TcaplusDB Container Environment Deployment
1. Version Introduction
TcaplusDB Local (Container Version) is a version made available to users for local development and debugging (downloadable version based on Docker deployment). Compared with the Tencent Cloud TcaplusDB version, the local version can facilitate users to debug code in the development environment that cannot connect to the cloud environment, and has no difference in functions from the Tencent Cloud version. This file mainly describes how to deploy the TcaplusDB local Docker environment in the local development environment.
If you encounter problems during deployment, you can consult the FAQ list first. If there is no solution in the list, you can contact the TcaplusDB team to help solve problems.
2. Compatibility Description
It has been tested for compatibility in the following environments. It is recommended to deploy the local version of TcaplusDB in an environment that has been tested for compatibility.
| Environments that have completed compatibility test |
|---|
| Windows10 + WSL2 (CentOS 7) |
| MAC + Parallels + Debian |
| CentOS 7 |
| TLinux1.2/TLinux2.2 |
TcaplusDB Local version is required to run in the Linux environment as a Docker container. If the system is Windows or MAC, run the Linux virtual machine or WSL (Windows) on these systems first, and then run the Docker container in the virtual machine.
When deploying the TcaplusDB Local version in an environment without compatibility test, some unknown compatibility problems may occur.
3. Performance Description
| Host specification | Support QPS |
|---|---|
| 4-core 8 G memory | 2000 |
| 8-core 16 G memory | 4000 |
| >= 16-core 32 G memory | 8000 |
Description:
Since the number of tables is related to the number of shards in tcapsvr, a maximum of 512 tables can be created when only one group of machines is installed by default.
Supported QPS are test values based on default table engine parameters
bnum=7500000#xmsiz=12000000#xikmsiz=3000000#kmapow=6#kfapow=6#vmapow=6#vfapow=6#lnum=2, 1 kb data read and write, and 10 GB data scale, rather than the upper limit.
4. Environment Requirements
The deployment machine should have at least 4 cores of CPU, at least 8 G of total memory, and at least 6 G of free memory before starting the TcaplusDB Local container.
The recommended disk space for the deployment machine is 100 GB, and the available disk space for storing the Docker container should be greater than 50 GB.
Docker needs to be installed in advance to deploy the machine. See Appendix 1 for the installation method.
The user who creates a Docker container must have permission to execute docker commands through sudo.
The deployment machine should have SELinux turned off.
If the deployment machine is Windows+WSL2, vsyscall should be set correctly. See Appendix 3 for the setting method.
If the deployment machine is Debian (or Debian virtual machine), vsyscall should be set correctly. See Appendix 4 for the setting method.
5. Version Download
6. Version Deployment
6.1. Version Deployment in Linux Environment
6.1.1. Check and prepare the deployment environment
Download the environment scan script.
wget https://tcaplus-tools-1301716906.cos.ap-guangzhou.myqcloud.com/tcaplus_tools.sh
Run the environment scan script to scan for environment resources and check whether the TcaplusDB Local version meets the minimum operating requirements.
sh tcaplus_tools.sh check_environment
Example of environment scan results:
Environment Information:
OS : Tencent tlinux release 2.2 (Final)
CPU Arch : x86_64
CPU Cores : 8
Memory Total (MB) : 15787
Memory Free (MB) : 1025
Memory Cached (MB) : 4542
Disk Free (MB) : 15787
Docker Version : Docker version 18.09.7, build 2d0083d
LDD Version : ldd (GNU libc) 2.17
GCC Version : gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-4)
G++ Version : g++ (GCC) 4.8.5 20150623 (Red Hat 4.8.5-4)
Go Lang Version : Not found
Java Version : Not found
Python Version : Python 2.7.5
SELinux Setting : Disabled
Reference values of indicators in the environment scan results:
| Indicator | Reference value |
|---|---|
| OS | See Version Compatibility descriptions |
| CPU Arch | x86_64 |
| CPU Cores | >= 4 |
| Memory Total (MB) | >= 8192 (8G) |
| Memory Free (MB) | >= 6144 (6G) |
| Disk Free (MB) | >= 51200 (50G) |
| Docker Version | It is suggested >= Docker 1.13.1 or >= Docker CE 18.09.7. See Appendix 1 for the installation method. |
| SELinux Settings | Disabled For details about how to disable the function, see Appendix 2 |
Some processes of TcaplusDB rely on mdadm, which needs to be manually installed if it is not installed by default on the deployment machine.
