|
||||||
|
Output from rpm -qa, how to extract only the name of the package? If you need to migrate to a new computer and want to install the same packages that were on your old computer on the new one, the following pipes all the packages into a list that you can then use to install on the new computer. rpm -qa –qf “%{NAME}\n” > filelist.txt Then to install the packages yum -y install `cat filelist.txt` get WebLogic working with Apache as reverse proxy (mod_proxy) There is Oracle Reports running on a WebLogic Server. I want it to run on https but for some reason I don’t know it wasn’t working for the Internet Explorer 11 (and maybe other versions too). IE11 just timed out and never displayed any html nor could be establish a connection. It worked with Chrome and Firefox. With the usual Apache config for mod_proxy: … SetOutputFilter proxy-html <html> in it. <html> <html> enable „WebLogic Plug-In Enabled“ in Domain Structure > Environment > Servers > managed01 > Configuration General > Advanced RequestHeader set WL-Proxy-SSL true If you are using a WebLogic Cluster you need a proxy in front of the cluster to provide load balancing. You need either a hardware load balancer or a Web Server with the Oracle Web server plug-in. Even if you are not using a Cluster it makes sense to place a proxy in front of your WebLogic server to provide an extra layer of security. You can use the Apache HTTP Server with the mod_proxy module to configure a reverse proxy. You can of cause also use the Oracle Web server plug-in but it is a proprietary module that you have to manually download and install. mod_proxy is open source and since I mostly work with Red Hat servers it is available in the standard Red Hat channel. Red Hat provide security updates so “all” you have to do is run yum update once in a while to get the latest updates. If you use the Oracle Web server plug-in you have to manually check for updates. I often you use a solution where an Apache HTTP Server is placed in a DMZ network segment. SSL is terminated at the Apache server and mod_proxy is used to proxy requests to an WebLogic Server in the production network segment. In this example we have an Internet domain called theheat.dk. The public IP 217.116.232.220 is NATed to 10.10.10.1 on winterfell. Apache has already been installed on winterfell, the mod_proxy module loaded, SSL configured and WebLogic is running on wintermute. In your VirtualHost in the ssl.conf (httpd-ssl.conf on Windows) file you can add something like this: ProxyPass / http://10.10.10.10:8001/ Another example: ProxyPass /service/ws1 http://10.10.10.10:8001/ws1_v101 ProxyPass /service/ws2 http://10.10.10.10:8001/ws2_v300 If you need more control over what is proxied you can use the ProxyPassMatch directive. If you want to prevent a path from being proxied you can use the ! directive. Configure the WebLogic Server to use a proxy The above will proxy the requests to the WebLogic Server. In some situations it will not work though. The WebLogic Server does not know that there is a proxy in front of it and sometimes it will return URLs to the end-user that contains the server name. In this example it will return URLs that contain wintermute. Since wintermute is not known on the Internet it will fail. I have experienced this when an ADF application session timeout and asks the user to log in again. The URL returned to the user is wrong. Another example is the WSDL for a Web Service. The location of the end-point and references to XML schemas will use the host name. To remedy this you can configure WebLogic so I knows that there is a proxy in front of it. WebLogic will use this information and dynamically change the references, so they uses the proxy information. First you must enable the WebLogic Plug-In. We are not using the WebLogic Plug-In but we still need to enable it here. Domain Structure > Environment > Servers > managed01 > Configuration General > Advanced: Domain Structure > Environment > Servers > managed01 > Protocols > HTTP:
