test of the email server
|
||||
- Add prompt on resource list page
|
||||
- Optimize the project overview page and project basic information
|
||||
- Simplify the service creation process
|
||||
- Simplify the workload creation process
|
||||
- Support real-time status update in the resource list
|
||||
- optimize YAML editing
|
||||
- Support image search and image information display
|
||||
- Add the pod list to the workload page
|
||||
- Update the web terminal theme
|
||||
- Support container switching in container terminal
|
||||
- Optimize Pod information display, and add Pod scheduling information
|
||||
- More detailed workload status display
|
||||
|
||||
### Bug Fixes
|
||||
|
||||
- Fix the issue where the default request resource of the project is displayed incorrectly
|
||||
- Optimize the web terminal design, make it much easier to find
|
||||
- Fix the Pod status update delay
|
||||
- Fix the issue where a host could not be searched based on roles
|
||||
- Fix DevOps project quantity error in workspace detail page
|
||||
- Fix the issue with the workspace list pages not turning properly
|
||||
- Fix the problem of inconsistent result ordering after query on workspace list page
|
||||
|
|
@ -1,122 +0,0 @@
|
|||
---
|
||||
title: "Upgrade KubeSphere and Kubernetes"
|
||||
keywords: "kubernetes, docker, kubesphere, jenkins, istio, prometheus"
|
||||
description: "Upgrade KubeSphere and Kubernetes in Linux machines"
|
||||
|
||||
linkTitle: "Upgrade KubeSphere and Kubernetes"
|
||||
weight: 100
|
||||
---
|
||||
|
||||
KubeSphere 2.1.1 was released on Feb 23rd, 2020, which has fixed known bugs and brought some enhancements. For the users who have installed versions of 2.0.x or 2.1.0, make sure to read the user manual carefully about how to upgrade before doing that, and feel free to raise any questions on [GitHub](https://github.com/kubesphere/kubesphere/issues).
|
||||
|
||||
## What's New in 2.1.1
|
||||
|
||||
## Installer
|
||||
|
||||
### UPGRADE & ENHANCEMENT
|
||||
|
||||
- Support Kubernetes v1.14.x、v1.15.x、v1.16.x、v1.17.x，also solve the issue of Kubernetes API Compatibility#[1829](https://github.com/kubesphere/kubesphere/issues/1829)
|
||||
- Simplify the steps of installation on existing Kubernetes, and remove the step of specifying cluster's CA certification, also specifying Etcd certification is no longer mandatory step if users don't need Etcd monitoring metrics
|
||||
- Backup the configuration of CoreDNS before upgrading
|
||||
|
||||
### BUG FIXES
|
||||
|
||||
- Fix the issue of importing apps to App Store
|
||||
|
||||
## App Store
|
||||
|
||||
### UPGRADE & ENHANCEMENT
|
||||
|
||||
- Upgrade OpenPitrix to v0.4.8
|
||||
|
||||
### BUG FIXES
|
||||
|
||||
- Fix the latest version display issue for the published app #[1130](https://github.com/kubesphere/kubesphere/issues/1130)
|
||||
- Fix the column name display issue in app approval list page #[1498](https://github.com/kubesphere/kubesphere/issues/1498)
|
||||
- Fix the searching issue by app name/workspace #[1497](https://github.com/kubesphere/kubesphere/issues/1497)
|
||||
- Fix the issue of failing to create app with the same name of previously deleted app #[1821](https://github.com/kubesphere/kubesphere/pull/1821) #[1564](https://github.com/kubesphere/kubesphere/issues/1564)
|
||||
- Fix the issue of failing to deploy apps in some cases #[1619](https://github.com/kubesphere/kubesphere/issues/1619) #[1730](https://github.com/kubesphere/kubesphere/issues/1730)
|
||||
|
||||
## Storage
|
||||
|
||||
### UPGRADE & ENHANCEMENT
|
||||
|
||||