#centos
yum install -y mdadm
#debian
apt-get update
apt-get install -y mdadm
6.1.2. Check whether Docker is started
Log in to the deployment machine.
Check whether the Docker service is installed and started normally. See Appendix 1 for the installation method of Docker.
# Check the Docker service status and check whether the service has been started
sudo service docker status
# Start the Docker service command
sudo service docker start
6.1.3. Import Local image
Import the TcaplusDB Local image to the local Docker image library.
sudo docker load -i tcaplusdb-local-3.55.1.tar.gz
View the local Docker image library to check whether the image is successfully imported.
sudo docker images
If the image information is displayed in the local Docker image library, the Docker image is imported successfully.
REPOSITORY TAG IMAGE ID CREATED SIZE
tcaplusdb-local 3.55.1 bb42e8ecff11 3 days ago 8.43GB
6.1.4. Create a Local container
Example of container creation command:
sudo docker run -itd --memory=8g --storage-opt size=50G --shm-size=2G --privileged -e TCAPLUS_CONTAINER_PUBLIC_IP="x.x.x.x" -e TCAPLUS_CONTAINER_OMS_PASSWORD='***' -p 80:80 -p 9999:9999 -p 13755-13765:13755-13765 --name test tcaplusdb-local:3.55.1
Description of container startup parameters:
| Parameter | Description |
|---|---|
| --shm-size=2G | [Optional] Specify the shared memory size of the container, with a minimum requirement of 2 G. When the shared memory is not specified or it is detected that the shared memory is less than 2G when the container is started, it will be forced to be set to the minimum value of 2 G. |
| --privileged | [Required] Specify to start the container in privileged mode. The startup of TcaplusDB Local involves the modification of some system parameters. If the container is not started in privileged mode, the startup will fail. |
| --name test | [Required] Specify the container name. The name can be used as a parameter for viewing container logs and starting or stopping the container. |
| -e TCAPLUS_CONTAINER_PUBLIC_IP="x.x.x.x" | [Optional] Specify the IP address that the TcaplusDB container provides services to. This IP address needs to be set if TcaplusDB is to be accessed from outside the container. This value is generally the external IP address of the host. |
| -e TCAPLUS_CONTAINER_OMS_PASSWORD="*" | [Required] Specify the login password of tcaplus for the TcaplusDB OMS user. For security reasons, the TcaplusDB container does not provide a default login password, which needs to be specified by the user when the container is started. [Note] After the TcaplusDB service is started, users can log in to the TcaplusDB OMS page through the tcaplus username and password. |
| -p 80:80 -p 9999:9999 -p 13755-13765:13755-13765 | [Optional] When running the container in bridge network mode, a mapping relationship between the port in the container and the port of the host needs to be established through - p if service is to be accessed through the port of the host outside the container. [Note] At present, ports that need to be exposed in the Local version of TcaplusDB are mainly 80, 9999, 13755-13765. See Appendix 5 for ports monitored by TcaplusDB. |
| tcaplusdb-local:3.55.1 | [Required] Image name and tag. |
| --cpus=4 | [Optional] If necessary, this parameter can be used to specify the number of CPU cores that can be used by the container. The parameter value should be greater than or equal to 4. |
| --memory=8G | [Optional] If necessary, this parameter can be used to specify the available memory size of the container. The parameter value should be greater than or equal to 8 G. |
| --storage-opt size=50G | [Optional] If necessary, this parameter can be used to specify the available disk size of the container. The parameter value should be greater than or equal to 50 G. [Note] For some Docker versions, the disk size available to the container is 10 G by default. If the disk size is used up soon after the service runs for a period of time, you need to increase the disk size by modifying this parameter. |
| --network User defined network name | [Optional] In some environments, the host network segment may conflict with the container default network segment. At this time, the customized Docker network should be created and used by the container specified by this parameter to avoid network conflicts. [Note] Docker containers usually default to 192.168 or 172.17 container network segments. When the host happens to be in these network segments, a Docker network should be created that is different from the host network segment. For example, if the host network segment is 172.17, a Docker network that is on network segment 192.168 can be created, and specify this network when creating containers. For details of the Docker network creation method, see Appendix 6. |
For more container parameters and descriptions, see:
sudo docker run --help
The TcaplusDB service is automatically started after the container is created. In this case, you can run the "docker logs" command to view service startup logs to learn about service startup progress and status. The service usually takes about 60 seconds to start.
