Tell me about Yourself

(Q)Tell me about yourself:

Ø  I'm Raghu from Bangalore working as a middleware administrator in Trigent Software

Having 3.5 years of experience.

Ø  I've been working with multiple middleware technologies like Oracle Web Logic server, JBoss, tomcat and apache.

Ø  Done the Installation & Configuration and hands on experience in applying the security patches.

Ø  Done Weblogic (10.3.3 to 10.3.6) and JDK (jdk1.6.0_201) to (jdk1.7.0_99) up gradations, configured domains, clusters and machines.

Ø  Involved in OS patching (Quarterly patches released by REDHAT Linux) activities, Database refresh activities and will do the production changes in maintenance window time.

Ø  I've generated the csr and key files, provided them to the Security team to get the SSL Certificates.

Ø  We usually provide the csr files to the company like digicert and they would provide us the Root certificate along with the renewed (Extending the certificate validity time) certificate file.

Ø  Once we receive the Certificate File and Root CA certificate, we will load the renewed certificate file, Key file and Root CA certificate file in apache config (httpd.conf).

Ø   We will also configure the SSL Cipher suites in the Apache SSL section along with the TLS and SSL protocols.

Ø  Since the SSLv2 and SSLv3 are vulnerable, we have configured the TLS protocols.

Egg: TLSv1.1 and TLSv1.2

Qns:-

***

1) What is your role in UNIX OS PATCHING?

Ø  We will stop the applications or the managed servers and other processes

Running on the servers, which are needed for patching activity.

Ø  Patching will be applied by UNIX team and they will reboot the VMs after the patching activity.

Ø  Once the servers are up, we will check the uptime and then we will see what services are up along with the server reboot.

Ø  Then we will start the remaining services on the server.

Ø  Auto Reboot: We can setup the auto reboot of services under /etc/init.d/

Ø  Once the services are started on the server, we will validate to make sure all the services are up and accessible.

Ø  The application team/dev team also validates the applications and its functionality.

Ø  Tools: For AppDynamics and Splunk we will have monitoring agents

Installed/configured on our Virtual Machines.

Ø  For AppDynamics we will also configure the hosts in Managed Server Startup arguments.

***

2) What is your role in Database Refresh Activities?

Ø  We will stop the applications/servers that are related with the specific data base

That is getting refreshed and when the refresh is completed, we will start all the applications/servers which are down.

Ø  Validation will happen as usual.

3) What is a cluster?

Cluster is nothing but group of managed servers on a single machine or on multiple machines.

There are two types of clusters:-

1. Horizontal clustering

2. Vertical clustering

Horizontal Clustering:-

In this we will have a managed server Configured on one server and other managed server on another machine.

Vertical Clustering:-

 We can configure the cluster within the same virtual machine.

Benefits of Clustering:-   (Fail over, Load Balancing, Scalability, High Availability)

1.Fail over:

When a request is coming to one server and if the servers fails

(Goes down) after some time, the request will be sent to another server

On the same cluster and continue to serve the request without fail.

2.Load Balancing:-

When the load is high on one server,

The next request will automatically go to the next available server

On the cluster (uses Round Robin Algorithm).

(The request will be routing based on the algorithm we use.

In our case we are using Round Robin Algorithm.

That means the first request will go to the first server,

And 2nd request will go to the 2nd server,

3rd request will go the 3rd server

And again 4th request will go to the 1st server

Where we have 3 managed servers configured in a cluster.)

3. Scalability:-

When a cluster is up, and you may need an additional server

to serve further requests since the existing servers in the cluster are busy

In serving the requests and can't serve any further new requests,

Then we can create a new managed server and add it to the cluster

Without disturbing the existing servers in the cluster.

4. High Availability: -

The application should be available on as many servers as possible on the cluster

So that the application will be available 24*7 to the customers.

3) How do you deployment on a cluster? Or how do you do deployment in web logic?

Ø  We will do the deployments in staging mode.

Ø  We usually create a folder called "upload" under the admin server directory and place the .war files and .ear files, through web logic admin console we will click on "Lock & Edit".

Ø  And go to deployments section in the right pane of the console.

And click on "Install", will choose the application from the upload location,

And target that deployment to a specific cluster and proceed further and then we activate the changes.

Ø  Once the changes are completed we will start the deployment.

Ø  If we are doing a deployment for the first time, and the server is in running state, the state of the application would be "prepared" and then we will just need to START the application.

Ø  Then it will go to active state.

Ø  If we are doing the re-deployment (a new version of the existing application),

Then we will just place the new application on the upload folder,

By taking the back up of existing file.

