Skip to content

Commit

Permalink
update
Browse files Browse the repository at this point in the history
  • Loading branch information
@chongyuanyin
chongyuanyin committed Aug 8, 2024
1 parent 0f470bd commit 3319631
Show file tree
Hide file tree
Showing 11 changed files with 120 additions and 52 deletions.
87 changes: 53 additions & 34 deletions ecp/en_US/cluster/add_manage.md
View file
Original file line number Diff line number Diff line change
@@ -1,20 +1,27 @@
# Add EMQX Clusters

ECP supports adding clusters by creating (recommended) or adding existing EMQX clusters. It is recommended to add clusters by creating with ECP, which offers more extensive functionality and allows for license and connection quota sharing.
ECP supports adding clusters by creating (recommended) or adding existing EMQX clusters:

- Creating clusters with ECP offers more extensive functionality and allows for license and connection quota sharing.
- Adding clusters into ECP allows easy management for existing clusters. ECP supports management for EMQX v4 Enterprise Edition (4.4.6 and above) and EMQX v5 Enterprise Edition (5.6.0 and above).

There are functional differences between creating (**Hosted Clusters**) and managing clusters (**Managed Clusters**) on the ECP platform, as shown in the table below.

|Function|Hosted Clusters|Managed Clusters|
|:--------:|:----:|:----:|
|Start/Stop|✅|❌|
|Horizontal Scaling|✅|❌|
|Vertical Scaling|✅|❌|
|Update Network Type|✅|❌|
|Update Connection Limit|✅|❌|
|Upgrade/Downgrade|✅|❌|
|Cluster Transfer|✅|✅|
|Delete|✅|✅|
|Log|✅|✅|
|Function|Hosted v4 Clusters|Managed v4 Clusters|Managed v5 Clusters|
|:--------:|:----:|:----:|:----:|
|Start/Stop|✅|❌|❌|
|Deletion|✅|✅|✅|
|Horizontal Scaling|✅|❌|❌|
|Vertical Scaling|✅|❌|❌|
|Update Network Type|✅|❌|❌|
|Update Connection Limit|✅|❌|✅*|
|Upgrade/Downgrade|✅|❌|❌|
|Log|✅|✅|✅|
|Cluster Monitor|❌|❌|✅|
|Cluster Alarm|❌|❌|✅|
|Cluster Transfer|✅|✅|✅|

\* For managed clusters, **Update Connection Limit** feature applies to EMQX v5.7.0 and above.

## Create a Hosted Cluster

Expand All @@ -33,6 +40,30 @@ The newly-created clusters will be listed in the **Cluster List** panel with the

<img src="./_assets/cluster-running.png" alt="cluster-running" style="zoom:50%;" />

## Status for Hosted Cluster

You can start or stop a cluster as your business requirement changes.

1. Log in as system admin, organization admin, or project admin.
2. On the target cluster, click the more icon and select **Stop**/**Start**.

Hosted EMQX cluster can be in the following states:

| Status | Description |
| ------------------ | ------------------------------------------------------------ |
| Creating | Intermediate state during the process of new cluster creation |
| Updating | Intermediate state during cluster OM operations, such as horizontal or vertical scaling, network type modifications, connection number modifications, cluster upgrade or downgrade |
| Starting | When starting the service |
| Running | Normal running state of the cluster |
| Stopping | When stopping the service or an intermediate state after deleting a cluster |
| Stopped | After stopping or deleting |
| Syncing Status | Intermediate state during horizontal or vertical scaling, cluster upgrade or downgrade, network type modifications, connection number modifications |
| Downgraded Running | One or more nodes of the cluster are unavailable, but the overall cluster is still usable |
| Error | The most recent task executed by the cluster failed (can auto-recover), or a cluster fault or dirty data occurred (this state rarely appears)<!--shall we remove the dirty data part?--> |
| Nonexistent | The task to create the cluster was not successfully issued |

For clusters in the state of Error, you can click the more icon and click **Try Fix**. If the problem is successfully solved, the cluster state will be Running; or consider deleting the cluster or reaching out to EMQ's technical support.

## Add an Existing Cluster

ECP also provides the capability to manage existing EMQX clusters. ECP supports the management of EMQX v4 (version 4.4.6 and above) and EMQX v5 (version 5.6.0 and above).
Expand Down Expand Up @@ -109,30 +140,18 @@ ECP also provides the capability to manage existing EMQX clusters. ECP supports
![cluster-v5-dashboard](./_assets/cluster-v5-dashboard.png)


## Cluster Status

You can start or stop a cluster as your business requirement changes.