When "Run Success" is displayed at the end of the container startup log, the service has been started successfully and the service is available.
sudo docker logs container name
6.1.5. Enter the Local container
If you need to enter the container to view information or locate faults, run the following command.
sudo docker exec -it container name /bin/bash
6.1.6. Close the Local container
If you need to temporarily close the container to release resources, run the following command to close the container.
sudo docker container stop --time=120 container name
Note: Since the TcaplusDB process needs a certain time to exit, if it is forced to close, the data engine file will be damaged, thus affecting the restart of the container. Here, you need to specify that the specified closing timeout is 120 seconds to prevent the TcaplusDB process from being forcibly killed by the Docker before normal close and exit (the Docker's default closing timeout is 10 seconds. If the container is not closed and exited after the timeout, it will be forcibly closed).
6.1.7. Start the Local container
For the closed container, if you want to start the service again, run the following command to start it.
sudo docker container start container name
6.2. Version Deployment in Windows 10 Environment
6.2.1. Special Notes
Deployment depends on Powershell to run relevant commands. It is important to execute PowerShell as an administrator to avoid prompting permission problems.
| Dependency | Description | Download address |
|---|---|---|
| Windows10 | Build Version >= 18362, which can be viewed by running thevercommand on the CMD terminal of Win10 |
|
| WSL2 | Windows Subsystem for Linux 2, Microsoft's Linux deployment plan, which can be upgraded to WSL2 if you were previously using WSL1 | |
| Linux Kernel Update Package | Update package for Linux kernel components for WSL2 | Download |
| CentOS7 Distro | Linux Kernel for WSL2, the system carrier of docker service | Download |
| Docker-Desktop | Docker management platform on Windows, combined with WSL2 to manage images and containers | Download |
6.2.2. Deploy WSL2
For details about WSL2, see Microsoft Official Website. For details about WSL2 deployment, see Official Website. The following describes the specific process. If the user already has a WSL1 environment, please go to "Upgrade WSL1 to WSL2 part"
Step 1: Enable WSL2 components
You can use the command to open PowerShell as an administrator and run the following command:
dism.exe /online /enable-feature /featurename:Microsoft-Windows-Subsystem-Linux /all /norestart
Step 2: Enable virtualization features
WSL2 depends on the virtualization feature and the Hyper-v feature at the same time. It needs to be enabled to open
- Enable virtualization features
dism.exe /online /enable-feature /featurename:VirtualMachinePlatform /all /norestart
- Enable Hyper-v features
dism.exe /online /enable-feature /featurename:Microsoft-Hyper-V /all /norestart
# It is critical and needs to be executed. If not executed, a BIOS problem will be reported
bcdedit /set hypervisorlaunchtype auto
Step 3: Restart the computer
After configuring the virtualization feature, restart the computer for it to take effect.
Step 4: Install the Linux kernel update package
Double click the Linux Kernel Update Package downloaded above to install it.
Step 5: Set WSL2 to support low version CentOS images, such as Centos6
Create a new file .wslconfig in the C:\Users\
[wsl2]
kernelCommandLine = vsyscall=emulate
Step 6 Set the default WSL version number. Open PowerShell as an administrator and run the following command:
wsl --set-default-version 2
If the setting is successful, there is no problem.
6.2.3. Upgrade WSL1 to WSL2
If you have enabled the WSL function and use WSL1, the upgrade is similar. Except step 1, the upgrade can be set according to steps 2, 3, 4, 5 and 6 above. For the existing Linux Distro, you can modify the version of the Distro to WSL2. For example, the name of the existing Distro is CentOS7. The modification method is as follows:
wsl --set-version CentOS7 2
6.2.4. Deploy CentOS7 Distro
The TcaplusDB image works well with the CentOS7 series. You can deploy a CentOS7 kernel. Download CentOS7 Distro according to the above resource preparation stage. Download it to the specified directory and unzip it. Right click to install as an administrator. Set Centos7 as the default Distro, open PowerShell as an administrator, and run:
wsl --set-default CentOS7
Check the current Distro status:
wsl -l -v
6.2.5. Deploy Docker Desktop
Click the Docker-Desktop downloaded in the resource preparation stage to install it. During installation, it will prompt "Enable WSL2", and you can select it directly. For details, see Official Deployment Document.
