Redeploy in cluster in production environment

There is a cluster consisting two node in a production environment. The purpose is to make it possible, to upgrade a deployed application to a new version, with the highest avaibility. Is it possible to do it faster than the standard Redeploy method? Couse it lasts an hour at least, and during that time the application cannot be accessed.
Ideal would be that it could be done without service breakdown. /On one node the application keeps running, on the other at first it is redeployed, than the while the second is restarting the former stops and then that is redeployed too./
Is it Possible? Or any other way to solve that?

Hello Mathers,
          Why does it take an hour to redoply your app, are you precompiling JSPs on startup or do you have a vast number of EJBs, lots of connection pools with connections that are all created on deployment etc?
          Can you explain your deployment process and are you using WLS staging?
          A redeployment of the application I am currently working on (three ear files approx 100 EJBs and 3 web apps) takes about 15 minutes on each managed server, so we have about 15 minutes of down time. But I think I can get this down to 5 minutes without too much trouble.
          cheers
          Hussein Badakhchani
          www.orbism.com

Similar Messages

Create a new portlet in Deployed Portal in production environment

are there any ways to import or create a new portlet in Deployed Portal in production environment?
Edited by: user8322798 on May 1, 2011 7:26 AM

This can done via WSRP proxy portlets and streaming desktops. First, you'll need to have a WSRP producer setup somewhere. This could be another WLP webapp with portlets, or another server altogether. Or, you can use the JSR 286 WSRP import tool from within the Portal Administration Console (I think it's under Services | WSRP | Import Tool) -- this will allow you upload .war(s) of JSR 168 or 286 portlets, which will be turned into WSRP producer(s).
Then, you can use the Portal Administration Console to register a WSRP Producer, and then add portlets from the producer to your desktop (http://download.oracle.com/docs/cd/E15919_01/wlp.1032/e14235/chap_fed_books_pages.htm#FPGWP690). Additionally, once the producer has been registered in Portal Adminstration Console, an adminstrator user can use the Dynamic Visitor Tools from within the streaming desktop itself to add wsrp proxy portlets to the desktop (http://download.oracle.com/docs/cd/E15919_01/wlp.1032/e14243/dvt.htm#PDGWP691).
It is not possible to add new local .portlet files to a deployed application in production mode. This requires adding the file artifacts to the .ear/.war and redeploy the application.
Greg

Best practice for a deplomyent (EAR containing WAR/EJB) in a productive environment

Hi there,
I'm looking for some hints regarding to the best practice deployment in a productive
environment (currently we are not using a WLS-cluster);
We are using ANT for buildung, packaging and (dynamic) deployment (via weblogic.Deployer)
on the development environment and this works fine (in the meantime);
For my point of view, I would like to prefere this kind of Deploment not only
for the development, also for the productive system.
But I found some hints in some books, and this guys prefere the static deployment
for the p-system.
My question now:
Could anybody provide me with some links to some whitepapers regarding best practice
for a deployment into a p-system ??
What is your experiance with the new two-phase-deploment coming up with WLS 7.0
Is it really a good idea to use the static deployment (what is the advantage of
this kind of deployment ???
THX in advanced
-Martin

Hi Siva,
What best practise are you looking for ? If you can be specific on your question we could provide appropriate response.
From my basis experience some of the best practices.
1) Productive landscape should have high availability to business. For this you may setup DR or HA or both.
2) It should have backup configured for which restore has been already tested
3) It should have all the monitoring setup viz application, OS and DB
4) Productive client should not be modifiable
5) Users in Production landscape should have appropriate authorization based on SOD. There should not be any SOD conflicts
6) Transport to Production should be highly controlled. Any transport to Production should be moved only with appropriate Change Board approvals.
7) Relevant Database and OS security parameters should be tested before golive and enabled
8) Pre-Golive , Post Golive should have been performed on Production system
9) EWA should be configured atleast for Production system
10) Production system availability using DR should have been tested
Hope this helps.
Regards,
Deepak Kori

MSDN and Production Environment (again)

