Automatic deployment method for geophysical prospecting processing interpretation software terminal

Abstract

A geophysical prospecting processing interpretation software terminal automatic deployment method belongs to the technical field of petroleum geophysical prospecting computer application, and realizes batch terminal Red Hat Linux operating system installation through network-based linux system unattended installation technology application, and realizes installation and setting of multiple linux version geophysical prospecting software through network-based geophysical prospecting software mirror image installation. And compiling a shell script to realize batch management of a terminal system and software. Therefore, technical innovation in three aspects of system installation, software installation and batch management is realized, the operation and maintenance efficiency of the system is improved, the software installation time of the system is greatly shortened, the error occurrence probability is reduced, and the operation and maintenance efficiency of the system is improved. Automatic installation of the system and the software is achieved, the system and the software are installed in one step, the whole installation and configuration process of each terminal is less than half an hour, and multiple workstations can be installed at the same time.

Description

Technical Field

[0001] The invention belongs to the technical field of petroleum geophysical exploration computer application, and in particular relates to an automatic deployment method for geophysical exploration processing and interpretation software terminal. Background Art

[0002] In terms of distributed systems and high-performance computing, Linux-based PC clusters have good performance-price ratio and high reliability, portability, and scalability. These advantages are also suitable for the requirements of seismic data processing in the field of petroleum geophysical exploration. Therefore, Linux-based PC clusters are used as the hardware and software platform for seismic data processing to build an advanced system environment, which has achieved good practical and economic value.

[0003] At present, in the processing of geophysical data, computer equipment is mainly based on CPU clusters and GPU clusters, supplemented by tower or rack-mounted high-performance servers and NAS storage, which can handle most of the current seismic data processing work. However, the number of cluster nodes is large, and a variety of different batch management methods are required to implement node systems, software installation and operation and maintenance.

[0004] In terms of explanation, the traditional stand-alone mode has a simple structure, using a graphic workstation to access the local area network, and each workstation runs independently. The user experience of running the software in stand-alone mode is smooth, and it is the simplest and most stable way to use and run. The configuration method of a single workstation is to maximize the satisfaction of business needs, but there are many defects and problems: each device needs to install the system and software separately. Once a problem occurs in the system, all software and environment need to be reconfigured. The operation and maintenance management is complicated, and there are problems such as high maintenance difficulty and low system operation and maintenance efficiency. These problems have seriously affected the use and maintenance of geophysical computers.

[0005] Generally, Linux system installation is to boot Linux with a floppy drive or CD-ROM drive, and then install Linux through the local CD-ROM drive. However, this installation method is not applicable in the following situations: no floppy drive or CD-ROM drive, non-standard floppy drive or CD-ROM drive, and a large number of terminals need to install Linux at the same time. If you install one by one through the CD-ROM drive, it is not only inefficient but also not conducive to maintenance.

[0006] In addition, installing and configuring each machine one by one is not only time-consuming, but also very likely to cause errors. In the process of system maintenance, various problems often arise and the system or software needs to be restored. Each time, from restoring the system to reinstalling various processing and interpretation software, it is particularly time-consuming and labor-intensive. It needs to be installed and set up step by step. According to this method, it takes about one or two days to install the operating system and application software. Summary of the invention

[0007] In order to solve the above problems, the present invention proposes: a method for automatic deployment of geophysical processing and interpretation software terminal, which adopts a network-based Linux system unattended customized installation method, comprising the following steps:

[0008] S1. PXE protocol for booting the Red Hat Linux installation program from the network on a PC;

[0009] S2. The network installation function provided by the Red Hat Linux installation program refers to accessing the installation media through the network;

[0010] S3, the unattended installation feature provided by the Red Hat Linux installation program.

[0011] Furthermore, the following software and hardware resources are used to complete the automated installation: a PC machine as the boot and installation server; a PC machine to be installed, whose network card must have PXE support; a built local area network, and the above multiple machines have been connected to the same subnet; a set of Red Hat Linux installation media to be installed.

