This is an automated email from the ASF dual-hosted git repository.
morningman pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/doris-website.git
The following commit(s) were added to refs/heads/master by this push:
new b73ad829e7d [docs](4.x) Clarify Operator tokens and management
credentials (#4654)
b73ad829e7d is described below
commit b73ad829e7d8423bd93d39be8478b9d684a6fbec
Author: Linying Assad <[email protected]>
AuthorDate: Wed Nov 18 17:53:56 2026 +0800
[docs](4.x) Clarify Operator tokens and management credentials (#4654)
## Summary
Clarify how to configure and rotate the MetaService HTTP token,
including the restart requirement and its distinction from Doris
management credentials. Document how missing or mismatched `authSecret`
credentials prevent BE registration, provide complete Secret examples,
and explain how to recover failed BE Pods after correcting credentials.
Also correct related internal links in the affected pages. This update
is intentionally limited to the English and Chinese 4.x documentation;
current and 3.x are outside the requested scope.
## Validation
- Compiled all six changed documents with the Docusaurus Markdown
processor.
- Parsed all six complete Secret YAML examples and verified matching
English/Chinese code examples.
- Ran the documentation link checker for changed files: no errors or
warnings.
- Passed `git diff --check`.
- Previewed the Chinese 4.x documentation locally and confirmed the
rendered content.
## Versions
- [ ] dev
- [x] 4.x
- [ ] 3.x
- [ ] 2.1 or older (not covered by version/language sync gate)
## Languages
- [x] Chinese
- [x] English
## Docs Checklist
- [x] Checked by AI
- [x] Test Cases Built
- [x] Updated required version and language counterparts, or explained
why not
- [x] If only one language changed, confirmed whether source/translation
counterparts need sync
---
.../doris-operator/status-and-troubleshooting.md | 39 +++++++++++++++++++
.../separating-storage-compute/config-cluster.md | 44 +++++++++++++++++-----
.../separating-storage-compute/config-ms.md | 13 +++++--
.../doris-operator/status-and-troubleshooting.md | 39 +++++++++++++++++++
.../separating-storage-compute/config-cluster.md | 40 ++++++++++++++++----
.../separating-storage-compute/config-ms.md | 13 +++++--
6 files changed, 164 insertions(+), 24 deletions(-)
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/install/deploy-on-kubernetes/doris-operator/status-and-troubleshooting.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/install/deploy-on-kubernetes/doris-operator/status-and-troubleshooting.md
index 51f17329bdd..6c730d033cd 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/install/deploy-on-kubernetes/doris-operator/status-and-troubleshooting.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/install/deploy-on-kubernetes/doris-operator/status-and-troubleshooting.md
@@ -202,6 +202,45 @@ kubectl logs deployment/${operator_deployment_name} -n
${operator_namespace}
如果集群已设置 root 密码,或者使用了非默认管理账号,需要确认 `authSecret` 与实际 Doris 凭证一致。
+#### 存算分离集群中 BE 因凭据错误无法注册
+
+以下步骤适用于使用 `authSecret` 登录 FE、通过 SQL 注册节点的 BE 镜像(例如 Doris 4.1.5)。
+
+| 现象 | 排查方向 |