I started this on another forum before I found this one, but this seems a more suitable place.
The definition of "production environment" seems rather odd. In some responses on this forum it appears to refer to "soft" systems whereas the latest MSDN licence refers to environment and physical kit.
Below is a conversation I had over email with someone from MSDN and I find the whole thing utterly bizarre. I cannot for the life of me see how this helps anyone apart from MS being able to charge for non-production software. It renders having a powerful desktop
for local lab experimentation pointless as you're not allowed to install anything and effectively doubles the hardware cost to small companies if they have to buy a separate server for any testing work (yes, best practice and all that, but budgets...) or pay
out for a Windows Datacenter licence.
Question:
“If a physical machine running one or more virtual machines is used entirely for development and test, then the operating system used on the physical host system can be MSDN software. However, if the physical machine or any of the VMs hosted on that physical
system are used for other purposes, then both the operating system within the VM and the operating system for the physical host must be licensed separately.”
Is this actually saying that if I have a physical server licenced with a purchased (not MSDN) Server 2012 R2, running Hyper-V with, say, a production file server VM on it, that ALL Windows VMs on that machine must have purchased licences even if they
are only for development & testing purposes?
Is this saying that all production and development Windows VMs must be only completely separate hardware, cluster, SAN, etc otherwise you must pay for full licences for the VMs?
Or does it just mean that the bare metal licence (plus any additional ones required for running further production VMs) must be purchased if the VMs are a mix of production and development?
Answer:
We kindly inform that any products licensed under the developer tools model (e.g. SQL/BizTalk developer and/or MSDN) must be installed on their own separate physical hardware.
You are not allowed to run test or development products on a server where production workloads are running at the same time. Kindly run your developer software on a device/host that is dedicated to testing and development.
Explanation:
The Product Use Rights (PUR) say that the developer software is not licensed for use in a production environment. Even if the PUR does not have a separate definition of production environment, a production environment is a set of resources for network, physically
dedicated hardware and software to provide "live" service. If the intent was to say that the same physical server could be used for both development and production - it would say "not licensed for use in a production OSE," instead
it says environment.
See current PUR, page 51:
Developer Tools (User License)
You have the rights below for each license you acquire.
# You must assign each license to a single user.
# Each Licensed User may run an unlimited number of copies of the Developer Tools software and any prior version on any device.
# The Licensed User may use the software for evaluation and to design, develop, test, and demonstrate your programs. These rights include the use of the software to simulate an end user environment to diagnose issues related to your programs.
# The software is not licensed for use in a production environment. # Additional rights provided in license terms included with the software are additive to these product use rights, provided that there is no conflict
with these product use rights, except for superseding use terms outlined below.
Question:
Classifying an entire physical infrastructure as "production" in these days of virtualisation and shared storage really does not make any sense at all. Not using the software for production purposes makes perfect sense, but not being able to locate
it alongside production OS installs is mad. Does this only apply to the server running the VM (CPU and RAM)? If the VHDX is hosted on shared SAN storage does the SAN have to be dedicated to non-production storage?
Answer:
We kindly inform that after double-checking the case we would like to confirm the development software cannot be run on the same hardware with production software.
We have also received a feedback from the responsible team regarding your request about a dedicated SAN (Storage Area Network) for MSDN software.
They have confirmed that the SAN has to be dedicated to the development and testing environment if it is used to run the software acquired through MSDN.
Question:
OK, so if I have my desktop (which is a production environment as I use it for email and other day to day office tasks), can I turn on Hyper-V and install an MSDN Windows Server 2012 instance for development purposes?
Answer:
We kindly inform it is not allowed to install and run software from MSDN subscriptions in production environments. Please do not install MSDN software on a desktop in a production environment:
"[.] The customer will need to run the developer software on a device/host that is dedicated to testing and development.
Explanation:
The Product Use Rights (PUR) say that the developer software is not licensed for use in a production environment. Even if the PUR does not have a separate definition of production environment, a production environment is a set of resources for network, physically
dedicated hardware and software to provide "live" service. If the intent was to say that the same physical server could be used for both development and production - it would say "not licensed for use in a production OSE," instead
it says environment.
See current PUR, page 51:
Developer Tools (User License)
You have the rights below for each license you acquire.
- You must assign each license to a single user.
- Each Licensed User may run an unlimited number of copies of the Developer Tools software and any prior version on any device.
- The Licensed User may use the software for evaluation and to design, develop, test, and demonstrate your programs. These rights include the use of the software to simulate an end user environment to diagnose issues related to your programs.
- The software is not licensed for use in a production environment.
- Additional rights provided in license terms included with the software are additive to these product use rights, provided that there is no conflict with these product use rights, except for superseding use terms outlined below.