Ø  And then will restart the cluster for which that application is already targeted.

Ø  When the managed servers are restarting it will take a copy of the new application

From the upload location (which is on the admin server)

To the "stage" directory of this managed server wherever it resides.

4) How do you configure domains on multiple servers?

Ø   We will install the web logic server on all the machines.

Ø  Then we will pack the domain on the first machine (Where we have the admin server running).

And will unpack on all other machines.

Ø  Before we unpack the domain, we will create a folder structure,

Until the domains directory (/apps/oracle/Weblogic 10.3.6/user_projects/domains)

In all machines.

***

5) Why the disk space will be full suddenly and what happens?

Ø  If there is any debugging enabled at application level

Or server level or log level, the disk space will be full.

Ø  The server may hang up (\Ex: /apps mount or file system)

If it reaches 100%.

6) What is WLS in Weblogic?

Ø  It manages system-level details so you can concentrate on business logic and presentation.

Ø  J2EE Platform. Weblogic Server contains Java 2 Platform, Enterprise Edition (J2EE) technologies.

Ø  J2EE is the standard platform for developing multitier enterprise applications

Ø  Based on the Java programming language.

7) Is Weblogic a middleware?

Ø  Oracle Weblogic Server forms part of Oracle Fusion Middleware portfolio and supports Oracle, DB2, Microsoft SQL Server, and MySQL Enterprise and other JDBC-compliant databases.

Ø  Oracle Weblogic Platform also includes: JRockit, a custom JVM.

8) What is the difference between Weblogic and Tomcat?

Ø  Key Difference: Tomcat and Weblogic are two servers.

Ø  Tomcat is a web server from Apache Software Foundation,

Whereas Weblogic is an application server from Oracle.

Ø  Tomcat is capable of running servlets and JSP,

Tomcat is a web container that runs the web applications, based on servlet and JavaServer pages.

9) Can we create multiple domains in web logic?

Organizing Domains: -

Ø  You can use a single Oracle Weblogic  Server installation, to create and run multiple domains

Ø  Or you can use multiple installations to run a single domain.

10) What is Web logic Server?

Ø  Weblogic Server contains Java 2 Platform, Enterprise Edition (J2EE) technologies,

J2EE is the standard platform for developing multitier enterprise applications,

based on the Java programming language. ...

Ø  J2EE applications are based on standardized, modular components.

***

11) What is Web logic Node Manager?

Ø  Node Manager is a Weblogic Server utility that enables you to start, shut down, and restart Administration Server and Managed Server instances from a remote location.

Ø  A Node Manager process is not associated with a specific Weblogic domain, but with a machine.

***

12) What are managed servers in Weblogic?

ü  Domains include a special Weblogic Server instance called the Administration Server, which is the central point from which you configure and manage all resources in the domain.

ü  Usually, you configure a domain to include additional Weblogic Server instances called Managed Servers.

13) How do I change my Weblogic Admin listening address?

To do so, you use Oracle Weblogic Server Administration Console:

1.       In the Change Center, click Lock & Edit.

2.       In the Domain Structure section, expand Services, then JDBC, and select Data Source…..

3.       Select the data source you want to change…..

4.       Select the Connection Pool tab.

5.       To change the IP address, modify the URL field. For example:

jdbc:oracle:thin:@hostname.domainname.com:1522/orcl

Clicks save.

6.       Restart the servers that use this data source.

7.       (Click the Target tab to see the servers that use this data source.)

***

14) What is domain in Weblogic?

Ø  A domain is an interrelated set of Weblogic Server resources Managed as a unit.

Ø  A domain includes one or more administration servers and managed servers.

Ø  Various clients use the administration server to configure the system.

Ø  The managed server is used to run actual applications.

15) What is admin server in Weblogic?

Ø  The basic administrative unit for a Weblogic Server installation is called a domain.

Ø  A domain is a logically related group of Weblogic Server resources that you manage as a unit.

Ø  A domain always includes only one instance of Weblogic Server, Called the Administration Server.

***

16) How do I start Weblogic admin server?

Start Weblogic administration console:-

Ø  If Weblogic Admin Server is not already running, from a command prompt,

Go to the [appserver root]\user_projects\domains\[domainname] directory and enter the following command: ...

Ø  On the login screen, type your administrator user name and password, and click Log In.

17) How do I start/stop manager server from Weblogic admin console?

To start or stop servers by using Oracle Weblogic Administration Console:-

1. Log in to the Oracle Weblogic Server Administration Console.

2. On the left pane, under Domain Structure, select Environment, Servers.

3. On the right pane, under Summary of Servers, click the Control tab.

Select the server name.

18) Can we start managed server without Node Manager?