- If wsl2 is not selected during installation, open Settings of Docker-Desktop after installation and check whether "Use the WSL2 based engine" is selected in General.
- You can view the currently installed Distro on the "Resources" section of the "Settings" of Docker Desktop. It will automatically associate all the installed Distros. In addition to associating the default Distro, you can also choose to associate other Distros.
After the Docker Desktop is installed, you can view the status through wsl - l - v:
PS C:\Windows\system32> wsl -l -v
NAME STATE VERSION
* CentOS7 Running 2
docker-desktop Running 2
docker-desktop-data Running 2
From the above information, the default Distro is CentOS7, and there are two default Distros: docker-desktop, docker-desktop-data.
6.2.6. Import Local image
Confirm that the default Linux of WSL2 is CentOS7. If not, refer to "CentOS7 Distro Deployment" above. Open PowerShell as administrator and run:
docker load -i tcaplusdb-local-3.60.1.tar.gz
After importing, check whether the image is OK.
PS C:\Windows\system32> docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
tcaplusdb-local 3.60.1 xxxxxxxxxxxx 2 hours ago 8.43GB
6.2.7. Create a Local container
In the WSL + Docker environment, WSL cannot directly access the IP address in the container, so the TCAPLUS_CONTAINER_PUBLIC_IP parameter must be specified when the WSL is deployed, and the parameter value of TCAPLUS_CONTAINER_DEPLOY_MODE must be specified as WSL, otherwise the Tcaplus service inside the container will not be accessible outside the container.
#TCAPLUS_CONTAINER_PUBLIC_IP is the ip of the specified proxy. The way to obtain it is to enter the CentOS7 distro environment and use ifconfig to get the ip of eth0,
docker run -itd --privileged -e TCAPLUS_CONTAINER_PUBLIC_IP="192.168.53.2" -e TCAPLUS_CONTAINER_OMS_PASSWORD='***' -e TCAPLUS_CONTAINER_DEPLOY_MODE="WSL" -p 8080:80 -p 13755-13765:13755-13765 -p 9999:9999 --shm-size=3G --name test tcaplusdb-local:3.60.1
The above container creation command also exposes the ports in the container to the host:
- 80: The port of the Web operation and maintenance platform is exposed as 8080, which can be directly accessed through the browser through localhost: 8080
- 13755-13765: Port of tcaplusdb proxy
- 9999: The tcaplusdb directory service address port is the port used for app connection
View the created container:
[root@ballenwen-PC4 e]# docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
4a6ce2b1811c tcaplusdb-local:3.60.1 "sh /data/tcaplus/tc…" 2 seconds ago Up 1 second 0.0.0.0:13755-13765->13755-13765/tcp, 0.0.0.0:8080->80/tcp test
After the container is created, enter the container and wait 20 to 60 s for the tcaplus process to automatically start. You can run the following:
tcaplus_docker 172.17.0.2 root> ps -ef | grep tcaplus
root 1 0 0 21:32 pts/0 00:00:00 sh /data/tcaplus/tcaplus_run.sh
root 883 1 0 21:32 ? 00:00:00 /data/tcaplus/tsf4g_release/apps/tconnd/bin/tconnd --id=1.1.3 --use-bus --bus-key=1688 --conf-file=../cfg/tconnd/tconnd.xml --tlogconf=../cfg/tconnd/tconnd_log.xml --daemon start