+|------|----------|
+| Pod 事件出现 `FailedMount`,提示 Secret 不存在 | 检查 Secret 名称及命名空间;Secret
必须与集群位于同一命名空间 |
+| BE 注册日志持续出现 `Access denied` / `1045`,节点未加入 | 检查 Secret 的
`username`、`password` 是否与数据库一致;脚本先尝试 root 空密码,单次报错不一定表示后续凭据认证失败 |
+| 使用管理账户可以登录 FE,但节点管理 SQL 被拒绝 | 检查账户是否具备所需节点管理权限 |
+| Secret 已修正,但 BE 仍使用旧凭据注册失败 | 启动脚本已将凭据读入变量,注册重试不会重新读取 Secret |
+
+先查看目标 BE Pod 的事件和注册日志:
+
+```shell
+kubectl describe pod ${be_pod_name} -n ${namespace}
+kubectl logs ${be_pod_name} -c compute -n ${namespace}
+```
+
+恢复步骤:
+
+1. 按 [配置认证](../separating-storage-compute/config-cluster.md) 修正 Secret 及
`spec.authSecret` 引用。Secret 类型为 `kubernetes.io/basic-auth`,包含 `username` 和
`password`。如果 root 已设置密码,使用已有的有效账户,或由管理员预先创建并授权管理账户;更新 Secret 不会修改数据库密码。
+2. 从能够访问 FE 的环境使用同一账户连接 FE,验证密码及 `SHOW BACKENDS`
等所需管理操作。以下命令通过提示输入密码,`${fe_query_port}` 为 FE 查询端口(默认 9030):
+
+ ```shell
+ mysql -h ${fe_host} -P ${fe_query_port} -u ${management_user} -p
+ ```
+
+3. 如果 BE 仍因旧凭据注册失败,逐个重建这些 Pod,使启动脚本重新读取 Secret。`${be_pod_name}` 必须是已确认注册失败的目标
Pod,不要批量重启正常运行的计算组:
+
+ ```shell
+ kubectl delete pod ${be_pod_name} -n ${namespace}
+ ```
+
+4. 等待 StatefulSet 重建该 Pod、Pod 恢复 Ready,并在 FE 执行以下 SQL,确认对应节点已加入且 `Alive` 为
`true`,再处理下一个失败 Pod:
+
+ ```sql
+ SHOW BACKENDS;
+ ```
+
### ComputeGroup 缩容未完成
在存算分离集群中,ComputeGroup 缩容可能长时间停留在 `Decommissioning` 或 `ScaleDownFailed`。
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-cluster.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-cluster.md
index 62cef1c83d5..610d13ee789 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-cluster.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-cluster.md
@@ -18,10 +18,16 @@
### 为什么需要配置管理凭证
-Doris 节点的管理需要通过用户名、密码以 MySQL 协议连接活着的 FE 节点进行操作。Doris 实现[类似 RBAC
的权限管理机制](../../../admin-manual/auth/security-overview),节点的管理需要用户拥有
[Node_priv](../../../admin-manual/auth/authorization/internal#所有权限) 权限。
+Doris 节点的管理需要通过用户名、密码以 MySQL 协议连接活着的 FE 节点进行操作。Doris 实现[类似 RBAC
的权限管理机制](../../../admin-manual/auth/security-overview),节点的管理需要用户拥有
[Node_priv](../../../admin-manual/auth/authorization/internal#所有权限项) 权限。
Doris Operator 默认使用拥有所有权限的 root 用户无密码模式对 DorisDisaggregatedCluster
资源配置的集群进行部署和管理。当 root 用户添加密码后,需要在 DorisDisaggregatedCluster 资源中显式配置拥有 Node_Priv
权限的用户名和密码,以便 Doris Operator 对集群进行自动化管理操作。
+:::caution root 设置密码后的管理凭据
+root 设置密码后,必须向 Operator 和节点启动脚本提供有效的管理凭据,推荐使用 `spec.authSecret`。对于通过 FE SQL
注册的 BE,凭据缺失或不匹配可能导致节点无法加入集群。Secret 必须与 `DorisDisaggregatedCluster`
位于同一命名空间,并包含与数据库有效管理账户一致的 `username` 和 `password`。修改 Secret
不会修改已有数据库账户的密码;数据库改密时也需要同步更新 Secret。
+
+root 已有密码时,如需使用专用管理账户,应由管理员预先创建并授权,不要依赖启动脚本自动创建。
+:::
+
### 三种密码配置场景对比
根据集群所处阶段和管理需求的不同,存在三种配置场景,可参考下表选择:
@@ -138,9 +144,15 @@ Doris Operator 提供使用 [Basic Authentication
Secret](https://kubernetes.io/
按照如下格式配置需要使用的 Basic Authentication Secret:
```yaml
+apiVersion: v1
+kind: Secret
+metadata:
+ name: ${secretName}
+ namespace: ${namespace}
+type: kubernetes.io/basic-auth
stringData:
- username: root
- password: ${password}
+ username: root
+ password: "${password}"
```
其中,`${password}` 为 root 设置的非加密密码。
@@ -196,9 +208,15 @@ spec:
按照如下格式配置需要使用的 Basic Authentication Secret:
```yaml
+apiVersion: v1
+kind: Secret