The little yellow triangles with the exclamation mark tell us that we need to restart the WebLogic server for the changes to take effect. But it will still not work. We have told WebLogic what the frontend HTTPS host and port is, but we have terminated SSL at the Apache proxy and uses HTTP between Apache and WebLogic. We do this by inserting an tag in the HTTP header in the Apache configuration: RequestHeader set WL-Proxy-SSL true If you are using a cluster you should set Frontend Host and Frontend HTTPS Port for the cluster not the individual Managed Server. For more information: ProxyPreserveHost In some situations you need to access your application from both the Internet and from an internal network segment using the internal server names. To accomplish this you need to change the Apache configuration: ProxyPreserveHost On You also need to remove the Frontend HTTPS Host from you WebLogic Server or Cluster. Now I can access a Web Service via both: The WSDL will either contain reference to theheat.dk or winterfell depending of which URL I use to access the WSDL with. Apache HTTP Server Plugin configuration to Weblogic 11g (10.3.5) 1. Install Apache HTTP server (Version 2.2.29). Note – Installation might need ‘root’ Here is the Apache reference doc for Installation (http://httpd.apache.org/docs/2.2/install.html) Download required version of Apache software distribution from nearest mirror site (Mirror I used) http://mirrors.sonic.net/apache/httpd/httpd-2.2.29.tar.gz 2. Configure ./configure –prefix=/opt/apache2.x –with-ldap –enable-mods-shared=”all ssl ldap cache proxy authn_alias mem_cache file_cache authnz_ldap charset_lite dav_lock disk_cache isapi suexec” –enable-rewrite=shared –enable-mime-magic –enable-info –enable-speling –enable-usertrack –enable-example –enable-maintainer-mode 3. make 4. make install Note – When we re-run the Apache installation with updated modules, sometimes you may run into libtool error Error : libtool: install: error: cannot install `libaprutil-1.la’ to a directory not make clean Issue#2 : you may run into below error while starting Apache instance after loaded weblogic shared Object files: [root@NA9F2RFV1 bin]# ./apachectl -k start httpd: Syntax error on line 126 of /opt/thirdparty/Middleware/apache2.x/conf/httpd.conf: Cannot load /opt/thirdparty/Middleware/apache2.x/modules/mod_wl_22.so into server: libstdc++.so.5: cannot open shared object file: No such file or directory Solution : Run below Command to install dependent libraries. sudo yum install compat-libstdc++-33.x86_64 >Installing : compat-libstdc++-33-3.2.3-69.el6.x86_64 1/1 >Installed: Issue 3: you may see below error while restart Apache. [root@NA9F2RFV1 bin]# ./apachectl -k stop Solution : you need to add ‘hostname’ to /etc/hosts file 127.0.0.1 NA9F2RFV1.localdomain localhost localhost4 localhost4.localdomain4 5. vi $APACHE_HOME/conf/httpd.conf a) Load weblogic module copy shared object file into $Apache_Home/modules directory. choose the correct .so file based on the Apache version and OS (32 bit/64 bit) you can copy these .so files from weblogic distribution ($WLS_HOME/wlserver_10.3/server/plugin/) LoadModule weblogic_module modules/mod_wl_22.so 6. Define IF module for Weblogic WebLogicHost 192.168.56.101 Configure SSL for Apache
Note – After you generate certs and configure Apache to access over https, you may see below issue in browser. ******************************* Your browser sent a request that this server could not understand. To have HTTP (non-ssl working) between apache plugin and weblogic, you need to enable this option “WebLogic Plug-In Enabled” using weblogic console under Servers > AdminServer > General > Advanced. Configure SSL between Apache and Weblogic11g (Using Trusted CA) Web Logic Server Configuration 1. Generate a private key jdk_home\bin\keytool -genkey -alias Example: /opt/thirdparty/Middleware/Oracle/jdk1.6.0_24/bin/keytool -genkey -alias sslcert -keyalg RSA -keysize 2048 -keystore mykeystore.jks 2. Generate a certificate request (CSR file). jdk_home\bin\keytool -certreq -keyalg RSA -alias Example: /opt/thirdparty/Middleware/Oracle/jdk1.6.0_24/bin/keytool -certreq -keyalg RSA -alias sslcert -file certreq.csr -keystore mykeystore.jks Paste the csr file and get the trail certificate(save as public.pem) and intermediate CA (save as intermediate.pem) and Root CA (save as rootCA.pem) from the email sent from Verisign website( http://www.symantec.com/ssl-certificates/?inid=vrsn_symc_ssl_index&searchdomain=google.com&searchterms= 4. Import Certificates into keystores: 1. Import CA into keystore: keytool -import -alias verisignCA -file CA.pem -keystore Example: /opt/thirdparty/Middleware/Oracle/jdk1.6.0_24/bin/keytool -import -alias rootCA -file rootCA.pem -keystore mykeystore.jks –trustcacerts 2. Import intermediate CA into keystore: keytool -import -alias verisignIntermediateCA -file Intermediate.pem -keystore Example: /opt/thirdparty/Middleware/Oracle/jdk1.6.0_24/bin/keytool -import -alias intermediateCA -file intermediateCA.pem -keystore mykeystore.jks –trustcacerts 3. Import the public key into your keystore. It will go on the same alias as the private key: keytool -import -alias Example: /opt/thirdparty/Middleware/Oracle/jdk1.6.0_24/bin/keytool -import -alias localhost -file public.pem -keystore mykeystore.jks –trustcacerts 4. To view the keystore: keytool -list -keystore mykeystore.jks -v 5. keystore configuration in weblogic console. From the Admin console, go to your server page, and in the Keystore & SSL tab choose: Custom Identity and Custom Trust Custom Identity Custom Identity Key Store Type: jks Custom Trust Custom Trust Key Store Type: jks Custom Trust Key Store Pass Phrase: Go to SSL TAB : Private Key Alias: Restart your server and now try https://localhost:7002/console Error You might notice below error in Apache error log while accessing https: apache url [Wed Dec 10 13:55:26 2014] [error] [client 192.168.56.1] ap_proxy: trying GET /console at backend host ‘192.168.56.101/7002; got exception ‘WRITE_ERROR_TO_SERVER [os error=0, line 806 of ../nsapi/URL.cpp]: ‘ Solution : Convert this .crt into .pem openssl x509 -inform der -in MyWeblogicCAToTrust.cer -out MyWeblogicCAToTrust.pem Configuration Complete!! Apache-Weblogic SSL Cannot load /u01/wls_plugin/lib/mod_wl_24.so into server: libopmnsecure.so: cannot open shared object file: No such file or directory I’ve configured weblogic proxy plugin 12c for linux according to documentation but I run into “Cannot load /u01/wls_plugin/lib/mod_wl_24.so into server: libopmnsecure.so: cannot open shared object file: No such file or directory” The error: [root@localhost ~]# apachectl restart My settings: [root@localhost ~]# vi /etc/httpd/conf.d/whatever.conf — LD_LIBRARY_PATH=/u01/wls_plugin/lib This solution worked for me: [root@localhost ~]# vi /etc/ld.so.conf [root@localhost ~]# ldconfig [root@localhost ~]# apachectl restart Configure Apache with multiple weblogic server instances Instance A – http://test_server:7001/App1(Domain I) I have App1 deployed on instance1 which is running in domain I. Similarly App2 deployed on instance 2, app3 deployed on instance 3. Needless to say all domains running in different ports(non-clustered environment). Now I would like to configure my Apache to forward request to respective URLs based on what do they type in the URL. If they type http://test_server/App1 should go to instance A. When I use #### for Instance 1/App1 #### for Instance2/App2