root 899 1 0 21:32 ? 00:00:00 /data/tcaplus/tsf4g_release/apps/tcenterd/bin/tcenterd --id=1.1.1 --use-bus --bus-key=1688 --tlogconf=../cfg/tcenterd/tcenterd_log.xml --conf-file=../cfg/tcenterd/tcenterd.xml -D start
root 923 1 2 21:32 ? 00:00:01 /data/tcaplus/tsf4g_release/apps/tcm/bin/tcmcenter --id=1.1.2 --use-bus --timer=10 --bus-key=1688 --bus-beat-gap=60 --conf-file=../cfg/tcm/tcmcenter.xml --tlogconf=../cfg/tcm/tcmcenter_log.xml --conf-format=3 -D start
root 940 1 3 21:32 ? 00:00:02 /data/tcaplus/tsf4g_release/apps/tagent/bin/tagent --id=1.2.1 --tlogconf=../cfg/tagent/tagent_log.xml --conf-file=../cfg/tagent/tagent.xml -D start
root 962 1 0 21:32 pts/0 00:00:00 /data/tcaplus/tcaplus_service/bin/tbusd --epoll-wait=1 --use-tsm --idle-sleep=1 --conf-file=../cfg/tbusd/tbusd.xml --tlogconf=../cfg/tbusd/tbusd_log.xml --no-bus --daemon --id=0.0.0.1 --business-id=0 -D start
root 1320 1318 0 21:32 ? 00:00:00 /opt/lampp/bin/rotatelogs /data/tcaplus/oms/tcapoms/www/log/error/error_log_%Y%m%d 86400 480
root 1324 1318 0 21:32 ? 00:00:00 /opt/lampp/bin/rotatelogs /data/tcaplus/oms/tcapoms/www/log/accesslogs/access_log_%Y%m%d 86400
tcaplus 1712 1318 0 21:32 ? 00:00:00 /opt/lampp/bin/httpd -k start -E /opt/lampp/logs/error_log -DSSL -DPHP
tcaplus 1713 1318 0 21:32 ? 00:00:00 /opt/lampp/bin/httpd -k start -E /opt/lampp/logs/error_log -DSSL -DPHP
tcaplus 1714 1318 0 21:32 ? 00:00:00 /opt/lampp/bin/httpd -k start -E /opt/lampp/logs/error_log -DSSL -DPHP
tcaplus 1715 1318 0 21:32 ? 00:00:00 /opt/lampp/bin/httpd -k start -E /opt/lampp/logs/error_log -DSSL -DPHP
tcaplus 1716 1318 0 21:32 ? 00:00:00 /opt/lampp/bin/httpd -k start -E /opt/lampp/logs/error_log -DSSL -DPHP
tcaplus 1717 1318 0 21:32 ? 00:00:00 /opt/lampp/bin/httpd -k start -E /opt/lampp/logs/error_log -DSSL -DPHP
tcaplus 1718 1318 0 21:32 ? 00:00:00 /opt/lampp/bin/httpd -k start -E /opt/lampp/logs/error_log -DSSL -DPHP
tcaplus 1719 1318 0 21:32 ? 00:00:00 /opt/lampp/bin/httpd -k start -E /opt/lampp/logs/error_log -DSSL -DPHP
tcaplus 1720 1318 0 21:32 ? 00:00:00 /opt/lampp/bin/httpd -k start -E /opt/lampp/logs/error_log -DSSL -DPHP
root 1895 1 5 21:32 ? 00:00:02 /data/tcaplus/tcaplus_service/bin/1_2_1_1/swiftd --use-tsm --idle-sleep=1 --conf-file=../../cfg/access_1_2_1_1/swift_tcaproxy.xml --conf-format=3 --tlogconf=../../cfg/access_1_2_1_1/tcaproxy_log.xml -D --id=1.2.1.1 --bus-key=2088 --business-id=0 -D start
root 1898 1 4 21:32 ? 00:00:02 /data/tcaplus/tcaplus_service/bin/1_3_1_1/swiftd --use-tsm --idle-sleep=1 --conf-file=../../cfg/access_1_3_1_1/swift_tcaproxy.xml --conf-format=3 --tlogconf=../../cfg/access_1_3_1_1/tcaproxy_log.xml -D --id=1.3.1.1 --bus-key=2088 --business-id=0 -D start
root 1914 1 31 21:32 ? 00:00:16 /data/tcaplus/tcaplus_service/bin/1_1_2_1/tcapsvr --use-tsm --idle-sleep=1 --nthread=8 --bus-check-gap=0 --threadChannelSize=20480000 --conf-file=../../cfg/storage_1_1_2_1/tcapsvr.xml --conf-format=3 --tlogconf=../../cfg/storage_1_1_2_1/tcapsvr_log.xml --daemon --id=1.1.2.1 --bus-key=2088 --business-id=0 -D start
root 1931 1 30 21:32 ? 00:00:15 /data/tcaplus/tcaplus_service/bin/1_1_2_2/tcapsvr --use-tsm --idle-sleep=1 --nthread=8 --bus-check-gap=0 --threadChannelSize=20480000 --conf-file=../../cfg/storage_1_1_2_2/tcapsvr.xml --conf-format=3 --tlogconf=../../cfg/storage_1_1_2_2/tcapsvr_log.xml --daemon --id=1.1.2.2 --bus-key=2088 --business-id=0 -D start