- Support CSI plugins of Alibaba Cloud and Tencent Cloud
|
||||
|
||||
### BUG FIXES
|
||||
|
||||
- Fix the paging issue of storage class list page #[1583](https://github.com/kubesphere/kubesphere/issues/1583) #[1591](https://github.com/kubesphere/kubesphere/issues/1591)
|
||||
- Fix the issue that the value of imageFeatures parameter displays '2' when creating ceph storage class #[1593](https://github.com/kubesphere/kubesphere/issues/1593)
|
||||
- Fix the issue that search filter fails to work in persistent volumes list page #[1582](https://github.com/kubesphere/kubesphere/issues/1582)
|
||||
- Fix the display issue for abnormal persistent volume #[1581](https://github.com/kubesphere/kubesphere/issues/1581)
|
||||
- Fix the display issue for the persistent volumes which associated storage class is deleted #[1580](https://github.com/kubesphere/kubesphere/issues/1580) #[1579](https://github.com/kubesphere/kubesphere/issues/1579)
|
||||
|
||||
## Observability
|
||||
|
||||
### UPGRADE & ENHANCEMENT
|
||||
|
||||
- Upgrade Fluent Bit to v1.3.5 #[1505](https://github.com/kubesphere/kubesphere/issues/1505)
|
||||
- Upgrade Kube-state-metrics to v1.7.2
|
||||
- Upgrade Elastic Curator to v5.7.6 #[517](https://github.com/kubesphere/ks-installer/issues/517)
|
||||
- Fluent Bit Operator support to detect the location of soft linked docker log folder dynamically on host machines
|
||||
- Fluent Bit Operator support to manage the instance of Fluent Bit by declarative configuration through updating the ConfigMap of Operator
|
||||
- Fix the issue of sort orders in alert list page #[1397](https://github.com/kubesphere/kubesphere/issues/1397)
|
||||
- Adjust the metric of container memory usage with 'container_memory_working_set_bytes'
|
||||
|
||||
### BUG FIXES
|
||||
|
||||
- Fix the lag issue of container logs #[1650](https://github.com/kubesphere/kubesphere/issues/1650)
|
||||
- Fix the display issue that some replicas of workload have no logs on container detail log page #[1505](https://github.com/kubesphere/kubesphere/issues/1505)
|
||||
- Fix the compatibility issue of Curator to support ElasticSearch 7.x #[517](https://github.com/kubesphere/ks-installer/issues/517)
|
||||
- Fix the display issue of container log page during container initialization #[1518](https://github.com/kubesphere/kubesphere/issues/1518)
|
||||
- Fix the blank node issue when these nodes are resized #[1464](https://github.com/kubesphere/kubesphere/issues/1464)
|
||||
- Fix the display issue of components status in monitor center, to keep them up-to date #[1858](https://github.com/kubesphere/kubesphere/issues/1858)
|
||||
- Fix the wrong monitoring targets number in alert detail page #[61](https://github.com/kubesphere/console/issues/61)
|
||||
|
||||
## DevOps
|
||||
|
||||
### BUG FIXES
|
||||
|
||||
- Fix the issue of UNSTABLE state not visible in the pipeline #[1428](https://github.com/kubesphere/kubesphere/issues/1428)
|
||||
- Fix the format issue of KubeConfig in DevOps pipeline #[1529](https://github.com/kubesphere/kubesphere/issues/1529)
|
||||
- Fix the image repo compatibility issue in B2I, to support image repo of Alibaba Cloud #[1500](https://github.com/kubesphere/kubesphere/issues/1500)
|
||||
- Fix the paging issue in DevOps pipelines' branches list page #[1517](https://github.com/kubesphere/kubesphere/issues/1517)
|
||||
- Fix the issue of failing to display pipeline configuration after modifying it #[1522](https://github.com/kubesphere/kubesphere/issues/1522)
|
||||
- Fix the issue of failing to download generated artifact in S2I job #[1547](https://github.com/kubesphere/kubesphere/issues/1547)