+metadata:
+ name: ${secretName}
+ namespace: ${namespace}
+type: kubernetes.io/basic-auth
stringData:
- username: ${DB_ADMIN_USER}
- password: ${DB_ADMIN_PASSWD}
+ username: ${DB_ADMIN_USER}
+ password: "${DB_ADMIN_PASSWD}"
```
| 参数 | 说明 |
@@ -234,7 +252,7 @@ spec:
## 场景三:集群部署后设置 root 用户密码
-Doris 集群在部署后若未设置 root 用户的密码,需要配置一个具有
[Node_priv](../../../admin-manual/auth/authorization/internal#所有权限) 权限的用户,便于
Doris Operator 自动化的管理集群节点。建议不要使用 root
用户,请参考[用户新建和权限赋值章节](../../../sql-manual/sql-statements/account-management/CREATE-USER)来创建新用户并赋予
Node_priv 权限。创建用户后,再通过环境变量或者 Secret 配置新的管理用户和密码,并在 DorisDisaggregatedCluster
资源中配置。
+Doris 集群在部署后若未设置 root 用户的密码,需要配置一个具有
[Node_priv](../../../admin-manual/auth/authorization/internal#所有权限项) 权限的用户,便于
Doris Operator 自动化的管理集群节点。建议不要使用 root
用户,请参考[用户新建和权限赋值章节](../../../sql-manual/sql-statements/account-management/CREATE-USER)来创建新用户并赋予
Node_priv 权限。创建用户后,再通过环境变量或者 Secret 配置新的管理用户和密码,并在 DorisDisaggregatedCluster
资源中配置。
配置流程概览:
@@ -294,9 +312,15 @@ spec:
按照如下格式创建 Basic Authentication Secret:
```yaml
+apiVersion: v1
+kind: Secret
+metadata:
+ name: ${secretName}
+ namespace: ${namespace}
+type: kubernetes.io/basic-auth
stringData:
- username: ${DB_ADMIN_USER}
- password: ${DB_ADMIN_PASSWD}
+ username: ${DB_ADMIN_USER}
+ password: "${DB_ADMIN_PASSWD}"
```
| 参数 | 说明 |
@@ -327,7 +351,7 @@ spec:
其中,`${secretName}` 为部署的 Basic Authentication Secret 的名称。
:::tip 提示
-部署后设置 root 密码,并配置新的拥有管理节点的用户名和密码后,会引起存量服务滚动重启一次。
+修改 CR 中的 `spec.authSecret` 引用或 `spec.adminUser` 会改变相关 Pod
模板,可能触发滚动更新;仅修改数据库密码或同名 Secret 的内容,不等于触发 Pod 滚动重启。通过 FE SQL 注册的 BE 启动脚本在启动时读取
Secret 到变量,注册重试期间不会重新读取。修正凭据并验证能够登录 FE 后,如果 BE 仍因旧凭据注册失败,应逐个重建这些注册失败的 BE
Pod,再通过 `SHOW BACKENDS` 确认节点已加入且 `Alive` 为 `true`。具体排查步骤见 [authSecret
配置问题](../doris-operator/status-and-troubleshooting.md#authsecret-配置问题)。
:::
## 挂载 Kerberos 认证文件
@@ -395,4 +419,4 @@ spec:
### 第 4 步:在 Hive Catalog 中使用 Kerberos 认证
-Kerberos 配置完成后,你可以在创建 Hive Catalog 时启用 Kerberos 认证。具体配置请参考 [Hive Catalog
配置文档](../../../lakehouse/catalogs/hive-catalog#配置-catalog)。
+Kerberos 配置完成后,你可以在创建 Hive Catalog 时启用 Kerberos 认证。具体配置请参考 [Hive Catalog
配置文档](../../../lakehouse/catalogs/hive-catalog.mdx#配置-catalog)。
diff --git
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-ms.md
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-ms.md
index a8ebab14952..d0fac147f26 100644
---
a/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-ms.md
+++
b/i18n/zh-CN/docusaurus-plugin-content-docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-ms.md
@@ -181,11 +181,18 @@ spec:
### 自定义管理 HTTP token {#ms-http-token}
-`http_token` 用于校验 MetaService 管理 HTTP API 的请求,默认值为 `greedisgood9999`。请将示例中的
`<custom-http-token>` 替换为自己的值,并保留 `doris_cloud.conf` 中的其他配置。将文件放在 ConfigMap 的
`doris_cloud.conf` 键下,再按上方步骤通过 `spec.metaService.configMaps` 引用该 ConfigMap。
+`http_token` 用于 MetaService 管理 HTTP API 的请求鉴权,默认值为 `greedisgood9999`。请将示例中的
`<custom-http-token>` 替换为自定义随机字符串。将 `doris_cloud.conf` 放在 ConfigMap
的同名键下,再按上方步骤通过 `spec.metaService.configMaps` 引用该 ConfigMap。
-所有 MetaService 副本应使用相同 token,并重启以加载启动配置。API 客户端通过 `token` 查询参数传入该值。Doris
Operator 也会从引用的 MS ConfigMap 中读取 `http_token`,用于管理 HTTP 请求;只修改 Pod 内的文件会导致