root 1938 1 4 21:32 ? 00:00:02 /data/tcaplus/tcaplus_service/bin/1_3_8_1/swiftd --use-tsm --idle-sleep=1 --conf-file=../../cfg/tcaprestproxy_1_3_8_1/swift_tcaprestproxy.xml --conf-format=3 --tlogconf=../../cfg/tcaprestproxy_1_3_8_1/tcaprestproxy_log.xml -D --id=1.3.8.1 --business-id=0 -D start
root 1944 1 4 21:32 ? 00:00:02 /data/tcaplus/tcaplus_service/bin/tcapdir/swiftd --use-tsm --idle-sleep=1 --conf-file=../../cfg/tcapdir/tcapdir_swift.xml --tlogconf=../../cfg/tcapdir/tcapdir_log.xml -D --id=1.0.4.1 --business-id=0 -D start
root 2000 1 29 21:32 ? 00:00:14 /data/tcaplus/tcaplus_service/bin/1_2_5_1/tcapdb --use-tsm --idle-sleep=1 --nthread=8 --threadChannelSize=10000000 --timer=3000 --conf-file=../../cfg/tcapdb_1_2_5_1/tcapdb_cfg.xml --conf-format=3 --tlogconf=../../cfg/tcapdb_1_2_5_1/tcapdb_log.xml -D --id=1.2.5.1 --bus-key=2088 --business-id=0 -D start
root 2253 2140 0 21:33 pts/1 00:00:00 grep tcaplus
If creating a container fails, check for the following reasons:
- Check machine resources: The available disk size is above 50 G, and the available memory is 4 to 6 G
- Check system parameters: Whether to create C:\Users\
.wslconfig - View container creation logs: docker logs
6.2.8. Close the Local container
If you need to temporarily close the container to release resources, run the following command to close the container.
sudo docker container stop --time=120 container name
Note: Since the TcaplusDB process needs a certain time to exit, if it is forced to close, the data engine file will be damaged, thus affecting the restart of the container. Here, you need to specify that the specified closing timeout is 120 seconds to prevent the TcaplusDB process from being forcibly killed by the Docker before normal close and exit (the Docker's default closing timeout is 10 seconds. If the container is not closed and exited after the timeout, it will be forcibly closed).
7. Version Test
After the version deployment is completed, you can connect to TcaplusDB through SDK, OMS, etc. for data reading and writing or cluster management.
TcaplusDB Local version creates test services, game zones and tables by default. You can log in to TcaplusDB OMS to view relevant information.
7.1. Connect through OMS
For this section, refer to: Using the TcaplusDB OMS Console.
The login address of OMS is http://\
The default user name for logging in to OMS is tcaplus, and the password is the parameter value of the TCAPLUS_CONTAINER_OMS_PASSWORD specified when starting the container.
7.2. Connect through SDK
For this section, please refer to: TcaplusDB TDR Table C++SDK Instructions
8. FAQ
8.1. Container Startup Log: Failed to write channels, for shmctl
Reason:
The shared memory size of the container is set too small. The default shared memory size of some Docker versions is 32 M, while the minimum value required for the TcaplusDB container is 2 G.
Solution:
Use images of the latest 3.53.1 version or a later version.
When starting the container, the shared memory size of the specified container can be displayed through -- shm size=2G.
