Shulou(Shulou.com)06/02 Report--

The Spring Festival in 2020 must be the most unforgettable Spring Festival for everyone, and everyone is concerned about the flu epidemic. I can't go out. I can only watch movies and read books at home. As an operation and maintenance worker, I came across a multi-cloud operation and maintenance platform product called TiOps, which supports secure remote operation and maintenance management in multi-cloud environment. Our company has Ucloud and Aliyun, office environment and a lot of virtual machine development environment, just lack of a management software, this software recently launched activities for free 100CVM, 180-day license. I applied for a trial without hesitation.

Apply for

First talk about the application method, the application method is very simple, www.tyun.cn this website can be downloaded.

After going in, it is divided into the community version and the commercial version. Here I choose the community version for testing, and pay attention to it when choosing the other version. It is divided into two versions:

OVA image package

This is that you can quickly use the script to install the package by starting the vmware image directly.

This is to complete the installation by command

You will be notified of the download in less than a minute. You can see that the installation package is placed on Amazon S3, and the email also contains certificates and user manuals, which can be said to be very sweet.

Download and install

The size of the image is 2.3G. It is recommended to download it using IDM. My home bandwidth 1000m download only took more than 10 seconds, I do not know if 5G is also so fast.

Import virtual machine images using VMware Workstation

It will take about 2 minutes to complete the import. Mine is that the old mechanical hard drive is slower, and the solid-state hard drive is estimated to be ready in 1 minute.

The exciting time has come to turn on virtual machine use.

After turning on the virtual machine, you can see that the system is CentOS7, and then I don't know how to read the operation manual.

Open the OVA image deployment manual https://tyun.cn/docs/All_in_one_ova_installation.pdf

You can see that the account number of the console is root, the password is tiops@123, and then do the image initialization operation to change the IP address of the system and restart the network.

# vi / etc/sysconfig/network-script/ifcfg-ens160

# systemctl restart network restart the network and use the ssh tool link to initialize the work

After the login is successful, run the initialization.sh script in the / root directory and type y to continue

Start after initialization

#. / initialization.sh start Open the platform initialization operation manual https://tyun.cn/docs/Sample_init.pdf login platform interface http://192.168.10.210 enter the default user name admin password 321Cba.@ login

After logging in, it will show that the authorization expires and allows you to enter the authorized company and the authorization key, which is copied and pasted in the email.

After updating the authorization information, you will jump back to the login interface and log in again to enter the platform.

The interface is very clean and tidy. Let's take a look at what features use multi-cloud management.

Multi-cloud management includes three parts: cloud account, CVM and network.

Cloud accounts can be imported into public clouds, private clouds, and local area network hosts. Here, I will try to import a public cloud account.

If it is aws, it is divided into overseas and domestic. Here I import Aliyun's AK.

After import, there is a synchronization button on the side to synchronize all the CVMs.

If you do not know the area during import, you can select all of them, and the scanning time for all selection will be less than 1 minute.

At this point, you can see my test machine, and show that the status is not running. Check this server to synchronize the host

If you enter the CVM, you can see the host you just imported. Is it very simple? the status of the host here is turned off. Let's start it with the platform to see the effect.

You will reconfirm whether you want to turn on the mainframe before turning on the computer.

The page will show that the instance status is running

After running, the status of the instance is changed to running. Let's verify it on the Aliyun page.

Click on this host in the list of CVMs to see all its details.

It may be because I use elastic IP. The public network IP address is not shown here.

There is an estimated cost of your machine in the consumption record.

The security rules are the rules of the security group

The operation log contains all the records of the operations on the platform.

I don't know what the host setting is. Finally, take a look at the network in multi-cloud management.

Here is the network for setting up this platform. According to the user manual, it is a gateway for private network connection with different cloud environments. You need to install agent. Let's try to install agent.

This side shows that the gateway assistant has not been installed yet. Let's try one.

It is divided into automatic installation and manual installation. I will definitely choose automatic installation if I am lazy.

Enter the relevant information and click next.

Installation has been waiting for a long time

I don't know if the key is incompatible. I failed to install the key generated by Aliyun.

Now try manual installation.

The contents of the copy script are executed by the administrator on the server

Obviously I know the reason for the failure of the automatic installation, this is to download the installation package on my intranet host, but I Aliyun can't link to my intranet. It is hoped that merchants can put these installation packages on the public network for automatic download. Here, I downloaded the installation package and manually uploaded it to install it myself.

Enter http://192.168.10.210/expose/servers/download/install on the browser to download and upload to the server

Then enter the following command on the server

# export import_domain_key=0718EC78962342BB85621E15A3FDA8A1# export import_gateway_key=728E75E9F3D04995BF6BC47B01D1C4C0# tar-xzvf tyun_install.tar.gz-C / tmp# bash / tmp/install.sh proxy 192.168.10.210 http://192.168.10.210 4505 4506

When it comes to this command, I find that its installation still depends on the service of my intranet, and the gateway assistant installation can only be abandoned.