[0012] Furthermore, the steps of implementing the automated installation include:

[0013] S11, configure and start the server;

[0014] S12, configure and install the server;

[0015] S13. Boot the machine from the network to complete the installation.

[0016] Furthermore, the steps S11 and S12 need only be performed once. For each machine on which Red Hat Linux needs to be installed, corresponding data needs to be deployed in the startup server and the installation server, and step S11 needs to be performed once.

[0017] Furthermore, a network-based geophysical software image installation method is adopted. This method uses an installation template to create a software and system configuration tar package, writes a shell script, and executes the script through an nfs server. It can be installed on multiple workstations at the same time, greatly shortening the deployment time of batch workstations.

[0018] Furthermore, various new and old operating systems and versions process and interpret Linux software to achieve batch deployment, find the relevant system configuration, services, and variables during software installation, modify them one by one, and make them into template files. The implementation process includes the following steps:

[0019] S31. Use conventional methods to manually install a server and install various versions of processing and interpretation software: OpenWosks, Jason, Epos, Geoeast, configure the Oracle environment, system kernel configuration, network configuration, user configuration, etc. After the installation is completed and tested by the user without any problems, it will be used as a configuration template;

[0020] S32, making software tarballs, system configuration files and user file tarballs, including system configuration files, user home directories, cold backup tarballs after software installation, and work area path backup tarballs;

[0021] S33, write a shell script, use shell programming, create a new executable file and write the shell script to be executed. The script needs to be tested repeatedly and confirmed to be correct before it can be executed;

[0022] S34. Build an NFS server, edit the NFS server configuration file / etc / exports to allow other machines to access the directory through NFS, and put all the backed-up software tarballs, system configuration files, and user file tarballs into it;

[0023] S35. Install the operating system on the client network. Use the unattended installation technology of the Linux system to batch install the client Linux operating system through the network, install the network card and graphics card drivers, and set up the services that need to be started and stopped, as well as the necessary rpm packages;

[0024] S36. The client runs the installation script, mounts the network shared disk of the NFS server to the client locally and executes the software installation script. Multiple clients can perform the operation simultaneously.

[0025] Furthermore, batch operations are implemented using expect's shell programming. Expect is developed based on the Tcl language and is used in interactive operations and automated testing scenarios. It is used in environments where the same operations need to be performed on multiple servers, greatly improving the work efficiency of system operation and maintenance personnel.

[0026] Furthermore, send, expect and spwan are the most basic commands of the Expect language. The send command sends a string to the specified process, the expect command waits for the result returned by the process and decides the next operation based on the returned string; and the spwan command can initiate the running of a process.

[0027] Expect can be used to implement SSH login to run related commands, file management, database management and other operations, such as batch execution of system software installation, batch configuration of system files, database backup, hardware information reading, and batch shutdown.

[0028] The beneficial effects of the present invention are as follows: the present invention realizes batch terminal Red Hat Linux operating system installation through the application of network-based Linux system unattended installation technology, realizes installation and setting of multiple Linux versions of geophysical software through network-based geophysical software mirror installation, and writes shell scripts to realize batch management of terminal systems and software. Thus, system management technology innovation is realized in three aspects: system installation, software installation, and batch management, which greatly shortens the time of system software installation, reduces the probability of error occurrence, and improves system operation and maintenance efficiency. The system and software installation automation is realized, and the system and software are installed in one step. The entire installation and configuration process of each terminal takes less than half an hour, and multiple workstations can be installed at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 The software, hardware and settings required for the network-based unattended customized installation method for a Linux system provided by an embodiment of the present invention;

[0030] Figure 2 It is a schematic diagram of an unattended customized installation process of a network-based Linux system provided by an embodiment of the present invention;

[0031] Figure 3 This is a screenshot of an installation example of a network-based geophysical exploration software image provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to make the technical means and objectives adopted by the present invention easy to understand, the following is a specific implementation method to illustrate a method for automatic deployment of geophysical processing and interpretation software terminals from the three processes of system installation, software installation, and batch management.