|
||||
- Fix the issue of [data loss occasionally after restarting Jenkins]( https://kubesphere.com.cn/forum/d/283-jenkins)
|
||||
- Fix the issue that only 'PR-HEAD' is fetched when binding pipeline with GitHub #[1780](https://github.com/kubesphere/kubesphere/issues/1780)
|
||||
- Fix 414 issue when updating DevOps credential #[1824](https://github.com/kubesphere/kubesphere/issues/1824)
|
||||
- Fix wrong s2ib/s2ir naming issue from B2I/S2I #[1840](https://github.com/kubesphere/kubesphere/issues/1840)
|
||||
- Fix the issue of failing to drag and drop tasks on pipeline editing page #[62](https://github.com/kubesphere/console/issues/62)
|
||||
|
||||
## Authentication and Authorization
|
||||
|
||||
### UPGRADE & ENHANCEMENT
|
||||
|
||||
- Generate client certification through CSR #[1449](https://github.com/kubesphere/kubesphere/issues/1449)
|
||||
|
||||
### BUG FIXES
|
||||
|
||||
- Fix content loss issue in KubeConfig token file #[1529](https://github.com/kubesphere/kubesphere/issues/1529)
|
||||
- Fix the issue that users with different permission fail to log in on the same browser #[1600](https://github.com/kubesphere/kubesphere/issues/1600)
|
||||
|
||||
## User Experience
|
||||
|
||||
### UPGRADE & ENHANCEMENT
|
||||
|
||||
- Support to edit SecurityContext in workload editing page #[1530](https://github.com/kubesphere/kubesphere/issues/1530)
|
||||
- Support to configure init container in workload editing page #[1488](https://github.com/kubesphere/kubesphere/issues/1488)
|
||||
- Add support of startupProbe, also add periodSeconds, successThreshold, failureThreshold parameters in probe editing page #[1487](https://github.com/kubesphere/kubesphere/issues/1487)
|
||||
- Optimize the status update display of Pods #[1187](https://github.com/kubesphere/kubesphere/issues/1187)
|
||||
- Optimize the error message report on console #[43](https://github.com/kubesphere/console/issues/43)
|
||||
|
||||
### BUG FIXES
|
||||
|
||||
- Fix the status display issue for the Pods that are not under running status #[1187](https://github.com/kubesphere/kubesphere/issues/1187)
|
||||
- Fix the issue that the added annotation can't be deleted when creating service of QingCloud LoadBalancer #[1395](https://github.com/kubesphere/kubesphere/issues/1395)
|
||||
- Fix the display issue when selecting workload on service editing page #[1596](https://github.com/kubesphere/kubesphere/issues/1596)
|
||||
- Fix the issue of failing to edit configuration file when editing 'Job' #[1521](https://github.com/kubesphere/kubesphere/issues/1521)
|
||||
- Fix the issue of failing to update the service of 'StatefulSet' #[1513](https://github.com/kubesphere/kubesphere/issues/1513)
|
||||
- Fix the issue of image searching for QingCloud and Alibaba Cloud image repos #[1627](https://github.com/kubesphere/kubesphere/issues/1627)
|
||||
- Fix resource ordering issue with the same creation timestamp #[1750](https://github.com/kubesphere/kubesphere/pull/1750)
|
||||
- Fix the issue of failing to edit configuration file when editing service #[41](https://github.com/kubesphere/console/issues/41)
|
||||
|
|
@ -1,10 +0,0 @@
|
|||
---
|
||||
title: "Overview"
|
||||
keywords: "kubernetes, upgrade, kubesphere, v3.0.0"
|
||||
description: "Upgrade KubeSphere"
|
||||
|
||||
linkTitle: "Overview"
|
||||
weight: 50
|
||||
---
|
||||
|
||||
TBD
|
||||
|
|
@ -0,0 +1,45 @@
|
|||
---
|
||||
title: "FAQ"
|
||||
keywords: "kubernetes, upgrade, kubesphere, v3.0.0"
|
||||
description: "KubeSphere Upgrade FAQ"
|
||||
|
||||
linkTitle: "FAQ"
