This is an automated email from the ASF dual-hosted git repository.
wangdan pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-pegasus-website.git
The following commit(s) were added to refs/heads/master by this push:
new 39ef3343 update administration doc (#119)
39ef3343 is described below
commit 39ef334308db0ac5da8449fa69d878054d907e83
Author: Samunroyu <[email protected]>
AuthorDate: Mon Nov 3 15:37:23 2025 +0800
update administration doc (#119)
---
_docs/en/administration/administration.md | 35 +++++++++++++++++++++++++++++++
_docs/en/administration/index.md | 5 -----
_docs/zh/administration/administration.md | 8 +++----
3 files changed, 39 insertions(+), 9 deletions(-)
diff --git a/_docs/en/administration/administration.md
b/_docs/en/administration/administration.md
new file mode 100644
index 00000000..060c2d9a
--- /dev/null
+++ b/_docs/en/administration/administration.md
@@ -0,0 +1,35 @@
+---
+permalink: administration/index.html
+---
+
+Pegasus not only provides simple key-value storage interfaces; based on
stability considerations, we have added many features that help efficient
operations and maintenance.
+
+If you are new to Pegasus and want to deploy the compiled system on your
machines, you should first refer to [Configuration Management](config) to
adjust parameters according to your environment. Then refer to [Cluster
Deployment](deployment) to deploy Pegasus on your machine cluster.
+
+After a successful deployment, you should focus on observing the system’s
runtime status. Refer to [Monitoring](monitoring) to view the relevant metrics.
+
+During service operation, occasional anomalies (possibly due to network
issues, disk failures, system errors, or potential bugs) may cause cluster
fluctuations. If this results in uneven node load, you can perform operations
according to [Load Balancing](rebalance).
+
+If business volume grows beyond what your service resources can handle, you
can increase the number of nodes by following [Scale In/Out](scale-in-out). If
a machine experiences persistent failure, you can also remove the bad node via
[Scale In/Out](scale-in-out). If a specific SSD disk on a machine fails, refer
to [Bad Disk Maintenance](bad-disk) to remove the faulty disk.
+
+If you need to restart or upgrade the cluster, please refer to [Rolling
Update](rolling-update).
+
+During cluster operation, you must continuously monitor resource usage (disk,
memory, network) and make timely operational adjustments. Please refer to
[Resource Management](resource-management).
+
+If the business data volume is large or individual values are long, refer to
[Compression](compression) to improve disk utilization and boost read/write
performance.
+
+For some critical businesses, you can regularly perform [Cold
Backup](cold-backup) on related tables to ensure data safety. The cold backup
feature is also often used for fast large-scale data migration.
+
+In extreme situations, such as metadata loss or multiple nodes failing
simultaneously, the cluster may become inconsistent. If metadata is lost, refer
to [Meta Recovery](meta-recovery); if inconsistencies are caused by backup
loss, refer to [Replica Recovery](replica-recovery).
+
+If the cluster needs to rely on a new ZooKeeper and migrate ZooKeeper data,
please refer to [ZooKeeper Migration](zk-migration).
+
+Typically, a cluster serves many tables, and all tables have the same
priority. In other words, if a certain table’s request volume becomes
excessive, the service quality of other tables will decline. This is commonly
called a “multi-tenancy” problem. You can use [Throttling](throttling) to meet
each table’s SLA.
+
+If you need to migrate a table’s data to another table, please refer to [Table
Migration](table-migration).
+
+Table-level operational tasks also include [Table Soft
Delete](table-soft-delete), [Table Environment Variables](table-env),
[Partition Split](partition-split), [Manual Compaction](manual-compact), and
[Usage Scenario](usage-scenario).
+
+In addition, we support [Remote Commands](remote-commands) for administrators
to directly control servers, and provide an [HTTP API](http).
+
+From our experience operating cluster services for various businesses, we have
accumulated many [Operational Experiences](experiences) for your reference.
diff --git a/_docs/en/administration/index.md b/_docs/en/administration/index.md
deleted file mode 100644
index d30b1c12..00000000
--- a/_docs/en/administration/index.md
+++ /dev/null
@@ -1,5 +0,0 @@
----
-menubar: administration_menu
----
-
-TRANSLATING
diff --git a/_docs/zh/administration/administration.md
b/_docs/zh/administration/administration.md
index e47bc648..9cfe2e25 100644
--- a/_docs/zh/administration/administration.md
+++ b/_docs/zh/administration/administration.md
@@ -13,20 +13,20 @@ Pegasus 不仅仅只提供简单的 key value 存储接口，我们还基于稳
如果是随着业务量增大，超出了你的服务资源所能承受的能力，你可以参照 [集群扩容缩容](scale-in-out) 来增加节点数。
如果有机器发生持久性的故障，你也可以参照 [集群扩容缩容](scale-in-out) 剔除这个坏节点。
-如果是机器的某个SSD盘出故障，可以参照 [坏盘检修](bad-disk) 剔除这个坏盘。
+如果是机器的某个 SSD 盘出故障，可以参照 [坏盘检修](bad-disk) 剔除这个坏盘。
如果需要重启或升级集群，请参照 [集群重启](rolling-update)。
集群运行过程中，你需要时刻关注资源（磁盘、内存、网络）的使用情况，并及时做出运维调整，请参照 [资源管理](resource-management)。
-如果业务数据量很大或者单条value很长，你可以参照 [数据压缩](compression) 来提高磁盘利用率和提升读写性能。
+如果业务数据量很大或者单条 value 很长，你可以参照 [数据压缩](compression) 来提高磁盘利用率和提升读写性能。
对一些重要的业务，你可以将相关的表定期进行 [冷备份](cold-backup) 来保证数据安全性。冷备份的功能也常常用作于快速的大量数据迁移。
对于一些极端情况，譬如元数据丢失、多个节点同时宕机，可能会造成集群数据的不一致。
如果元数据丢失，建议你参考 [元数据恢复](meta-recovery)；对于数据备份丢失造成的不一致，建议你参考
[Replica数据恢复](replica-recovery)。
-如果集群要依赖一个新的Zookeeper，需要迁移Zookeeper数据，请参考 [Zookeeper迁移](zk-migration)。
+如果集群要依赖一个新的 Zookeeper，需要迁移 Zookeeper 数据，请参考 [Zookeeper迁移](zk-migration)。
通常一个集群内会服务于许多表，所有的表都有相同的优先级，换言之，有某个表的请求量过大，
其他表的服务质量就会下滑，这通常被称为 “多租户” 问题，你可以通过 [流量控制](throttling) 满足每个表的 SLA。
@@ -35,6 +35,6 @@ Pegasus 不仅仅只提供简单的 key value 存储接口，我们还基于稳
表级的运维操作还包括 [Table软删除](table-soft-delete) 、[Table环境变量](table-env)、[Partition
Split](partition-split)、[Manual Compact功能](manual-compact)、[Usage
Scenario功能](usage-scenario) 等。
-除此之外，我们还支持了 [远程命令](remote-commands)，方便运维人员对Server进行直接控制，并提供了 [HTTP接口](http)。
+除此之外，我们还支持了 [远程命令](remote-commands)，方便运维人员对 Server 进行直接控制，并提供了 [HTTP接口](http)。
我们在运维集群服务业务的过程中，也积累了很多 [运维经验](experiences)，你可以参考。
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