[0033] (1) Network installation of Red Hat Enterprise Linux Server release 6.8

[0034] The following steps describe how to set up a boot server and installation server on a server machine with Red Hat Enterprise Linux Server release 6.8 installed, and install Red Hat Enterprise Linux Server release 6.8 on a workstation.

[0035] Step 1: Configure the startup server

[0036] The purpose of the boot server is to help start the Red Hat Linux installation program on the machine to be installed. A DHCP server and a TFTP server need to be set up on the boot server. The required hardware and software and setting conditions are as follows: Figure 1 The former is to assign an IP address to the machine to be installed, and the latter provides a way for the machine to be installed to download the boot image. The installation process diagram is as follows Figure 2 shown.

[0037] Configure the DHCP server:

[0038] 1) Install the DHCP server package (RPM package name: dhcpd).

[0039] 2) Edit the DHCP server configuration file / etc / dhcpd.conf.

[0040] The file pxelinux.0 (in the root directory of the TFTP server) will be loaded into memory and run as the boot image by the PXE client in the network card ROM.

[0041] For each additional machine that needs to be installed, add a host entry in dhcpd.conf.

[0042] 3) Start the DHCP service.

[0043] Configure the TFTP server:

[0044] 1) Install the TFTP server package (RPM package name: tftpd).

[0045] 2) Edit the TFTP server configuration file / etc / xinetd.d / tftp.

[0046] 3) Copy the Linux installer's kernel / root file system files to the root directory of the TFTP server.

[0047] 4) Copy the boot image file pxelinux.0 to the root directory of the TFTP server.

[0048] 5) Write the configuration of the Red Hat Linux version in the default file.

[0049] 6) Start the TFTP service.

[0050] Step 2: Configure NFS Server

[0051] 1) Prepare ISO installation media and place the ISO files of the Red Hat Linux installation media in the / install directory.

[0052] 2) Edit the NFS server configuration file / etc / exports to allow other machines to access the directory / install through NFS.

[0053] 3) Restart the NFS service for the new configuration to take effect

[0054] Step 3: Network boot the machine to complete the installation

[0055] Install Red Hat Enterprise Linux Server release 6.8 on the machine to be installed in an unattended manner over the network. Start the machine to be installed and choose to boot from the network card. The specific method varies depending on the BIOS version. To automatically install which version of Linux, just modify the corresponding content in the default file. The operating system should contain / soft and / data partitions.

[0056] (2) Network-based geophysical interpretation software tar package image installation

[0057] The following steps list how to install the tar package image of the geophysical interpretation software on a batch workstation.

[0058] Step 1: Select a workstation that has been installed with the operating system RHEL6.8 and software such as OpenWosks R5000, Jason9, EPOS2015, Geoscope3.1, and HRS10.3 by conventional methods as a configuration template. The workstation has been configured with the Oracle environment, system kernel configuration, network configuration, and user configuration, and is used as a configuration template after being tested by the user.

[0059] Step 2: Create software tarballs, system configuration files, and user file tarballs, including system configuration files, user home directories, cold backup tarballs after software installation, and work area path backup tarballs.

[0060] The home directory specifically includes lmu1.tgz and home.tar;

[0061] The software installation cold backup tar package includes pa_jason91.tar, pa_OpenWorks.tgz, pa_ora10g.tgz, pb_geoscope3.1_revscope3.1.tgz, hrs10.3.tar, sw_epos15.tar, and LAM5000.tar;

[0062] Work area path backup file package pe_prj.tgz;

[0063] And the rpm installation package required by the application software, put it in the rpm_rh6 folder.

[0064] Find the system configuration files and user environment variable files required by the application software, and put them in the etc folder. The specific file names include:

[0065] User profiles: group, passwd, shadow

[0066] User environment variable files: .cshrc, launcher.dat, login, csh.login

[0067] System files: services, sysctl.conf, limits.conf, Xaccess, custom.conf

[0068] Database configuration files: oraInst.loc, oratab, rc.ow.linux

[0069] Service startup files: S98paradigmFlexNetLicenseServer, lam.linux, PG_lmgrd_startup_script, S85lmgrd.jason

[0070] Custom command files: revscope, geoscope, gftp-gtk

[0071] Step 3: Write a shell script. Using shell programming, create a new inst executable file and write the shell script to be executed. The file permissions are executable. (inst script omitted)

[0072] Step 4: Build an NFS server, edit the NFS server configuration file / etc / exports to allow other machines to access the directory through NFS, and put all the prepared software tar packages, rpm packages, etc folders and inst scripts into the shared disk path. Start the nfs service.