#### for Instance3/App3 [root@clusterserver1 ~]# yum repolist [root@clusterserver1 ~]# yum group list vim-filesystem x86_64 2:7.4.160-1.el7 base 9.6 k Transaction Summary Total download size: 692 M Next step is to change boot mode to graphical using graphical.target # systemctl set-default graphical.target CentOS 6.8?nginx 1.9.14?mysql 5.6.23?php 5.6.17 service iptables stop vi /etc/sysconfig/selinux yum -y install gcc gcc-c++ autoconf libjpeg libjpeg-devel libpng libpng-devel freetype freetype-devel libxml2 libxml2-devel zlib zlib-devel glibc glibc-devel glib2 glib2-devel bzip2 bzip2-devel ncurses ncurses-devel curl curl-devel e2fsprogs e2fsprogs-devel krb5 krb5-devel libidn libidn-devel openssl openssl-devel openldap openldap-devel openldap-clients openldap-servers make libtool* git tree bison pcre-devel perl gd gd-devel mkdir software wget http://ftp.gnu.org/pub/gnu/libiconv/libiconv-1.14.tar.gz tar xf libiconv-1.14.tar.gz wget http://heanet.dl.sourceforge.net/project/mhash/mhash/0.9.9.9/mhash-0.9.9.9.tar.gz wget http://downloads.sourceforge.net/project/mcrypt/Libmcrypt/2.5.8/libmcrypt-2.5.8.tar.gz?r=https%3A%2F%2Fsourceforge.net%2Fprojects%2Fmcrypt%2Ffiles%2FLibmcrypt%2F2.5.8%2F&ts=1468310750&use_mirror=freefr tar xf libmcrypt-2.5.8.tar.gz wget http://pkgs.fedoraproject.org/repo/pkgs/mcrypt/mcrypt-2.6.8.tar.gz/md5/97639f8821b10f80943fa17da302607e/mcrypt-2.6.8.tar.gz ln -s /usr/local/lib/libmcrypt.la /usr/lib/libmcrypt.la wget https://cmake.org/files/v3.6/cmake-3.6.0.tar.gz wget http://pkgs.fedoraproject.org/repo/pkgs/community-mysql/mysql-5.6.30.tar.gz/md5/ac8ba1db4454d2c144c7d892185a9328/mysql-5.6.30.tar.gz tar -xf mysql-5.6.30.tar.gz mkdir -p /usr/local/mysql cd mysql-5.6.30 cd /usr/local/mysql vi /etc/my.cnf [mysqld_safe] # CLIENT # [mysqld] # GENERAL # # MyISAM # # SAFETY # # DATA STORAGE # # BINARY LOGGING # # REPLICATION # # CACHES AND LIMITS # # INNODB # # LOGGING # cp support-files/mysql.server /etc/init.d/mysqld drop user ”@localhost; wget http://nginx.org/download/nginx-1.9.14.tar.gz groupadd -r nginx mkdir -pv /var/tmp/nginx/client nginx vim /etc/rc.d/init.d/nginx #!/bin/sh # Source function library. # Source networking configuration. # Check that networking is up. nginx=”/usr/local/nginx/sbin/nginx” NGINX_CONF_FILE=”/etc/nginx/nginx.conf” [ -f /etc/sysconfig/nginx ] && . /etc/sysconfig/nginx lockfile=/var/lock/subsys/nginx make_dirs() { start() { stop() { restart() { reload() { force_reload() { configtest() { rh_status() { rh_status_q() { case “$1″ in chmod +x /etc/rc.d/init.d/nginx ?PHP cp -frp /usr/lib64/libldap* /usr/lib/ php PHP php-fpm php-fpm web LNMP vim /usr/local/nginx/html/test.php nginx http://ip/test.php,LNMP $ make test rpm -qa | grep tcl yum install tcl Permit SSH Login for Root in Oracle Solaris 11 Open Terminal window and switch to root user. 1. Change the file /etc/ssh/sshd_config PermitRootLogin yes with PermitRootLogin no and save file. vi /etc/ssh/sshd_config PermitRootLogin yes 2. Comment out the “CONSOLE=/dev/console” line in /etc/default/login. vi /etc/default/login #CONSOLE=/dev/console 3.Remove “;type=role” from the root entry in /etc/user_attr or use the below command. rolemod -K type=normal root 4. Restart the Services. #svcadm restart svc:/network/ssh:default 5.Try SSH connection using root user You should be able to connect. So you’ve got a fresh installed Solaris 11 box – and you can’t login directly as root because it is a role… db04 console login: root To change this, then you need to make root a user instead of being a role Login as an administrator account and su to root As you can see in the configuration file root is a role To change it root@db04:~# rolemod -K type=normal root fallocate -l 4G /swapfile swapon /swapfile echo “/swapfile swap swap sw 0 0” >> /etc/fstab The problem with fallocate(1) is that it uses filesystem ioctls to make the allocation fast and effective, the disadvantage is that it does not physically allocate the space but swapon(2) syscall requires a real space. Reference : https://bugzilla.redhat.com/show_bug.cgi?id=1129205 I had faced this issue earlier with my box too. So instead of using fallocate, I had used dd as per the link suggests |
||||||
|
Copyright © 2026 - All Rights Reserved Powered by WordPress & Atahualpa |
||||||
Recent Comments