ü  You can't start or stop a Managed Server with WLST without first connecting to the Node Manager.

ü  However, you can use WLST commands to start the AdminServer without the Node Manager running.

***

19) How do I check my Node manager status?

To monitor Node Manager Status:-

1.       In the Administration Console, expand Environment and select Machines.

2.       Click the name of the machine where you want to view Node Manager Status.

3.       Click the Monitoring tab in the right pane.

4.       To view the contents of the Node Manager Log, click the Node Manager Log tab in the right pane.

20) How do I add a managed server in Weblogic?

To create a Managed Server in an existing domain:-

1. If you have not already done so, in the Change Center of the Administration Console, click Lock & Edit (see Use the Change Center).

2. In the left pane of the Console, select Environment > Servers.

3. In the Servers table, click New.

4. On the Create a New Server: Server Properties page:

5. Enter the name of the server in the Name field.

6. Listen Address: Your hostname

7. Listen Port: specify some port number

(Enter the port number from which you want to access the server instance.)

8. Specify whether or not this server will be a standalone server or will belong to a cluster.

9. To activate these changes, click Activate Changes.

10. Result:

The Administration Console updates the domain's config.xml file

With the new server configuration data.

21) How do I stop a managed server in Weblogic?

To start or stop the managed server using Oracle Enterprise Manager Console:-

1. Log in to the Oracle Enterprise Manager Console.

2. Navigate to Weblogic Domain, Domain Name, SERVER_NAME.

3. Right click, and navigate to Control.

4. Click Start Up to start the server. Click Shutdown to stop the server.

22) How many Types of Node Manager are there?

There are two types:-

1. Java based

2. Script Based

***

23) What is Weblogic console?

About the Administration Console:-

1.       The BEA Weblogic  Server Administration Console is a Web browser-based on,

Graphical user interface that you use to manage a Weblogic Server domain.

2.       A Weblogic  Server domain is a logically related group of Weblogic  Server resources,

That you manage as a unit.

What is the difference between an Alert and Incident?

Incident happens if there is only one issue/injection for business.

If there is a production incident the tool any application or failure that is affected in production environment and causing a problem so that is an incident.

Alert is we set the alerts.

Egg: Disk Space usage, Memory Usage for all that.

There is an alert for Disk Space and Disk Space is affecting your file system so it should not reach 100%.

So that’s why we set thresholds.

Disk Space – 85% = warning

Disk Space – 95% = critical

When the usage reaches 85 we will get an alert saying that this file system is consuming please take an action.

Immediately we will login to the server and verify the Disk Space what is consuming the disk if the logs are full and consuming more space we can remove the old log or archive log etc.

If there is any data that has not been used for long time or a temp files some back files which are not required those you can remove with that we can clear the Disk Space.

For memory alert what is the current consumption of the memory,

And if it is more than 90% of course it is good it still you used low memory which is available but at the same time may not exceed whatever is the limit.

We should find out which process is consuming more memory and we should reach out to the respected team or application team saying that this application is consuming more memory please let us know if we have to restart this program it is memory consumption on the server that is the final case whenever there is one issue reported then only you have to do this otherwise you should not restart any application.

So this is about the memory alerts, disk space alerts and there would be some alerts which are:

Let say it is for weblogic you will get some stuck threads whenever it starts 600 miili secs if it reaches more than that then it gives time out actually. So when you get more no of stuck threads on the server.

Probably you have to restart or it will automatically resolve by itself through stuck threads.

It continuous to increase the no of stuck threads then you may get An OutOfMemory also.

Most of the time OutOfMemory is heap space issue or it can be a Perm Gem space issue or in some cases you may get NativeMemoryError.

Heap Space Issue:-

YOU would recommend the Dev team or App team to increase the heap size Xmx or Xms values.

That is the resolution for memory consumption whatever it is.

If it is happening every time then we have to check the oracle support for weblogic.

They will guide you what is causing this issue some recommendation they would give you have to follow that.

Alert can be Incident why?

If we do not take an action when an alert is come an alert would be warning or critical whatever it could be .

So when the alert is coming you have to take an action lets say have not taken any action it is waning and it reaches critical then the critical alert will be triggered If you don’t take an action even it is critical alert then it will lead to incident.

Egg: File System Usage is 100%

Then the system hangs it.

At that time you should turn into the incident where you have to work out along with the Unix Team reboot the server and you have to clear the space .

So this difference between an incident and alert.

Incident means there is an issue and it is causing against the business .

If it is a critical incident or major incident that means business is affected very badly there would be a financial loss or there could be customer satisfaction during that particular time so that time it would be a very high incidents or a critical incident .