|
||||
weight: 200
|
||||
---
|
||||
|
||||
## How to upgrade Qingcloud CSI after upgrading?
|
||||
|
||||
Currently Qingcloud CSI will not be upgraded by KubeKey. You can run the following command to upgrade CSI manually after finishing KubeSphere upgrade:
|
||||
|
||||
```
|
||||
git clone https://github.com/yunify/qingcloud-csi.git
|
||||
cd qingcloud-csi/
|
||||
git checkout v1.1.1
|
||||
|
||||
kubectl delete -f deploy/disk/kubernetes/releases/qingcloud-csi-disk-v1.1.1.yaml
|
||||
kubectl delete sc csi-qingcloud
|
||||
|
||||
helm repo add test https://charts.kubesphere.io/test
|
||||
helm install test/csi-qingcloud --name-template csi-qingcloud --namespace kube-system \
|
||||
--set config.qy_access_key_id=KEY,config.qy_secret_access_key=SECRET,config.zone=ZONE,sc.type=2
|
||||
```
|
||||
|
||||
Wait until csi controller and daemonset are running
|
||||
|
||||
```
|
||||
$ kubectl get po -n kube-system | grep csi
|
||||
csi-qingcloud-controller-56979d46cb-qk9ck 5/5 Running 0 24h
|
||||
csi-qingcloud-node-4s8n5 2/2 Running 0 24h
|
||||
csi-qingcloud-node-65dqn 2/2 Running 0 24h
|
||||
csi-qingcloud-node-khk49 2/2 Running 0 24h
|
||||
csi-qingcloud-node-nz9q9 2/2 Running 0 24h
|
||||
csi-qingcloud-node-pxr56 2/2 Running 0 24h
|
||||
csi-qingcloud-node-whqhk 2/2 Running 0 24h
|
||||
```
|
||||
|
||||
Then run the following command to check csi image version is 1.2.x
|
||||
|
||||
```
|
||||
$ kubectl get po -n kube-system csi-qingcloud-controller-56979d46cb-qk9ck -ojson | jq '.spec.containers[].image' | grep qingcloud
|
||||
"csiplugin/csi-qingcloud:v1.2.0-rc.4"
|
||||
```
|
||||
|
|
@ -0,0 +1,28 @@
|
|||
---
|
||||
title: "Overview"
|
||||
keywords: "kubernetes, upgrade, kubesphere, v3.0.0"
|
||||
description: "KubeSphere Upgrade Overview"
|
||||
|
||||
linkTitle: "Overview"
|
||||
weight: 50
|
||||
---
|
||||
|
||||
## Kubernetes
|
||||
|
||||
KubeSphere v3.0.0 is compatible with Kubernetes 1.15.x, 1.16.x, 1.17.x and 1.18.x:
|
||||
|
||||
- If your KubeSphere v2.1.x is installed on Kubernetes 1.15.x+ , you can choose to only upgrade KubeSphere to v3.0.0 or upgrade Kubernetes (to a higher version) and KubeSphere (to v3.0.0) at the same time.
|
||||
|
||||
- If your KubeSphere v2.1.x is installed on Kubernetes 1.14.x, you have to upgrade Kubernetes (to 1.15.x+) and KubeSphere (to v3.0.0 ) at the same time.
|
||||
|
||||
{{< notice warning >}}
|
||||
There're some significant API changes in Kubernetes 1.16.x comparing with prior versions 1.14.x and 1.15.x, please refer to [Deprecated APIs Removed In 1.16: Here’s What You Need To Know](https://kubernetes.io/blog/2019/07/18/api-deprecations-in-1-16/) for more details. So if you plan to upgrade from Kubernetes 1.14.x/1.15.x to 1.16.x+, you'll have to migrate some of your workloads after upgrading.
