Use aVCS for automatic configuration Synchronization

Use two AX servers and upgrade to version 261. The latest aVCS feature of AX is used for automatic configuration synchronization. AVCS can be deployed on traditional HA, and can also be used with the VRRP-A to achieve automatic synchronization of Multi-virtual partition configuration (ADP with L2/L3virtualization ).
Basically divided into two steps, the first step to configure the VRRP-A, the second part to configure aVCS.
 
(1) VRRP-A
The configuration of the VRRP-A is quite simple, which is more advantageous than the traditional HA mode, the following is the vrrp-a configuration of the two machines in the lab:
AX1000-1-Active (config) # sh vrrp-a config
Vrrp-a device-id 1
Vrrp-a set-id 1
Vrrp-a enable
Vrrp-a vrid default
Floating-ip 192.168.10.3
Priority 100 device 1
Priority 50 device 2
!
Vrrp-a vrid 1
Priority 100 device 1
Priority 50 device 2
 
AX1000-2-Standby (config) # sh vrrp-a config
Vrrp-a device-id 2
Vrrp-a set-id 1
Vrrp-a enable
Vrrp-a vrid default
Floating-ip 192.168.10.3
Priority 50 device 1
Priority 100 device 2
!
Vrrp-a vrid 1
Priority 50 device 1
Priority 100 device 2
VRRP-A also has a variety of deployment modes options: Active-Standby (n + 1), Active-Active (n + 1)
Active-Standby (n + 1) is actually a one-master-multiple-slave mode, which breaks through the traditional HA's one-master-one-slave quantity limit;
 

The Active-Active (n + 1) mode is more flexible.
 

Now the configuration is described line by line:
1) device id is the unique identifier of each device.
2) set-id indicates that two or more AX instances belong to the same Active/Standby or Active/Active group. Unlike traditional HA instances, set-id is required. Remove the set-id in the test, and both AX instances are immediately declared active.
3) vrrp-a enable is to activate the VRRP-A Function
4) vrid (virtual router ID): The vrid required for VRP groups.
If vrid default is used (id is 0), all IP resources are automatically associated with this default vrid. With default vrid, ax automatically finds any connected port and sends a heartbeat packet to the Peer to detect the existence of this port.
For example, in the test, I set two VLANs to be interconnected with each other. No matter the cables in any vlan are unplugged, ax will automatically communicate with the peer end from the other connected port.
AX1000-1-Active (config) # sh vrrp-a de
Vrid default
Unit State Priority
1 (Local) Active 100
2 (Peer) Standby 50 *
Vrid that is running: default
VRRP stats
Peer: 2, vrid default
Port 1: received 106 missed 0
Port 6: received 8230 missed 1
Heartbeat missed: 2
......
Peer IP [2]: 192.168.10.2
 
AX1000-1-Active (config) # sh vrrp-a de
Vrid default
Unit State Priority
1 (Local) Active 100
2 (Peer) Standby 50 *
Vrid that is running: default
VRRP stats
Peer: 2, vrid default
Port 6: received 7897 missed 1
Heartbeat missed: 1
Total packets received ed from peer: 7897
......
Peer IP [2]: 1.1.1.2
5) The Floating ip address is the address that is provided to the upper-end router device or the lower-end server to specify AX as the next route address;
6) The higher the Priority value, the higher the Priority (150 by default)
 
(1) AVCS
AVCS is the new virtualization and High Availability Technology of A10, which is mainly used with VRRP-A to achieve automatic configuration synchronization.
AVCS configuration is also relatively simple, in the configuration of the VRRP-A and the basic configuration of the system (host name, vlan, interface, ve, route), and then configure aVCS as follows:
 
1) vcs enable activates the vcs Function
2) vcs chassis-id 1 is not configured. The default value is 1.
3) vcs device 1 sets the device id of the local machine.
4) interfaces ethernet 6 Configuration Management Interface
5) Ip 1.1.1.1 set the management address of the Local Machine
6) Floating-interface 1.1.1.3 set the shared management address of the vcs Group
7) Enable activates the configuration of the above vcs device 1
8) vcs reload restarts the vcs process for the vcs policy to take effect.

 

After reload, the configurations of the two servers are synchronized immediately. You can also see the configuration of the Peer device in the local unit:
Vcs enable
Vcs config-info 3a55bcb734003e3f 124
Vcs chassis-id 1
Vcs floating-ip 1.1.1.3/24
Vcs multicast-ip 224.0.0.210
Vcs device 1
Priority100
Interfaces management
Interfaces ethernet 6
Enable
Vcs device 2
Priority 50
Interfaces management
Interfaces ethernet 6
Enable
Vcs local-device 1
!
Hostname AX1000-1 device 1
Hostname AX1000-2 device 2
 
Technical details
1) determine the vMaster
Although the aVCS group also has a priority setting, the vMaster has its selection method, even if the priority values of each member are the same. The calculation sequence is as follows:
A) dynamic priority-dynamic priority. When A device is selected as A vmaster from the beginning, even if A new device with A higher priority is added to the avcs group, it will continue to be set as A vmaster, to ensure system (configuration and running status) Stability
B) Priority-select vmaster with the highest Priority
C) The lower the value of Device ID-Device ID, select vmaster.
 
2) Avcs configuration points
Avcs uses multicast to perceive each other in the same broadcast domain and each member. Therefore, before subordinates, it is necessary to determine that all AX instances must be deployed in the same L2 broadcast domain.
Avcs management interfaces cannot use ve, but can only use a single physical port ethernet port
The floating ip address of Avcs is mainly used to manage the avcs group, which is taken over by the vmaster. Of course, we can log on to the management address of each AX separately, however, you will find that only the vmaster ax can enter the configuration mode.
 
Summary: VRRP-A and aVCS meet the high availability requirements of the new generation, both convenient expansion, but also bring great convenience to system maintenance personnel.
This article is from the "ADC technology blog ".