1. Log in as system admin, organization admin, or project admin.
2. On the target cluster, click the more icon and select **Stop**/**Start**.
## Status for Managed Cluster

Managed EMQX cluster can be in the following states:

| Status | Description |
| ----------- | ------------------------------------------------------------ |
| Created | Cluster with no node registered yet |
| Registering | Intermediate state during cluster node registration |
| Running | Normal running state of the cluster |
| Deleting | Intermediate state before cluster deletion completes |
| Error | Abnormal running state of the cluster, or network connection issue between agent and cluster or between agent and ECP |

EMQX cluster can be in the following states:

| Status | Description |
| ------------------ | ------------------------------------------------------------ |
| Creating | Intermediate state during the process of new cluster creation |
| Updating | Intermediate state during cluster OM operations, such as horizontal or vertical scaling, network type modifications, connection number modifications, cluster upgrade or downgrade |
| Starting | When starting the service |
| Running | Normal running state of the cluster |
| Stopping | When stopping the service or an intermediate state after deleting a cluster |
| Stopped | After stopping or deleting |
| Syncing Status | Intermediate state during horizontal or vertical scaling, cluster upgrade or downgrade, network type modifications, connection number modifications |
| Downgraded Running | One or more nodes of the cluster are unavailable, but the overall cluster is still usable |
| Error | The most recent task executed by the cluster failed (can auto-recover), or a cluster fault or dirty data occurred (this state rarely appears)<!--shall we remove the dirty data part?--> |
| Nonexistent | The task to create the cluster was not successfully issued |

For clusters in the state of Error, you can click the more icon and click **Try Fix**. If the problem is successfully solved, the cluster state will be Running; or consider deleting the cluster or reaching out to EMQ's technical support.
For clusters in the state of Error, you can click the Error status icon to view possible cause.

<!--also the English for the status should be confirmed-->
6 changes: 3 additions & 3 deletions ecp/en_US/cluster/introduction.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -4,14 +4,14 @@

In ECP, clusters refer to the EMQX clusters deployed on cloud servers, which serve as high-performance message broker for IoT devices. Built on the MQTT protocol, EMQX offers a lightweight, reliable, and scalable solution for communication between IoT devices. It excels in real-time, high availability, and easy implementation. For a comprehensive understanding of EMQX, please refer to the detailed documentation available on [EMQX Enterprise](https://docs.emqx.com/en/enterprise/v4.4/).

With ECP's cluster management features, users can efficiently handle multiple clusters, create new ones, onboard existing ones, and perform various tasks like troubleshooting, scaling, modifying network types, adjusting connections, upgrading/downgrading, transferring ownership, and deletion. The platform's user-friendly interface offers cluster information overview and log access for improved visibility.
With ECP's cluster management features, users can efficiently handle multiple clusters, create new ones, onboard existing ones, and perform various tasks like troubleshooting, scaling, modifying network types, adjusting connections, upgrading/downgrading, transferring ownership, and deletion. The platform's user-friendly interface offers cluster information overview, monitor, alarm and log access for improved visibility.

## Access Cluster Workspace

After logging in, you can find the **Workspace** option in the ribbon area. Click on it to navigate to the **Workspace - Cluster** page. This page provides an overview of the EMQX clusters hosted or managed by ECP and displays this project's current number of members.
After logging in, you can find the **Workspace** option in the ribbon area. Click on it to navigate to the **Workspace - Cluster** page. This page provides an overview of the EMQX clusters hosted or managed by ECP.

:::tip
System admin, organization admin, project admin, and regular users all can access this page, however, regular users do not have access to the administration page, and the **Workspace - Cluster** page serves as their landing page.
System admin, organization admin, project admin, and regular users all can access this page.

For the permission of each role, see [Permissions and Roles](../acl/authorize.md#roles-and-permissions).
:::
Expand Down
43 changes: 35 additions & 8 deletions ecp/en_US/cluster/ops.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -86,7 +86,7 @@ To view the external IP and port, click on the Cluster name (or Cluster ID), or
![LoadBalancer](./_assets/cluster-loadbalancer.png)


## Update Connect Limit (Kubernetes Deployment)
## Update Connect Limit

You can dynamically modify the number of connections in the cluster according to business needs:

Expand All @@ -95,9 +95,9 @@ You can dynamically modify the number of connections in the cluster according to

However, please be noted that:

1. The number of connections is limited by the total number of connections allowed by the license.
2. If you are using the LoadBalancer network type, please avoid modifying the number of connections if not necessary, otherwise, it will cause the LoadBalancer's IP address to change.
1. For hosted cluster, if you are using the LoadBalancer network type, please avoid modifying the number of connections if not necessary, otherwise, it will cause the LoadBalancer's IP address to change.
2. For managed cluster, This feature applies to EMQX v5.7.0 and above. The license quota on ECP will be restored once unregistration or cluster deletion completes, and the connections assigned to EMQX cluster will be reclaimed then. Please reset the EMQX license via **Reset License** from EMQX dashboard.
3. For clusters in the state of Error, you can click the Error status icon to view possible cause.