|
||||
{{</ notice >}}
|
||||
|
||||
## How
|
||||
|
||||
A brand new installer [KubeKey](https://github.com/kubesphere/kubekey) is introduced in KubeSphere v3.0.0, with which you can install or upgrade Kubernetes and KubeSphere. More details about upgrading with [KubeKey](https://github.com/kubesphere/kubekey) will be covered in the following sections.
|
||||
|
||||
### KubeKey or ks-installer?
|
||||
|
||||
[ks-installer](https://github.com/kubesphere/ks-installer/tree/master) was the main installation tool as of KubeSphere v2. For users whose Kubernetes clusters were NOT deployed via [KubeSphere Installer](https://v2-1.docs.kubesphere.io/docs/installation/all-in-one/#step-2-download-installer-package), they should choose ks-installer to upgrade KubeSphere. For example, if your Kubernetes is hosted by cloud vendors or self provisioned, you should go for [Upgrade with ks-installer](../upgrade-with-ks-installer).
|
||||
|
|
@ -0,0 +1,44 @@
|
|||
---
|
||||
title: "Upgrade with ks-installer"
|
||||
keywords: "kubernetes, upgrade, kubesphere, v3.0.0"
|
||||
description: "Upgrade KubeSphere with ks-installer"
|
||||
|
||||
linkTitle: "Upgrade with ks-installer"
|
||||
weight: 150
|
||||
---
|
||||
|
||||
ks-installer is recommended for users whose Kubernetes clusters were not setup via [KubeSphere Installer](https://v2-1.docs.kubesphere.io/docs/installation/all-in-one/#step-2-download-installer-package), but hosted by cloud vendors. This tutorial guides to **upgrade KubeSphere only**. Cluster operator are responsible for upgrading Kubernetes on themselves beforehand.
|
||||
|
||||
## Prerequisite
|
||||
|
||||
- You need to have a KubeSphere cluster running version 2.1.1.
|
||||
|
||||
{{< notice warning >}}
|
||||
If your KubeSphere version is v2.1.0 or earlier, please upgrade to v2.1.1 first.
|
||||
{{</ notice >}}
|
||||
|
||||
- Make sure you read the release notes carefully
|
||||
|
||||
{{< notice warning >}}
|
||||
In v3.0.0, KubeSphere refactors many of its components such as Fluent Bit Operator, IAM, etc. Make sure you back up any important components in case you heavily customized them but not from console.
|
||||
{{</ notice >}}
|
||||
|
||||
## Step 1. Download YAML files
|
||||
|
||||
The following are configuration templates.
|
||||
|
||||
```
|
||||
wget https://raw.githubusercontent.com/kubesphere/ks-installer/release-3.0/deploy/kubesphere-installer.yaml
|
||||
wget https://raw.githubusercontent.com/kubesphere/ks-installer/release-3.0/deploy/cluster-configuration.yaml
|
||||
```
|
||||
|
||||
## Step 2. Modify the configuration file template
|
||||
|
||||
Sync the changes from the v2.1.1 to v3.0.0 into the config section of `cluster-configuration.yaml`. Note that the storage class and the pluggable components need to be consistent with the v2.1.1.
|
||||
|
||||
## Step 3. Apply YAML files
|
||||
|
||||
```
|
||||
kubectl apply -f kubesphere-installer.yaml
|
||||
kubectl apply -f cluster-configuration.yaml
|
||||
```
|
||||
|
|
@ -0,0 +1,107 @@
|
|||
---
|
||||
title: "Upgrade with KubeKey"
|
||||
keywords: "kubernetes, upgrade, kubesphere, v3.0.0"
|
||||
description: "Upgrade KubeSphere with kubekey"
|
||||
|
||||
linkTitle: "Upgrade with KubeKey"
|
||||
weight: 100
|
||||
---
|
||||
KubeKey is recommended for users whose KubeSphere and Kubernetes were both deployed by [KubeSphere Installer](https://v2-1.docs.kubesphere.io/docs/installation/all-in-one/#step-2-download-installer-package). If your Kubernetes cluster was provisioned by yourself or cloud providers, please refer to [Upgrade with ks-installer](../upgrade-with-ks-installer).