And for any critical incident or any high incident the bridge call will be opened everybody has to join there.

What is everybody?

First of all they will call the System Admins that means system is maintained or middleware team ,

So middleware team will check what is the server, what is the application, where it is ending is there any errors in the logs they will check everything .

And if system reports that there is some error in the logs and it is reporting that the oracle account in the log or there is some ORA Error then the system may suggests we have seen an error related to the database for ex it is oracle database.

We have to tell them:-

Please involve the Oracle Database Team

If it is mysql Error you should tell the:-

Please involve the MySQL Database Team.

So then they will join the bridge call and publish the issue.

Or it is network issue you see something relate to network firewall or network bridge then you will inform the there is network connectivity issue please involve the network team.

And we have to give stat updates, you know the issue and you are fixing let’s say you have to restart something tell them I’m going to restart this particular server , when you are stopping and starting whenever you are doing something you have to make a notes every time you should give you dates to the main team or AOC .

What is the change?

Change is nothing but a change to infrastructure or a configuration file or a set up anything can be a change and all the production changes would be done along with the cab approval.

What is cab? – Change Advisory Board

Change allegory board so that board that manager has to approve your change so that we can perform this change in a particular maintenance window.

Let yours say in some companies they follow the weekly cab.

Every Monday there is a cab.

And everybody will join there we have to present our cab we hold assign the reports we have to tell the justification also what is the change you are doing in weblogic we are performing some change from weblogic.

They will ask you why you are doing on production.

No this is done in non-production environment and what is the successful of art all these things they will ask?

You should provide them a sign of email for the testing team whenever it is that change we are performing in the lower environment on it is successful so that we want to fix this in the production environment also please approve.

Then change management will ask if there any other change along with this on it is going to affect anything.

No one says you don’t have an objection then they will approve the change.

In some cases let yours say from database change will be there then the database administrator will tell we have a change for the foreign client proposal at the same time there might be some deployment plan.

So you have to inform them so we will at least let yours know when you are going to finish it from then we will start your activity or you have to postpone your deployment to next week or something.

Because your deployment depends on database you upgrade or whatever so from change happening on the database side it should not affect you so it’s always not better to do both changes at same time.

Now you are aware of change management process.

What is problem management?

Problem management is nothing out if there is an incident from OutOfMemory issue happened and something happened and it keep happening very frequently so proficient management team will tell that  these is reoccurring issue and you should have a problem payout you for this?

What they will do they will a problem ticket to you and you have to work on it.

So many qns like: -----

What is the issue, what is the resolution, Is there a permanent fix, when you will get to you date the permanent fix or if you don’t have a fix how to proceed for that are you going to work with any vendor to get this fix or any development team is there who can fix this issue?????

So once the problem ticket is created it will be appended to the respective team who is going to work on a permanent fix so the problem ticket will be created only when there is reoccurring issue.

An issue happening very frequently.

Now you were aware of Problem Management, Incident Management, Change Management and what an alert is also .

How do apply a patch?

I will download the patch for the oracle support related to the you whatever you suggested once I download it I will place the downloaded bundle and there will be a directory they will not directly create a cache directory on their bsu then I will create one then I will place the war file on the xml file.

You Should have two files on the cache directory:-

One is patch jar file and another one is cat log file (xml file)

This is the info about the particular patch.

We should remove the readme.txt file from that place.

Once you create these files on the cache directory then we will you use the bsyou.sh for the patching and will give the patch path and then be patch list.

And If I want to see how the patching works would I will give –verbose.

And there is another command once we applied the patch we can see what are the patches applied on the server.

And also there is a command to remove the patch if a new patch is released we will remove the existing patch and apply the new patch.

Why can’t you keep the existing patch and applying the new patch?

Oracle always recommends that you cannot do this you have to remove the old patch and apply the new patch.

This is about patch.

What is the CSR?

CSR is Certificate Signing Request

So on any server which opens ssl command we will generate the csr and key files once you generate the csr and key files they will provide the csr file to the security team or to the team who will provide the certificate.

Once it is approved whatever the vendor it is they will give the certificate along with the Root CA Certificate so now we will have The Root CA Certificate, CRT certificate and the key file.

We will create these three certificates or the three files on the Apache we will place these three files in the pack and then we will load it please keep your renewed certificate file and Root CA Certificate file under the httpd.conf file in the ssl section then they will mention all these files.

So that is the procedure for certificate renewal.

How do you create the data source?

We create the data sources from the web logic console,,,,,

We will go to the messaging > create generic data source >new

Then we click on new it will show you few options that is generic and multi data source we will select the generic data source and we require the database name, database hostname, database YOUN & PWD.