8.2. Failed to Start the Container in Debian System, with the Container Exit Code = 139
Reason:
The current exit code of 139 is mostly a problem with the vsyscall setting of the master
Solution:
For the method of setting vsyscall in Debian, refer to http://helpcenter.onlyoffice.com/installation/mail-enabling-vsyscall.aspx;
For the method of setting vsyscall in Windows WSL2, refer to https://zhuanlan.zhihu.com/p/256317180
8.3. The SSH, PWD and Other Commands and Services in the Container Are Unavailability, Resulting in the Unavailability of Some Functions of TcaplusDB
Reason:
Most of the reasons for the current situation are that the SELinux mandatory security policy is enabled on the host, resulting in the unavailability of SSH, PWD and other commands or services in the container. The immediate impact is that some of the functions of TcaplusDB that depend on these commands and services are unavailable, such as table clearing.
Solution:
Close the SELinux mandatory security policy of the host.
Close temporarily.
setenforce 0Close permanently. Edit the /etc/selinux/config file, set SELINUX=Disabled, and restart the computer.
8.4. The Host Network Segment Conflicts with the Docker Container Network Segment, Resulting in the Client Being Unable to Vonnect to the Service
Reason:
In some environments, there may be conflicts between the host network segment and the container default network segment. The Docker container is usually 192.168 or 172.17 by default. When the host happens to be in these network segments, TcaplusDB will treat the request sent from the host as an intranet request and return the IP in the container to the client as the service IP. In some environments (such as WSL), the host cannot directly access the IP in the container, so the service status is normal, but the client cannot connect to the server.
Solution:
Create a Docker network that is on a different network segment from the host. For example, if the host network segment is 172.17, create a Docker network that is on network segment 192.168 and specify this network when creating containers. The method of creating a Docker network is as follows.
Run the following command to create a Docker network.
sudo docker network create --driver=bridge --subnet=192.168.10.0/24 subnet_192_168_10
192.168.10.0/24 indicates the Docker network segment and mask.
subnet 192 168_ 10 is the Docker network name, which is followed by network in the container creation command.
8.5. When deploying the WSL environment, the containers created before the restart cannot be accessed externally after the subsystem restarts
Reason:
After the WSL subsystem restarts, the IP assigned by the system to the subsystem changes, but the previously started containers are configured with the original subsystem IP as the Public IP of the container, so external access to the container cannot continue through the original subsystem IP.
Solution:
Use Tcaplus Docker to change the Public IP address of the container to the new subsystem IP address. The detailed steps are as follows:
- Download the Tcaplus Docker tool script.
wget https://tcaplus-tools-1301716906.cos.ap-guangzhou.myqcloud.com/tcaplus_tools.sh
- Modify the public IP of the container
sh tcaplus_tools.sh change_container_public_ip <container name> <new IP>
8.6. After adding a table to OMS, the command is not executed but is waiting for the queue to execute.
Solution:
View transactions that are not pending and executing, and if necessary, other transactions have completed or ended.
The tcapcenter has been started for more than 3 minutes. The tcapcenter starts to execute transactions 3 minutes after it is started.
If 1 and 2 are excluded, you can restart the tcaplus center process in the container. Restart method:
Enter the tcaplus docker container: docker exec - it container name /bin/bash
Switch to tcaplus user: su - tcaplus
Switch to/data/tcaplus/tcaplus Service/bin/directory: cd/data/tcaplus/tcaplus service/bin/
Start tcm console: ./start_tcmconsole.sh
Execute the restart command on the tcm console: restart 1.0.9.1
8.7. Container startup error: docker: Error response from daemon: --storage-opt is supported only for overlay over xfs with 'pquota' mount option
Reason:
The current disk mounting method does not support specifying how many disks to use
Solution:
Remove this parameter from the startup command. This parameter is not required -- storage opt size=50G
8.8. BIOS Problems
Problem description:
Please enable the Virtual Machine Platform Windows feature and ensure virtualization is enabled in the BIOS
Solution: Enable Hyper-v, use bcdedit to set virtual features for automatic registration, and reset the BIOS configuration
dism.exe /online /enable-feature /featurename:Microsoft-Hyper-V /all /norestart
bcdedit /set hypervisorlaunchtype auto
8.9. Container Creation Failure
Problem description: The container automatically exits after being created. It is found that WSL2 requires additional configuration parameters for Centos6.x image. Solution:
# Create a file in the user directory
c:\users\<username>.wslconfig
# Fill the file with the following
[wsl2]
kernelCommandLine = vsyscall=emulate
8.10. The Container Created before Restart Cannot Be Accessed Externally after Restart of Subsystem.
Use Tcaplus Docker to change the Public IP address of the container to the new subsystem IP address. The detailed steps are as follows:
Download the Tcaplus Docker tool script.