[0073] Step 5: Install the network card and graphics card driver on the client workstation where the operating system is installed, and set the services that need to be started and stopped, as well as the necessary rpm packages. Mount the network share disk of the NFS server to the client locally, and run the inst installation script on the terminal workstation where rhel6.8 has been installed. Multiple clients can be run at the same time.

[0074] After completing the above operations, the terminal can log in to the software user and start the corresponding software after restarting, and the software license is read by connecting to the license server.

[0075] Using expect's shell programming to implement batch operations

[0076] 1) Read the password from the configuration file and call the expect command. The script content is as follows

[0077] [root@dtydb0 scripts]#cat expect.sh

[0078] #! / bin / bash

[0079] for iin`awk'{print$1}'passwd.txt`

[0080] do

[0081] j=`awk-v I="$i"'{if(I==$1)print$2}'passwd.txt`

[0082] k=`awk-v I="$i"'{if(I==$1)print$3}'passwd.txt`

[0083] expect login.exp$i$j$k

[0084] done

[0085] 2) Password file, can be multiple hosts

[0086] cat passwd.txt

[0087] 10.4.12.1 root root

[0088] 10.4.12.2 root root

[0089] 3) The actual working exp file

[0090] 3.1 Use ssh to log in and run related commands

[0091] cat login.exp

[0092] [root@dtydb0 scripts]#cat login.exp

[0093] #! / usr / bin / expect -f

[0094] set ipaddress[lindex$argv 0]

[0095] set username[lindex$argv 1]

[0096] set passwd[lindex$argv 2]

[0097] if{$argc!=3}{

[0098] puts stderr$passwd

[0099] puts stderr$username

[0100] exit 1

[0101] }

[0102] set timeout 30

[0103] #Use the spawn command to activate the ssh program. The output of the simulated terminal will be read by expect, and the simulated terminal can also be input to the remote host from send

[0104] spawn ssh $username@$ipaddress

[0105] expect{

[0106] "yes / no"{send "yes\r";exp_continue}

[0107] "password:"{send"$passwd\r"}

[0108] }

[0109] The #expect statement waits for a string match on the remote host, and when a "yes / no" is matched, the following action is performed. expect searches for the pattern "*password:", where * allows any input to be matched, so it is very effective in avoiding specifying all the details. If the remote host has no action, expect will continue to run after detecting the pattern. Once the prompt is received, the next line sends the password to the current process.

[0110] expect "]*"

[0111] send"date\r"

[0112] send "sh test.sh\r"

[0113] send "exit\r"

[0114] send "exit\r"

[0115] expect eof{exit 1}

[0116] 3.2scp file copying example

[0117]

[0118]

[0119] 3.3 Batch sqlplus operation examples

[0120] --oracle sqlplus

[0121] [oracle@dtydb0 dailycheck]$cat romote_daily_sqlplus.exp

[0122] #! / usr / bin / expect -f

[0123] set tnsname[lindex$argv 0]

[0124] set username[lindex$argv 1]

[0125] set passwd[lindex$argv 2]

[0126] set timeout 30

[0127] spawn sqlplus$username / $passwd@$tnsname

[0128] expect "SQL>"

[0129] send "select * from dual;\r"

[0130] expect "SQL>"

[0131] puts"--------talbe spaces----------"

[0132] send"@ / home / oracle / dba / scripts / showtbs.sql;\r"

[0133] expect "SQL>"

[0134] puts"--------top sql----------"

[0135] send"@ / home / oracle / dba / scripts / ash_topsql.sql;\r"

[0136] expect "SQL>"

[0137] send "exit\r"

[0138] expect eof

[0139] In addition to the above examples, expect can also be used to automate other batch operations, such as batch execution of system software installation, batch configuration of system files, database backup, hardware information reading, batch shutdown, etc.