Operator 与 MS 配置不一致。
+API 客户端通过 `token` 查询参数传入该值。Doris Operator 从引用的 ConfigMap 读取 `http_token`,用于管理
HTTP 请求,因此请更新 ConfigMap,使 Operator 与 MetaService 的配置保持一致。FE、BE 的元数据 RPC 不使用此
token。对于通过 FE SQL 注册节点的镜像(例如 Doris 4.1.5),修改此 token 无需修改或重启 FE、BE。[Doris 管理账户凭据
`spec.authSecret`](config-cluster.md) 与 `http_token` 相互独立。
-FE、BE 的元数据 RPC 不需要在其配置中设置这个管理 HTTP token。`spec.authSecret` 中的 SQL 管理凭据与
`http_token` 相互独立。
+部署后可按以下步骤轮换 token:
+
+1. 更新引用的 ConfigMap 中 `doris_cloud.conf` 的 `http_token`,保留其他配置项。所有 MetaService
副本使用相同的新 token。
+2. 逐个重启 MetaService Pod,等待当前 Pod 恢复 Ready 后再处理下一个。单副本部署在重启期间会暂时无法提供
MetaService 服务。
+3. 使用需要 token 鉴权的只读管理 HTTP API,逐个验证各 MetaService 副本的新 token 已生效。仅检查 Pod Ready
不能确认 token 已切换。
+4. 同步更新调用这些 API 的运维工具。
+
+MetaService 在启动时加载 `http_token`。仅更新 ConfigMap 不会修改运行中进程的 token,也不能保证自动重启
Pod。轮换期间,各副本可能暂时使用不同的 token,管理 HTTP API 请求可能认证失败。请在此期间避开依赖这些 API 的节点管理操作。
## 5. 配置服务探测超时
diff --git
a/versioned_docs/version-4.x/install/deploy-on-kubernetes/doris-operator/status-and-troubleshooting.md
b/versioned_docs/version-4.x/install/deploy-on-kubernetes/doris-operator/status-and-troubleshooting.md
index e388c3caef6..b0ef2a50c09 100644
---
a/versioned_docs/version-4.x/install/deploy-on-kubernetes/doris-operator/status-and-troubleshooting.md
+++
b/versioned_docs/version-4.x/install/deploy-on-kubernetes/doris-operator/status-and-troubleshooting.md
@@ -202,6 +202,45 @@ kubectl logs deployment/${operator_deployment_name} -n
${operator_namespace}
If the cluster uses a non-default management account or has already set a root
password, verify that `authSecret` matches the actual Doris credentials.
+#### BE registration failures caused by credentials in a decoupled cluster
+
+The following steps apply to BE images that use `authSecret` to log in to FE
and register nodes through SQL, such as Doris 4.1.5.
+
+| Symptom | What to check |
+|---------|---------------|
+| Pod events show `FailedMount` and report that the Secret does not exist |
Check the Secret name and namespace; the Secret must be in the same namespace
as the cluster |
+| BE registration logs repeatedly show `Access denied` / `1045` and the node
does not join | Check that the Secret's `username` and `password` match the
database account; the script first tries root without a password, so a single
error does not necessarily mean the subsequent attempt with configured
credentials failed |
+| The management account can log in to FE, but node management SQL is rejected
| Check that the account has the required node management privileges |
+| The Secret is corrected, but BE still fails registration with old
credentials | The startup script has already read the credentials into
variables and does not reread the Secret during registration retries |
+
+First inspect the target BE Pod's events and registration logs:
+
+```shell
+kubectl describe pod ${be_pod_name} -n ${namespace}
+kubectl logs ${be_pod_name} -c compute -n ${namespace}
+```
+
+Recovery steps:
+
+1. Correct the Secret and the `spec.authSecret` reference as described in
[Configuring Authentication](../separating-storage-compute/config-cluster.md).