wget https://tcaplus-tools-1301716906.cos.ap-guangzhou.myqcloud.com/tcaplus_tools.sh
Modify the Public IP of the container.
sh tcaplus_tools.sh change_container_public_ip <container name> <new IP>
9. Limitations on Use
Distributed indexes can be created, but clusters cannot be created.
10. Appendix
10.1. Installing Docker
Docker is generally installed through system commands, such as yum and apt-get. Take CentOS7 for example. The installation method is as follows:
# CentOS 7 environment, CVM instance
yum install -y docker
# Debian 8.2 environment
apt-get update
apt-get install -y docker-ce containerd.io
# CentOS 8 environment, CVM instance, installed with dnf
dnf config-manager --add-repo=https://download.docker.com/linux/centos/docker-ce.repo
dnf repolist -v
dnf list docker-ce --showduplicates | sort -r
dnf install docker-ce --nobest
Note: During the installation of Docker, pay attention to the configuration of the image source. It can be configured as a domestic image source by referring to relevant online documents.
After installing Docker, you can use the Docker command to check whether the installation is successful.
If the installation is successful, execute the following command to start the docker service.
# Start the Docker service
sudo service docker start
10.2. Closing SELinux
Check SELinux status.
getenforce
Close temporarily.
setenforce 0
Close permanently. Edit the /etc/selinux/config file, set SELINUX=Disabled, and restart the computer.
10.3. Setting vsyscall (Windows+WSL2)
In the system disk - User - \< user name > directory of Windows 10, modify the .wslconfig file (for example, C:\Users\test.wslconfig). If you do not have this file, you need to create it first.
Modify/add the following:
[wsl2]
kernelCommandLine = vsyscall=emulate
10.4. Setting vsyscall (Debian)
See Descriptions.
10.5. Monitoring Ports and Descriptions in the Container
| Port | Component | Description |
|---|---|---|
| 21 | SFTP | FTP Service Port A new FTP is built to solve the problem of slow packet transfer function of TCM |
| 80 | TcapOMS | TcaplusDB OMS HTTP Service Port |
| 443 | TcapOMS | TcaplusDB OMS HTTPS Service Port |
| 3306 | MySQL | MySQL Service Port |
| 4500 | Tcapdb | TcaplusDB Tcapdb Service Port |
| 6000 | Tcapcenter | TcaplusDB Center Service Port |
| 8899 | Tcenterd | Tcenterd Service Port |
| 9010 | Tconnd | Tconnd Service Port |
| 9017 | Tbuspp | Tbuspp Service Port |
| 9020 | Tbuspp | Tbuspp Service Port |
| 9200 | Elasticsearch | Elasticsearch Service Port |
| 9300 | Elasticsearch | Elasticsearch Service Port |
| 9898 | Tcapdir | Tcapdir Service Port |
| 9999 | Tcapdir | Tcapdir Service Port |
| 13755 | Tcaproxy | Tcaproxy Service Port |
| 13756 | Tcaproxy | Tcaproxy Service Port |
| 31001 | Tcaprestproxy | Tcaprestproxy Service Port |
| 36000 | SSH | SSH Service Port |
| 45755 | Tcapsvr | Tcapsvr Service Port |
| 45760 | Tcapsvr | Tcapsvr Service Port |
| 46755 | Tcapsvr | Tcapsvr Service Port |
| 46760 | Tcapsvr | Tcapsvr Service Port |
10.6. Creating Docker Network
Run the following command to create a Docker network.
sudo docker network create --driver=bridge --subnet=192.168.10.0/24 subnet_192_168_10
192.168.10.0/24 indicates the Docker network segment and mask.
subnet 192 168_ 10 is the Docker network name, which is followed by network in the container creation command.