|
||||
|
||||
## Prerequisite
|
||||
|
||||
- You need to have a KubeSphere cluster running version 2.1.1.
|
||||
|
||||
{{< notice warning >}}
|
||||
If your KubeSphere version is v2.1.0 or earlier, please upgrade to v2.1.1 first.
|
||||
{{</ notice >}}
|
||||
|
||||
- Download KubeKey
|
||||
|
||||
```
|
||||
curl -O -k https://kubernetes.pek3b.qingstor.com/tools/kubekey/kk
|
||||
chmod +x kk
|
||||
```
|
||||
|
||||
- Make sure you read the release notes carefully
|
||||
|
||||
{{< notice warning >}}
|
||||
In v3.0.0, KubeSphere refactors many of its components such as Fluent Bit Operator, IAM, etc. Make sure you back up any important components in case you heavily customized them but not from console.
|
||||
{{</ notice >}}
|
||||
|
||||
- Make your upgrade plan. The two upgrading scenarios are documented below.
|
||||
|
||||
## Upgrade KubeSphere Only
|
||||
The following command brings your KubeSphere to v3.0.0 without touching Kubernetes control planes:
|
||||
|
||||
```
|
||||
./kk upgrade --with-kubesphere
|
||||
```
|
||||
|
||||
## Upgrade KubeSphere and Kubernetes
|
||||
|
||||
Upgrading steps are different for single-node clusters (all in one) and multi-node clusters.
|
||||
|
||||
{{< notice info >}}
|
||||
Upgrading with Kubernetes will cause helm to be upgraded from v2 to v3. If you want to continue using helm2, please backup it: `cp /usr/local/bin/helm /usr/local/bin/helm2`
|
||||
{{</ notice >}}
|
||||
|
||||
{{< notice info >}}
|
||||
When upgrading Kubernetes, KubeKey will upgrade from one MINOR version to the next MINOR version until the target version. For example, you may observe the upgrading process going through 1.16, 1.17 and 1.18, but not jumping to 1.18 from 1.16.
|
||||
{{</ notice >}}
|
||||
|
||||
### Allinone
|
||||
|
||||
The following command upgrades your single-node cluster to KubeSphere v3.0.0 and Kubernetes v1.17.9 (default):
|
||||
|
||||
```
|
||||
./kk upgrade --with-kubesphere --with-kubernetes
|
||||
```
|
||||
|
||||
To upgrade Kubernetes to a specific version, please explicitly provide the version after the flag `--with-kubernetes`. Available versions are:
|
||||
|
||||
- v1.15.12
|
||||
- v1.16.8, v1.16.10, v1.16.12, v1.16.13
|
||||
- v1.17.0, v1.17.4, v1.17.5, v1.17.6, v1.17.7, v1.17.8, v1.17.9
|
||||
- v1.18.3, v1.18.5, v1.18.6
|
||||
|
||||
### Multi-Nodes
|
||||
|
||||
#### Step1. Generate KubeKey configuration file
|
||||
|
||||
This commad creates KubeKey configuration file onto `config-sample.yaml` from your cluster.
|
||||
|
||||
```
|
||||
./kk create config --from-cluster
|
||||
```
|
||||
|
||||
{{< notice note >}}
|
||||
It assumes your kubeconfig is allocated in `~/.kube/config`. You can change it with the flag `--kubeconfig`.