Use a Secret of type `kubernetes.io/basic-auth` with `username` and `password`.
If root already has a password, use an existing valid account or have an
administrator create and grant privileges to a management account first.
Updating the Secret does not change the database password.
+2. From an environment that can reach FE, connect with the same account and
verify the password and required management operations such as `SHOW BACKENDS`.
The following command prompts for the password. `${fe_query_port}` is the FE
query port (9030 by default):
+
+ ```shell
+ mysql -h ${fe_host} -P ${fe_query_port} -u ${management_user} -p
+ ```
+
+3. If BE still fails registration because of old credentials, recreate those
Pods one at a time so that the startup script rereads the Secret.
`${be_pod_name}` must identify a Pod confirmed to have failed registration; do
not restart the entire healthy compute group:
+
+ ```shell
+ kubectl delete pod ${be_pod_name} -n ${namespace}
+ ```
+
+4. Wait for the StatefulSet to recreate the Pod and for it to become Ready.
Run the following SQL on FE to confirm that the corresponding node has joined
and its `Alive` field is `true` before proceeding to the next failed Pod:
+
+ ```sql
+ SHOW BACKENDS;
+ ```
+
### ComputeGroup scale-in does not finish
In compute-storage decoupled clusters, ComputeGroup scale-in may stay in
`Decommissioning` or `ScaleDownFailed`.
diff --git
a/versioned_docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-cluster.md
b/versioned_docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-cluster.md
index f8c6a9ed18c..d7309cacaf3 100644
---
a/versioned_docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-cluster.md
+++
b/versioned_docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-cluster.md
@@ -22,6 +22,12 @@ Managing Doris nodes requires connecting to a live FE node
over the MySQL protoc
By default, Doris Operator uses the root user with all privileges and no
password to deploy and manage the cluster configured by the
DorisDisaggregatedCluster resource. After a password is added to the root user,
you must explicitly configure a username and password with the Node_priv
privilege in the DorisDisaggregatedCluster resource so that Doris Operator can
perform automated management operations on the cluster.
+:::caution Management credentials after setting a root password
+Once root has a password, provide valid management credentials to Operator and
the node startup scripts, preferably through `spec.authSecret`. For BE nodes
that register through FE SQL, missing or mismatched credentials can prevent
them from joining the cluster. The Secret must be in the same namespace as
`DorisDisaggregatedCluster` and contain a `username` and `password` that match
a valid management account in Doris. Updating the Secret does not change an
existing database account pass [...]
+
+If root already has a password, an administrator must create and grant
privileges to a dedicated management account before using it. Do not rely on
the startup scripts to create it automatically.
+:::
+
### Comparison of the three password configuration scenarios
Depending on the cluster's stage and management requirements, there are three
configuration scenarios. Refer to the table below to choose:
@@ -138,9 +144,15 @@ Doris Operator supports using a [Basic Authentication
Secret](https://kubernetes
Configure the Basic Authentication Secret to be used in the following format:
```yaml
+apiVersion: v1
+kind: Secret
+metadata:
+ name: ${secretName}
+ namespace: ${namespace}
+type: kubernetes.io/basic-auth
stringData:
- username: root
- password: ${password}
+ username: root
+ password: "${password}"
```
Here, `${password}` is the unencrypted password set for root.
@@ -196,9 +208,15 @@ spec:
Configure the Basic Authentication Secret to be used in the following format:
```yaml
+apiVersion: v1
+kind: Secret
+metadata:
+ name: ${secretName}
+ namespace: ${namespace}
+type: kubernetes.io/basic-auth
stringData:
- username: ${DB_ADMIN_USER}
- password: ${DB_ADMIN_PASSWD}
+ username: ${DB_ADMIN_USER}
+ password: "${DB_ADMIN_PASSWD}"
```
| Parameter | Description |
@@ -294,9 +312,15 @@ spec:
Create a Basic Authentication Secret in the following format:
```yaml
+apiVersion: v1
+kind: Secret
+metadata:
+ name: ${secretName}
+ namespace: ${namespace}
+type: kubernetes.io/basic-auth
stringData:
- username: ${DB_ADMIN_USER}
- password: ${DB_ADMIN_PASSWD}
+ username: ${DB_ADMIN_USER}
+ password: "${DB_ADMIN_PASSWD}"
```
| Parameter | Description |
@@ -327,7 +351,7 @@ spec:
Here, `${secretName}` is the name of the deployed Basic Authentication Secret.