Along with JNDI name and data source name and which driver you have to select

If it is XE data source we have to select the jdbc XA.

And if it is non-XA data ource you have to select oracle thin non-XA .

And there are two types:-

One is Service Connection and instance Connection

We always go with the service connection for better performance.

So this is how we create the data source.

And we are creating the data source for different database

For oracle we will select the database as oracle and then we will select the driver.

If it is MySQL database yes you can have the drop down list in the drop down list you can see the MySql database you should select the MySql database and then proceed to select the MySql Driver there you can have the drivers along with that you proceed.

So like this we create the data soyource for any database.

These are the things You shoyould have to create the data soyource.

Db name

Db hostname

Db: - YOUN & PWD

JNDI name

And then targets which you have to create the data source for XA.

There is a cluster like c1.

You are creating the data source and you are targeting that data source to a particular cluster.

And when you are creating the data source if you want to check let yours check the checked connection is successful or not there is an option.

So You can click on the Test connection then it is showing succeeded msg that means you entered Un & pwd is correct.

If the pwd is wrong it will say the connection is refused or something invalid/username or password.

Then you have to correct it.

This is about the data source.

How do you deploy the web application in weblogic?

We will place the war file or ear file under the upload folder of AdminServer from there we will choose the application from the console and deploy an application on to the respective cluster.

Once we done the installation the application would go to prepared state for the first time when we are doing the deployment it will go to the prepared state so we have to start that application it will go to active state so once that deployment is done you can check the application by going to deployments tab or setting tab and then click on the link.

That is about the deployment.

The deployment on tomcat server what you do?

You will just place under webapps folder under that you place this war file and then you restart the tomcat server.

So that application will deploy.

For JBoss you can just go on to the console one you login to the console you can see the deployments choose the file and place it on the respective cluster and then activate That’s All.

So In JBoss you have two options:-

Enable and Disable the Application even through the application is targeted to particular cluster when you disable it goes to inactive state and when you enable it, it goes to active state.

And the data source creation in weblogic same like web logic.

How do you do the Apache and web logic integration?

We will copy the mod ssl22.so file from the weblogic modules folder to Apache modules folder and then we will load the module in the httpd.conf file along with that you have to provide cluster info in the web logic forward information you have to give the location path if required and if there is any specific request that have to go to https we have re-writhe the condition on re-write rule.

So that is how you do the integration.

One is loading the module placing the module in modules folder and then we have to configure the web logic forward info you will provide the listen address of web logic server and the port number if you have a cluster environment you will provide the listen address of particular manage server and port number,  

Listen address of second manages server and port.

So this is about the web logic and Apache integration.

A WebLogic domain is a fundamental administrative unit for WebLogic Server. It comprises one or more WebLogic Server instances with their resources, which are collectively managed and configured using a single Administration Server.

WebLogic Server instances are referred to as Managed Servers, in which Java Enterprise Solutions are deployed. These Managed Servers could be grouped into clusters for load balancing and failover, especially for critical applications. Multiple WebLogic domains could also be set up according to application boundaries, system administrator responsibilities and server’s geographical location. Another alternative is to set up a single domain to manage all WebLogic Server administration activities.

The WebLogic domain infrastructure consists of three parts, as illustrated below:

·         Administration Server: Guides the actions of managed servers

·         Managed Servers: Stand-alone servers ruled by the Administration Server

·         Clusters: Groups of managed servers managed by the Administration Server

Sample WebLogic Domain Infrastructure

1.    Administration Server: Each WebLogic domain must have one server instance set as the Administration Server. This server performs all the configuration changes of the domain and deployment of applications. The Administration Server centralizes domain management and access provision for all the WebLogic Server administration tools listed below:

·         Console: This is the Administration Server’s GUI.

·         Security

·         Application Deployment

·         Domain connectivity

2.    Managed Servers: These are all server instances that exist in a WebLogic domain apart from the Administration Server. They host application resources and components that are deployed and managed as part of the domain. In a WebLogic domain with only one server, that server will be used as both the Administration Server and the Managed Server.

3.    Clusters: A Cluster is more than one Managed Server grouped together to provide high availability and scalability for applications. Clustering is important to provide failover whenever a server instance is unavailable. Clustered servers can be run on one machine or multiple machines, but clients perceive the cluster as one single WebLogic Server instance.

The WebLogic infrastructure offers a single, high-performance, highly integrated, and reliable database management solution. WebLogic Domain provides a single development and runtime environment that combines common application infrastructure with user-friendly robust management and an application development framework for portal initiatives, custom development and enterprise application integration.