## Upgrade (Kubernetes Deployment)
Expand Down Expand Up @@ -128,17 +128,44 @@ For easier management, ECP provides a feature for transferring EMQX clusters acr
ECP offers a unified log feature.

1. Log in as system admin, organization admin, or project admin.

2. On the target cluster, click the more icon and select **Log**.
2. If the existing cluster is added into ECP, please enable and specify logs location when registering cluster node.
- parameter for enabling log collection: `--emqx-log-collection-enabled`
- parameter for cluster log directory path: `--emqx-log-collection-dir /opt/emqx/log`. If the cluster is installed via docker, please mount the log directory on the host machine into the container, and use the directory path on the host machine in the parameter.
3. On the target cluster, click the more icon and select **Log**.

You will be directed to the **Log** page, where you can view the log level, generated time, and log messages. For more information on logs, see [Logs](../log/introduction.md).

<img src="./_assets/cluster-log.png" alt="log" style="zoom:50%;" />

## Delete Clusters
## Delete Cluster

For unused clusters, it's advisable to delete them to save IT resources.

1. Log in as system admin, organization admin, or project admin.

2. On the target cluster, click the more icon and select **Delete** and confirm the action. ECP will first stop the cluster before proceeding with the deletion.

## Monitor Cluster

ECP provides the status overview for managed clusters on the **Cluster Monitor** page. For details, see [Monitor EMQX Clusters](../monitor/monitor_cluster.md)

:::tip

Monitoring on clusters feature applies to EMQX v5.

:::

## View Cluster Alarms

ECP provides the alarm management for cluster rules and connectors on the **Alarm** page.

To notify cluster alarms by email or Webhook, enable "Push EMQX Alarm" when creating notification.

<img src="./_assets/cluster-alarm-notification.png" style="zoom: 80%;" align="middle">

For other details of alarm, see [Alarms](../monitor/alarm_rules.md)

:::tip

Cluster alarms feature applies to EMQX v5.

::
2 changes: 1 addition & 1 deletion ecp/en_US/edge_service/edge_project_statistics.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -34,4 +34,4 @@ After finishing creating the edge service instances or adding existing edge serv

## Driver and Rule Lists

Underneath these two cards, you'll find the driver and rule lists for edge services, like list for not running drivers, list for abnormal drivers, list for not running rules. These lists provide name and type for each driver or rule, and details about instance it runs on, including edge service name, status, endpoint and version. You can filter edge services in the list by nam. Moreover, you can perform O&M actions per instance by clicking **Details** button from Action column.
Underneath these two cards, you'll find the driver and rule lists for edge services, like list for not running drivers, list for abnormal drivers, list for not running rules. These lists provide name and type for each driver or rule, and details about instance it runs on, including edge service name, status, endpoint and version. You can filter edge services in the list by name. Moreover, you can perform O&M actions per instance by clicking **Details** button from Action column.
Binary file added ecp/en_US/monitor/_assets/cluster-monitor.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified ecp/zh_CN/monitor/_assets/alert-notification.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified ecp/zh_CN/monitor/_assets/alert-rules.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added ecp/zh_CN/monitor/_assets/cluster-monitor.png
View file
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
15 changes: 10 additions & 5 deletions ecp/zh_CN/monitor/alarm_rules.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -50,7 +50,7 @@ ECP 支持对历史告警的清理，在**历史告警**选项卡中点击 **历

如果沉默时效对象设置为“单条告警级别”，在沉默时效时间范围内，ECP 不会就同一条告警反复发送通知；超过沉默时效后，将会继续推送告警通知。

如果沉默时效对象设置为“边缘服务实例级别”，在沉默时效时间范围内，同一个边缘服务上产生的任何告警都不会触发反复通知；只有超过沉默时效后，ECP 才会继续推送告警通知。
如果沉默时效对象设置为“边缘服务/集群实例级别”，在沉默时效时间范围内，同一个边缘服务或同一个集群上产生的任何告警都不会触发反复通知；只有超过沉默时效后，ECP 才会继续推送告警通知。

通知沉默时效只影响邮件及 Webhook 的告警推送，不论如何选择，所有的告警事件仍将展示在当前/历史告警选项卡中。

Expand All @@ -60,9 +60,9 @@ ECP 支持对历史告警的清理，在**历史告警**选项卡中点击 **历

![alert_rules](./_assets/alert-rules.png)