[0140] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and concepts of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for automatic deployment of geophysical processing and interpretation software terminals, characterized in that: The unattended custom installation method of the network-based Linux system includes the following steps: S1. PXE protocol for booting the Red Hat Linux installation program from the network on a PC; S2. The network installation function provided by the Red Hat Linux installation program refers to accessing the installation media through the network; S3, the unattended installation feature provided by the Red Hat Linux installation program.

2. The method for automatic deployment of geophysical processing and interpretation software terminal according to claim 1, characterized in that: The following hardware and software resources are used to complete the automated installation: a PC machine as the boot and installation server; a PC machine to be installed, whose network card must have PXE support; a built LAN, with the above machines connected to the same subnet; and a set of Red Hat Linux installation media to be installed.

3. The method for automatic deployment of geophysical processing and interpretation software terminal according to claim 2, characterized in that: The steps to achieve automated installation include: S11, configure and start the server; S12, configure and install the server; S13. Boot the machine from the network to complete the installation.

4. The method for automatic deployment of geophysical processing and interpretation software terminal according to claim 3, characterized in that: The steps S11 and S12 need only be performed once. For each machine that needs to install Red Hat Linux, the corresponding data needs to be deployed in the startup server and the installation server, and step S11 needs to be performed once.

5. The method for automatic deployment of geophysical processing and interpretation software terminal according to claim 4, characterized in that: A network-based geophysical software image installation method is adopted. This method uses the installation template to create the software and system configuration tar package, writes a shell script, and executes the script through the nfs server. It can be installed on multiple workstations at the same time, greatly shortening the deployment time of batch workstations.

6. The method for automatic deployment of geophysical processing and interpretation software terminal according to claim 5, characterized in that: Various new and old operating systems and versions of Linux software are processed and interpreted to achieve batch deployment. The relevant system configuration, services, and variables during software installation are found, and they are modified one by one and made into template files. The implementation process includes the following steps: S31. Use conventional methods to manually install a server and install various versions of processing and interpretation software: OpenWosks, Jason, Epos, Geoeast, configure the Oracle environment, system kernel configuration, network configuration, user configuration, etc. After the installation is completed and tested by the user without any problems, it will be used as a configuration template; S32, making software tarballs, system configuration files and user file tarballs, including system configuration files, user home directories, cold backup tarballs after software installation, and work area path backup tarballs; S33, write a shell script, use shell programming, create a new executable file and write the shell script to be executed. The script needs to be tested repeatedly and confirmed to be correct before it can be executed; S34. Build an NFS server, edit the NFS server configuration file / etc / exports to allow other machines to access the directory through NFS, and put all the backed-up software tarballs, system configuration files, and user file tarballs into it; S35. Install the operating system on the client network. Use the unattended installation technology of the Linux system to batch install the client Linux operating system through the network, install the network card and graphics card drivers, and set up the services that need to be started and stopped, as well as the necessary rpm packages; S36. The client runs the installation script, mounts the network shared disk of the NFS server to the client locally and executes the software installation script. Multiple clients can perform the operation simultaneously.

7. The method for automatic deployment of geophysical processing and interpretation software terminal according to claim 6, characterized in that: Use expect's shell programming to implement batch operations. Expect is developed based on the Tcl language and is used in interactive operations and automated testing scenarios. It is used in environments where the same operations need to be performed on multiple servers, greatly improving the work efficiency of system operation and maintenance personnel.

8. The method for automatic deployment of geophysical processing and interpretation software terminal according to claim 7, characterized in that: Send, expect and spwan are the most basic commands of the Expect language. The send command sends a string to the specified process, the expect command waits for the result returned by the process and decides the next operation based on the returned string; and the spwan command can initiate the running of a process. Expect can be used to implement SSH login to run related commands, file management, database management and other operations, such as batch execution of system software installation, batch configuration of system files, database backup, hardware information reading, and batch shutdown.

Images