diff --git a/docs/infrastructure/intranet/index.md b/docs/infrastructure/intranet/index.md
index 18c74f22..d2ddf392 100644
--- a/docs/infrastructure/intranet/index.md
+++ b/docs/infrastructure/intranet/index.md
@@ -1,40 +1,37 @@
 # Servers Intranet
 
-Servers Intranet connect all the servers together, including physics servers and virtual machines.
+Servers Intranet connects all the servers together, including physical servers and virtual machines.
 
 ## Network Topology
 
 <iframe frameborder="0" style="width:100%;height:500px;" src="https://viewer.diagrams.net/?lightbox=1&highlight=0000ff&edit=_blank&layers=1&nav=1&title=LUG%20Network#Uhttps%3A%2F%2Fdocs.ustclug.org%2Finfrastructure%2Fintranet%2Fimg%2Fnetwork.html"></iframe>
 
-以上架构图由 iBug 在 2023 年 11 月更新。以下信息是过时的，不过有参考价值。
+以上架构图由 iBug 在 2023 年 11 月更新。
 
-The network contains three parts:
+??? warning "以下是一些过时的信息，也许还有点参考价值"
 
-- Physical Switch in East Library Data Center
-- Virtual Switch on vSphere host machine
-- tincVPN
+    The network contains three parts:
 
-tincVPN is a mesh VPN, which can be abstracted as a virtual Switch.
+    - Physical Switch in East Library Data Center
+    - Virtual Switch on vSphere host machine
+    - tincVPN
 
-vm-nfs.s.ustclug.org runs a layer 2 bridge, connecting tincVPN and SRW2024(physical switch).
+    tincVPN is a mesh VPN, which can be abstracted as a virtual Switch.
 
-It is obvious that vm-nfs is a single point of failure of communicating between tinc host and vSphere virtual machine. I had tried to add another bridge node, but resulted in a broadcast storm. Maybe we can fix it by MPLS (merged in mainland kernel 4.3). But it isn't a right timing at this time.
+    vm-nfs.s.ustclug.org runs a layer 2 bridge, connecting tincVPN and SRW2024 (physical switch).
 
-## Network information
-
-The network contains two subnets:
+    It is obvious that vm-nfs is a single point of failure of communicating between tinc host and vSphere virtual machine. I had tried to add another bridge node, but resulted in a broadcast storm. Maybe we can fix it by MPLS (merged in mainline kernel 4.3). But it isn't a right timing at this time.
 
-* 10.254.0.0/21
-* 10.254.10.0/24
+## Network information
 
-Every server binds one and only one IP address in 10.254.0.0/21, used to communicate with each other.
+The network contains one single subnet: 10.254.0.0/21
 
-10.254.10.0/24 is used for 1to1 IP mapping. At this time, it just used between linode(10.254.10.2) and blog(10.254.10.1).
+Every server and service binds to one and only one IP address, used to communicate with each other.
 
 ### Address planning
 
-* 10.254.0.0/24: physical server and virtual machine
-* 10.254.1.0/24: docker container
-* 10.254.6.0/24: LUGi emergency entrypoint (managed by yzf)
-* 10.254.7.0/24: LUGi entrypoint (via board.s)
-* other address: not used yet.
+- 10.254.0.0/24: Physical servers and virtual machines
+- 10.254.1.0/24: Docker containers
+- 10.254.6.0/24: LUGi emergency entrypoint (via vpnstv.s, managed by yzf)
+- 10.254.7.0/24: LUGi entrypoint (via board.s)
+- Others: not used yet.