|
||||
{{</ notice >}}
|
||||
|
||||
#### Step 2. Modify the configuration file template
|
||||
|
||||
Modify `config-sample.yaml` to fit your cluster setup. Make sure you replace the following fields correctly.
|
||||
|
||||
- `hosts`: Fill connection information among your hosts.
|
||||
- `roleGroups.etcd`: Fill etcd memebers.
|
||||
- `controlPlaneEndpoint`: Fill your load balancer address (Optional)
|
||||
- `registry`: Fill image registry information (Optional)
|
||||
|
||||
{{< notice note >}}
|
||||
Please refer to the Cluster section of [config-example.yaml](https://github.com/kubesphere/kubekey/blob/master/docs/config-example.md) for more information.
|
||||
{{</ notice >}}
|
||||
|
||||
#### Step 3. Upgrade your cluster
|
||||
The following command upgrades your cluster to KubeSphere v3.0.0 and Kubernetes v1.17.9 (default):
|
||||
|
||||
```
|
||||
./kk upgrade --with-kubesphere --with-kubernetes -f config-sample.yaml
|
||||
```
|
||||
|
||||
To upgrade Kubernetes to a specific version, please explicitly provide the version after the flag `--with-kubernetes`. Available versions are:
|
||||
|
||||
- v1.15.12
|
||||
- v1.16.8, v1.16.10, v1.16.12, v1.16.13
|
||||
- v1.17.0, v1.17.4, v1.17.5, v1.17.6, v1.17.7, v1.17.8, v1.17.9
|
||||
- v1.18.3, v1.18.5, v1.18.6
|
||||
|
|
@ -0,0 +1,10 @@
|
|||
---
|
||||
title: "What Changed in 3.0.0"
|
||||
keywords: "kubernetes, upgrade, kubesphere, v3.0.0"
|
||||
description: "KubeSphere Upgrade"
|
||||
|
||||
linkTitle: "What Changed in 3.0.0"
|
||||
weight: 250
|
||||
---
|
||||
|
||||
There are some changes in access control in 3.0. We simplified the definition of custom roles, aggregated some closely related permission items into permission groups. The custom role will not change during the upgrade, custom roles that match the new policy rules can be used directly, otherwise you need to manually modify them according to the document.
|
||||
|
|
@ -5,7 +5,7 @@ createTime: '2019-06-25'
|
|||
snapshot: 'https://pek3b.qingstor.com/kubesphere-docs/png/20190930095954.png'
|
||||
---
|
||||
|
||||
日志通常含有非常有价值的信息，日志管理是云原生可观察性的重要组成部分。不同于物理机或虚拟机，在容器与 Kubernetes 环境中，日志有标准的输出方式(stdout)，这使得进行平台级统一的日志收集、分析与管理水到渠成，并体现出日志数据独特的价值。本文将介绍云原生领域比较主流的日志管理方案 EFK 、 KubeSphere 团队开发的 FluentBit Operator 以及 KubeSphere 在多租户日志管理方面的实践。此外还将介绍受 Prometheus 启发专为 Kubenetes 日志管理开发，具有低成本可扩展等特性的开源软件 Loki。
|
||||
日志通常含有非常有价值的信息，日志管理是云原生可观察性的重要组成部分。不同于物理机或虚拟机，在容器与 Kubernetes 环境中，日志有标准的输出方式(stdout)，这使得进行平台级统一的日志收集、分析与管理水到渠成，并体现出日志数据独特的价值。本文将介绍云原生领域比较主流的日志管理方案 EFK 、 KubeSphere 团队开发的 FluentBit Operator 以及 KubeSphere 在多租户日志管理方面的实践。此外还将介绍受 Prometheus 启发专为 Kubernetes 日志管理开发，具有低成本可扩展等特性的开源软件 Loki。
|
||||
|
||||
## 什么是可观察性
|
||||
|
||||
|
|
|
|||
Loading…
Reference in New Issue