Updated at: 2022-12-09 03:49:50
Prepare three servers with available networks. Upload installation package into these three servers under /backupsoft and unpack the packages.
§ For the first node, that is master node, run command:
./install.sh --product=product type --server-type=MASTER --self-ip= xxx.xxx.xxx（ipv4） --cluster-id=108 --cluster-vip= xxx.xxx.xxx（ipv4） --install-db=yes --db-id=109 --db-vip=xxx.xxx.xxx（ipv4） --ntp-server= xxx.xxx.xxx（same as cluster-vip） --ssl=no
Take AnyBackup Enterprise and intranet IPV4 address as an example,
./install.sh --product=Enterprise --server-type=MASTER --self-ip=192.168.10.58 --cluster-id=253 --cluster-vip=192.168.10.253 --install-db=yes --db-id=254 --db-vip=192.168.10.254 --ntp-server=192.168.10.253 --ssl=no
§ For the second node, that is slave node, run command:
./install.sh --product=product type --server-type=SLAVE --self-ip= xxx.xxx.xxx（ipv4） --cluster-id=108 --cluster-vip= xxx.xxx.xxx（ipv4） --install-db=yes --db-id=109 --db-vip=xxx.xxx.xxx（ipv4） --ntp-server= xxx.xxx.xxx（same as cluster-vip） --ssl=no
Take AnyBackup Enterprise and intranet IPV4 address as an example,
./install.sh --product=Enterprise --server-type=SLAVE --self-ip=192.168.10.59 --cluster-id=253 --cluster-vip=192.168.10.253 --install-db=yes --db-id=254 --db-vip=192.168.10.254 --ntp-server=192.168.10.253 --ssl=no
§ For the third node, that is slave node as well, run command:
./install.sh --product=product type --server-type=SLAVE --self-ip= xxx.xxx.xxx（ipv4） --cluster-id=108 --cluster-vip= xxx.xxx.xxx（ipv4） --install-db=yes --db-id=109 --db-vip=xxx.xxx.xxx（ipv4） --ntp-server= xxx.xxx.xxx（same as cluster-vip） --ssl=no
Take AnyBackup Enterprise and intranet IPV4 address as an example,
./install.sh --product= Enterprise --server-type=SLAVE --self-ip=192.168.10.60 --cluster-id=253 --cluster-vip=192.168.10.253 --install-db=yes --db-id=254 --db-vip=192.168.10.254 --ntp-server=192.168.10.253 --ssl=no
Mind the following:
In a cluster environment, be sure to configure consul service after the installation.
To configure the consul, do the following:
A. Stop consul service using command systemctl stop ABConsulService.service.
B. Enter the folder /backupsoft/AnyBackupServer/ConsulService/conf by the command cd /backupsoft/AnyBackupServer/ConsulService/conf and ls to unfold it.
C. Open the file server.json by command vi /server.json or vim /server.json and assign a value (the number of nodes in the management domain) to the parameter “bootstrap_except”.
D. Start consul service using command
systemctl start ABConsulService.service.
Check whether the leader is elected by command curl http://127.0.0.1:9691/v1/status/leader.
[Node IP:9696] indicates that the leader is elected.
Check whether the node is added into the consul successfully using the command curl http://127.0.0.1:9691/v1/status/peers.
[Node IP:9696] indicates that the node has been added.
E. Repeat the configuration procedure A~D on nodes subsequently one after another and check whether leaders are elected and nodes are added into the consul.
If installation fails, do as follows:
During cluster installation, if some configurations are not appropriate or the environment is not properly checked, you may fail to install one of the three nodes. Before reinstallation, please uninstall other successfully installed nodes, remove the installation package and residual information, and unpack the software package again. If among node A, B and C, A and B have been installed successfully, and an error is reported because environment checkout is not performed for node C, you have to uninstall all nodes and reinstall them. As IP will be recorded in the database once the installation is started, even though installation of node C failed, its IP has been recorded. Reinstalling it directly will cause error.
Problems that may occur during installation:
The error “Failed to connect DB” may be reported in node installation. See the following figure.
Cause: DBVIP disappears because KAD service on MASTER node is restarted. When MASTER VIP appears again, the installation can be performed as normal.
Resolving: DBVIP will be scanned four times. If it does not reappear within the 4 scannings, installation of the SLAVE nodes will fail. This is probably due to improper installation of MASTER node and you need to check its VIP and DBIP. Run command ip addr on the MASTER server to see whether VIP and DBIP can be generated. Then ping MASTER VIP and DBIP from the SLAVE node. If MASTER VIP and DBIP can be pinged, the installation can be performed as normal.