It is better to install agent on the server. The gateway assistant is not installed in my test environment, and it is not a big problem. If you really want to install it, you can try to map out port 80 of the server. My telecom cat does not have an administrator password, so it cannot be mapped. Back to the point, let's start to install agent, click not installed, and remember to install Agent in batches, which requires a gateway helper server, and the gateway helper needs to be able to log in to other servers without a password

Select the uninstalled menu to automatically jump to the operation and maintenance management. Here I will use the automatic installation first.

It's a tragedy. You still have to install a gateway assistant. All right, I'll take a look at the manual installation commands.

Or do you have a gateway assistant?

All right, I don't need agent for the time being. Anyway, I'll just have a server for trial.

Hardware management

Next let's take a look at hardware management.

Hardware management is divided into computer room management, cabinet management, server management and network equipment management.

The management of the computer room is edited by hand. I'll randomly add a demonstration here.

Cabinet management is the same as computer room management by manual input and editing.

Server management here should refer to the server of the physical machine, but also manually edited and added

Network device management is divided into switch, router and firewall, which is also manually edited and input, so there is no demonstration here.

Business tree management

Business tree management corresponds to asset allocation. For example, some resources belong to A department and some belong to B department.

You can't see any hosts after editing.

Later, the allocation option was found in more operations of the host in the CVM in multi-cloud management, but the result still could not be assigned that it was not an application node.

You can't add hosts until you continue to add the business tree. I don't know why it is designed to go to the application node group.

Operation and maintenance management

Operation and maintenance management is divided into host management, script distribution, command distribution, file distribution, job scheduling. Let's try host management one by one: you can see the business tree and the list of hosts, and you can reset passwords, boot, shut down, restart, script distribution, file distribution, and open sessions. Let's try to start a conversation first.

Click Connect after entering login information

Login is a bit slow. I don't know if it is the first time to log in. There are file transfer and session details on the right. File transfer is easy to add directly to the page and you can transfer files.

And then talk about script distribution.

This script is distributed with parameter input boxes that support shell and python, as well as two types of windows.

Let's upload the edited shell script test.sh first.

Since we don't have agent installed, we distribute the test with ssh, and remember to add the host to execute after editing the script.

Finally, click on the immediate distribution below to see the effect.

The execution error said that the login information could not be empty. I looked for it and saw that the login information of the host could be created in the system settings.

Then try script distribution again, login information is still not displayed, or go to the host settings in the CVM to see the binding login information, and finally know what the host settings are for.

After binding, the script is executed again and the script is distributed successfully. In the script distribution history, you can see that it took a total of 1 minute and 09 seconds to execute a script with an echo command. I really don't understand why ssh is so slow.

Then try to distribute the command.

The execution of the command is fast for about 2-3 seconds, and there are two places where the command is entered in batch execution, and each host can execute the command separately.

Next, try file distribution.

Here you can add local files as well as files on the remote host. Let's try to send the files on the remote host to itself.

Let's first create a file in the root home directory

# echo aaa > aaa.txt

Then add the source file to point to this file

Then add the target host and path, where we use / tmp

Finally, click to distribute now

It soon showed that the distribution was successful. We went to the host to see if the file was successfully distributed. Unfortunately, it failed, as can be seen from the file distribution history.

We can't see the reason for the failure here, so let's try using local files.

The local file was uploaded successfully, but it took a total of 50 seconds to send a very small file and finally try the homework choreography.

After adding it, you can see that it has job scheduling, scheduling tasks, execution log, and operation log.

The steps to be added to the job scheduling are execution scripts and file distribution.

If you are adding a script, you can set the parameters and timeout of the script on the right

The target host needs to be added.

In order to achieve the demonstration effect, I will add it twice in the script here.

Click to execute immediately to test the effect first, it is a pity to report an error again, saying that there are no steps. I see a script library below. Do you want to add scripts to the script library first?

I created a script library called test and uploaded the script distributed by the previous script. Note that the script must be published here, otherwise the script will not be seen in the job choreography.

Then go back to the job schedule to execute the job again or report the same problem. Click to check in and find that the steps created before are gone. I don't know why. Later, I know that I forgot to save the job in the lower right corner. Here is another important point to say. Jobs are in order of execution. It can be completed the first job and then executed the second or the first execution successfully and then executed the second. The difference between the two is obvious: one will continue to execute regardless of whether you succeed or fail, and the second is that the first execution must be successful before the second. This may be used in some similar release scenarios.

After saving, we click execute immediately to see the effect.

The execution failed. From the log, the script should be uploaded to the / tmp directory and renamed and then executed. It may fail because the upload is too slow.

I'll set the timeout up and look at it again, and the result is still the same. This estimate is a bug.

There is also a scheduled task in the job scheduling that can be executed regularly, and these jobs can be refined to every second.

Script library

The script library has just demonstrated that this function is still good. For example, I have a script that initializes the server, which can be executed in batches without having to execute it one by one every time, and the script can settle down and continue to use the operation log next time.

The operation log is still very detailed, whether it is the execution log of platform operation or script, as well as the video playback of the host computer.

User center

User center permissions are divided in detail, including accounts, users, user groups, and you can create different roles, such as authorization by resources and functions, etc.

System setup

It can set host login information and the security policy of operation and maintenance. Overall, this cloud platform management tool is fully functional and can meet the needs of many small and medium-sized companies, but there are still many areas that need to be optimized.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.