:::tip Tip
-After deployment, setting the root password and configuring the new username
and password with node management privileges will trigger one rolling restart
of the existing services.
+Changing the `spec.authSecret` reference or `spec.adminUser` in the CR changes
the affected Pod templates and may trigger a rolling update. Changing only the
database password or the contents of a Secret with the same name does not imply
a rolling restart. BE startup scripts that register through FE SQL read the
Secret into variables at startup and do not reread it during registration
retries. After correcting the credentials and verifying that they can log in to
FE, recreate BE Pods tha [...]
:::
## Mounting Kerberos authentication files
@@ -395,4 +419,4 @@ spec:
### Step 4: Use Kerberos authentication in Hive Catalog
-Once Kerberos is configured, you can enable Kerberos authentication when
creating a Hive Catalog. For specific configuration, refer to the [Hive Catalog
configuration
documentation](../../../lakehouse/catalogs/hive-catalog#配置-catalog).
+Once Kerberos is configured, you can enable Kerberos authentication when
creating a Hive Catalog. For specific configuration, refer to the [Hive Catalog
configuration
documentation](../../../lakehouse/catalogs/hive-catalog.mdx#configuring-catalog).
diff --git
a/versioned_docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-ms.md
b/versioned_docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-ms.md
index 81d148a282a..3344d2d218b 100644
---
a/versioned_docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-ms.md
+++
b/versioned_docs/version-4.x/install/deploy-on-kubernetes/separating-storage-compute/config-ms.md
@@ -181,11 +181,18 @@ Apply the above configuration to the
[DorisDisaggregatedCluster resource](./inst
### Customize the management HTTP token {#ms-http-token}
-`http_token` authenticates MetaService management HTTP API requests. Its
default value is `greedisgood9999`. Replace `<custom-http-token>` in the
example with your own value and keep the rest of your `doris_cloud.conf`
settings. Store the file under the ConfigMap key `doris_cloud.conf` and
reference that ConfigMap through `spec.metaService.configMaps`, as shown above.
+`http_token` authenticates MetaService management HTTP API requests. Its
default value is `greedisgood9999`. Replace `<custom-http-token>` in the
example with a custom random string. Store `doris_cloud.conf` under the
ConfigMap key of the same name and reference that ConfigMap through
`spec.metaService.configMaps`, as shown above.
-Apply the same token to all MetaService replicas and restart them to load the
startup configuration. API clients must pass that value in the `token` query
parameter. Doris Operator also reads `http_token` from the referenced MS
ConfigMap for its management HTTP requests; changing only a file inside a Pod
leaves the Operator's configuration out of sync.
+API clients pass the token in the `token` query parameter. Doris Operator
reads `http_token` from the referenced ConfigMap for management HTTP requests,
so update the ConfigMap to keep the Operator and MetaService in sync. FE and BE
metadata RPCs do not use this token. For images that register nodes through FE
SQL, such as Doris 4.1.5, changing this token requires no changes or restarts
in FE or BE. The [Doris management account credentials in
`spec.authSecret`](config-cluster.md) are in [...]
-FE and BE metadata RPCs do not require this management HTTP token in their
configuration. The SQL management credentials in `spec.authSecret` are separate
from `http_token`.
+To rotate the token after deployment:
+
+1. Update `http_token` in `doris_cloud.conf` in the referenced ConfigMap,
preserving the other settings. Use the same new token for all MetaService
replicas.
+2. Restart MetaService Pods one at a time and wait for each Pod to become
Ready before proceeding to the next. A single-replica deployment temporarily
loses MetaService availability during the restart.
+3. Verify the new token on each MetaService replica using a read-only
management HTTP API that requires token authentication. Pod readiness alone
does not confirm that the token has changed.
+4. Update the operations tools that call these APIs.
+
+MetaService loads `http_token` at startup. An update to the ConfigMap alone
does not change the token in a running process or guarantee an automatic
restart. During rotation, replicas may temporarily use different tokens and
management HTTP requests may fail authentication. Avoid node management
operations that depend on these APIs during this interval.
## 5. Configure service probe timeouts
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]