§ For the first node, that is master node, run command:
./install.sh --product=product type --server-type=MASTER --self-ip= xxx.xxx.xxx（ipv4） --cluster-id=108 --cluster-vip= xxx.xxx.xxx（ipv4） --install-db=yes --db-id=109 --db-vip=xxx.xxx.xxx（ipv4） --ntp-server= xxx.xxx.xxx（same as cluster-vip） --ssl=no
Take AnyBackup Enterprise and intranet IPV4 address as an example,
./install.sh --product=Enterprise --server-type=MASTER --self-ip=192.168.10.58 --cluster-id=253 --cluster-vip=192.168.10.253 --install-db=yes --db-id=254 --db-vip=192.168.10.254 --ntp-server=192.168.10.253 --ssl=no
§ For the second node, that is slave node, run command:
./install.sh --product=product type --server-type=SLAVE --self-ip= xxx.xxx.xxx（ipv4） --cluster-id=108 --cluster-vip= xxx.xxx.xxx（ipv4） --install-db=yes --db-id=109 --db-vip=xxx.xxx.xxx（ipv4） --ntp-server= xxx.xxx.xxx（same as cluster-vip） --ssl=no
Take AnyBackup Enterprise and intranet IPV4 address as an example,
./install.sh --product=Enterprise --server-type=SLAVE --self-ip=192.168.10.59 --cluster-id=253 --cluster-vip=192.168.10.253 --install-db=yes --db-id=254 --db-vip=192.168.10.254 --ntp-server=192.168.10.253 --ssl=no
§ For the third node, that is slave node as well, run command:
./install.sh --product=product type --server-type=SLAVE --self-ip= xxx.xxx.xxx（ipv4） --cluster-id=108 --cluster-vip= xxx.xxx.xxx（ipv4） --install-db=yes --db-id=109 --db-vip=xxx.xxx.xxx（ipv4） --ntp-server= xxx.xxx.xxx（same as cluster-vip） --ssl=no
Take AnyBackup Enterprise and intranet IPV4 address as an example,
./install.sh --product= Enterprise --server-type=SLAVE --self-ip=192.168.10.60 --cluster-id=253 --cluster-vip=192.168.10.253 --install-db=yes --db-id=254 --db-vip=192.168.10.254 --ntp-server=192.168.10.253 --ssl=no
Mind the following:
In a cluster environment, be sure to configure consul service after the installation.
To configure the consul, do the following:
A. Stop consul service using command systemctl stop ABConsulService.service.
B. Enter the folder /backupsoft/AnyBackupServer/ConsulService/conf by the command cd /backupsoft/AnyBackupServer/ConsulService/conf and ls to unfold it.
C. Open the file server.json by command vi /server.json or vim /server.json and assign a value (the number of nodes in the management domain) to the parameter “bootstrap_except”.
D. Start consul service using command
systemctl start ABConsulService.service.
Check whether the leader is elected by command curl http://127.0.0.1:9691/v1/status/leader.
[Node IP:9696] indicates that the leader is elected.
Check whether the node is added into the consul successfully using the command curl http://127.0.0.1:9691/v1/status/peers.
[Node IP:9696] indicates that the node has been added.
E. Repeat the configuration procedure A~D on nodes subsequently one after another and check whether leaders are elected and nodes are added into the consul.
If installation fails, do as follows:
During cluster installation, if some configurations are not appropriate or the environment is not properly checked, you may fail to install one of the three nodes. Before reinstallation, please uninstall other successfully installed nodes, remove the installation package and residual information, and unpack the software package again. If among node A, B and C, A and B have been installed successfully, and an error is reported because environment checkout is not performed for node C, you have to uninstall all nodes and reinstall them. As IP will be recorded in the database once the installation is started, even though installation of node C failed, its IP has been recorded. Reinstalling it directly will cause error.
Problems that may occur during installation:
The error “Failed to connect DB” may be reported in node installation. See the following figure.
Cause: DBVIP disappears because KAD service on MASTER node is restarted. When MASTER VIP appears again, the installation can be performed as normal.
Resolving: DBVIP will be scanned four times. If it does not reappear within the 4 scannings, installation of the SLAVE nodes will fail. This is probably due to improper installation of MASTER node and you need to check its VIP and DBIP. Run command ip addr on the MASTER server to see whether VIP and DBIP can be generated. Then ping MASTER VIP and DBIP from the SLAVE node. If MASTER VIP and DBIP can be pinged, the installation can be performed as normal.
< Previous:
Next: >