ECP 目前支持的告警规则包括边缘服务触发的规则，及 ECP 触发的规则两类。由边缘服务触发的规则包括：驱动异常、规则异常、NeuronEX 重启；由 ECP 触发的规则包括：NeuronEX 离线、邮件发送失败、Webhook 发送失败。规则的详情请参见[统一运维服务 - 告警规则列表](../monitor/rules.md)。
ECP 目前支持的告警规则包括边缘服务触发的规则，云端集群触发的规则，及 ECP 端通用规则三类。由边缘服务触发的规则包括：驱动异常、规则异常、NeuronEX 离线、NeuronEX 重启；由云端集群触发的规则包括：EMQX 规则异常、EMQX 连接器异常；ECP 端通用规则包括：邮件发送失败、Webhook 发送失败。规则的详情请参见[统一运维服务 - 告警规则列表](../monitor/rules.md)。

除 **NeuronEX 重启**告警规则外，您可以对告警规则的触发条件和恢复条件进行设置。如果希望告警更为敏锐，您可以设置较小的触发值；如果希望告警不要过于频繁地触发，您可以设置较大的触发值。触发值和恢复值目前可设置的上限为10。
除 **NeuronEX 重启**告警规则外，您可以对其他告警规则的触发条件和恢复条件进行设置。如果希望告警更为敏锐，您可以设置较小的触发值；如果希望告警不要过于频繁地触发，您可以设置较大的触发值。触发值和恢复值目前可设置的上限为10。

您也可以对所有的告警规则设置“严重”或“一般”的告警级别用于标识严重程度。如果通知范围指定为“仅通知严重告警事件”，那么一般级别的告警将只会显示在当前/历史告警选项卡中，不会推送到邮件及 Webhook 中。

Expand All @@ -72,16 +72,20 @@ ECP 目前支持的告警规则包括边缘服务触发的规则，及 ECP 触

![alarm-notification-config](./_assets/alarm-notification-config.png)

ECP 支持设置一个或多个告警推送，不同的告警推送通过指定标签关联到边缘服务上，当这些关联的边缘服务上产生告警时，告警将发送到相应的邮件及 Webhook 上。
ECP 支持设置一个或多个告警推送，不同的告警推送通过指定标签关联到边缘服务上，或者在打开“EMQX 告警推送”的前提下关联到集群上。当这些关联的边缘服务或 EMQX 集群产生告警时，告警将发送到相应的邮件及 Webhook 上。

<img src="./_assets/alert-notification.png" style="zoom: 50%;" align="middle">
<img src="./_assets/alert-notification.png" style="zoom: 80%;" align="middle">

### 被告警的边缘服务标签

如果选择“全部”，项目中任何边缘服务上产生的告警都将被发送到当前告警推送设置的邮件及 Webhook 上。也可以选择一个或多个标签，只有关联了标签的边缘服务产生的告警才会发送到设置的邮件及 Webhook 上。

请注意：如果告警是项目级别的，如邮件发送失败告警、Webhook 发送失败告警，则所有的告警推送中配置的邮件及 Webhook 都将收到通知。

### EMQX 告警推送

如果开启 EMQX 告警推送，项目中任何集群上产生的告警都将被发送到当前告警推送设置的邮件及 Webhook 上。

### 邮件通知

1. 滑动**邮件通知**开关为启用状态。
Expand Down Expand Up @@ -159,3 +163,4 @@ Content-Type: application/json
"uuid": "of9MHKAj"
}
```

15 changes: 15 additions & 0 deletions ecp/zh_CN/monitor/monitor_cluster.md
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,15 @@
# EMQX 集群监控

ECP 在 **集群监控** 页面提供了项目级别的 EMQX 纳管集群的运行状态统计信息。

![cluster-monitor](./_assets/cluster-monitor.png)

## 基础统计信息

- 连接数：项目内所有纳管集群的总连接数量、在线连接数量。
- 规则：项目内所有纳管集群的规则总数、运行中规则数量及停止的规则数量。
- 连接器：项目内所有纳管集群的连接器总数、已连接的连接器数量及断开连接的连接器数量。

### 规则和连接器列表

您还可在统计卡片下查看目前项目 ECP 纳管的 EMQX 集群中，停止运行的规则和断开连接的连接器的详细信息。您可以通过集群的名称对列表内容进行筛选。此外，您也可以通过点击操作栏中的“详情”按钮，进入该集群的 Dashboard 进行查看和操作。
Loading

0 comments on commit 3319631

Please sign in to comment.