Hi Mike,
It sucks that MSDN software can't be run in a production environment, that means you have to have two entirely separate hardware environments, which are costly, and it seems unnecessary.
That's essentially it. I'm not saying for one second that it should be used for production purposes, just that it's physical location shouldn't be relevant. Also, the word "environment" is a very bad choice in the documentation simply because it's very open
to interpretation.
A production environment is defined as an environment that is accessed by end users of an application (such as an Internet Web site) and that is used for more than
Acceptance Testing of that application
or Feedback. Some scenarios that constitute production
environments include:
Environments that connect to a production database.
Environments that support disaster-recovery or backup for a production environment.
Environments that are used for production at least some of the time, such a server that is rotated into production during peak periods of activity.
So I dont think (here's that inconclusive language) but am not sure that your desktop machines count as production environments, based on that, unless end users are connecting to them. (I dearly hope they are not!)
My reading is based on the "Other Guidance" section:
"If a physical machine running one or more virtual machines is used entirely for development and test, then the operating system used on the physical host system can be MSDN software. However, if the physical machine or any of the VMs
hosted on that physical system are used for other purposes, then both the operating system within the VM and the operating system for the physical host must be licensed separately."
<o:p>This is the crux of the matter and the interpretation of "licensed separately". A (to my mind) sensible reading of that would be "if you're running any production purpose VMs on a server then the physical host OS must be a full licence
[presuming it's Server 2012 and not, say, VMWare or Hyper-V 2012] as must all production purpose VMs on that server". This has been getting interpreted by others (I'm not the first) and backed up by MS as meaning that if you want to run any dev/test VMs on
a server that also runs production VMs then you can't use MSDN for those dev/test VMs.</o:p>
Also, there is a section
here, on the MSDN Licensing help page that says (with my added emphasis):
Many
MSDN subscribers use a computer for mixed use—both design, development, testing, and demonstration of your programs (the use allowed under the MSDN subscription license) and some other use. Using the software in any other way, such as for doing email,
playing games, or editing a document is another use and is not covered by the MSDN subscription license.
When this happens, the underlying operating system must also be licensed normally by purchasing a regular copy of Windows such as the one that came with a new OEM PC.
Now to me, it seems this might be saying that the underlying operating system on a work
machine cannot be licensed using MSDN if that work machine is going to be doing non-msdn things in addition to MSDN things. It doesn't say "This can't happen" it just says "When this happens, the underlying
OS must be licensed normally..."
So, based on what I'm reading it seems that this quote from you might not be true:
"We can't install a local MSDN instance of Server 2012 or 8.1 for dev and test under Hyper-V on desktops
because desktops used for email, writing documents, etc are production. "
I wouldn't have expected this to be true either, but this is the response I was given. It may well be
that my question was misunderstood. I hope this is the case otherwise one of the big reasons for turning on Hyper-V on expensive, powerful desktops enabling the running of personal test environments goes out the window!
Thanks for your time on this.

Struggling with oracle clustered production environment

Hi Experts,
I have a requirement to upgrade Oracle Data Integrator(ODI) from 10.1.3.5 to 11.1.1.6.3 We have a Clustered production environment where N1 will be up when N2 be down and viceversa.
Here N1 and N2 are the ODI servers as well as DB(11g Release 2) servers. They both access the SHARED CLUSTERED database.
From ODI we will generally point the Oracle clustered IP(Virtual IP) which will internally point either N1 or N2 whichever is active.
ODI application wise we are clear about the procedure.
Having some issues on DB related activities.
1. Should I break the cluster definitely? Cant I do the activity without breaking the cluster?
2. Do I need to point N1, N2, Clustered IP (Virtual IP) while doing the activities?
3. Since its a clustered database, do I need to db related activities once or twice? (Twice means, manually on both the servers)
4. As they are using same file structures (RAC), If the Virtual IP points N1 by default, assume that I create two new users and log in is success. Now i manually point N2 and what will happen if I try to log in to the two new users from N2. will it work?
5. If it will not work, then what would be the solution for this?
Please someone suggest about the clustered environment production problem.
As this is high priority, early response would be highly appreciated.
Need Expert's suggestions for the sequence of steps that to be carried out for a successful migration in clustered environment.
Many thanks in advance..!

This is an ODI question.
Please discontinue your question here and continue it in your duplicate thread here Struggling with oracle clustered production environment
Apart from that, ODI is a front-end tool. If you have any clue about RAC: for ODI it doesn't matter whether your database is RAC or not. the issue is you installed a front-end tool in a RAC environment.
Finally:
As this is high priority, early response would be highly appreciated.For high priority questions there is paid support. This is a forum of volunteers.
Asking for high priority is IMO insulting and rude.
Sybrand Bakker
Senior Oracle DBA

Is standalone OC4J scalable for production environment

Hi,
Can our team use an OC4J standalone (a cluster perhaps) to deliver Web applications on a production environment?
The Web app comprises JSPs and Servlets only at the moment, in future we are considering EJBs and other J2EE components.
Will standalone OC4J scale well and is performance good enough for real world applications?
Thank you
Llies Meridja

OC4J standalone is the exact same code that is packaged inside of Oracle9iAS so will have the same characteristics of Oracle9iAS. The benchmarks that Oracle markets with the ECPerf and SpecJAppServer is same thing ... so you should have no issues with performance etc.
You might have seen that we recently released a Java Edition of Oracle9iAS, which focuses primarily on the Java features rather than the extended features like Portal, Wireless etc. If you look at this URL, you will see the feature set comparison of this minimal Java version compared to the other versions:
http://www.oracle.com/ip/deploy/ias/index.html?pkgsum.html
Further, as I understand it, as long as you license one of the Java Edition, Standard Edition or Enterprise Edition, you will be supported running the corresponding instance stand-alone instance of OC4J if you want to work outside of the out-of-the-box framework which Oracle9iAS provides for OC4J instances to run in.
Mike.

Topology for production environment

this question is related to production environment deployment and clustering.
we have 1 physical server where we have webcenter domain. this domain has 1 admin server and 1 managed server (custom web center portal managed server).
our plan is to deploy our application on this managed server.
this is no clustered environment and not recommended (no failover, no load balancing).
what shall we do?
shall we create another managed server (custom web center portal managed server) and deploy our application on these 2 managed servers? there will be a load balancer in front to take care of transferring request to both the servers...
what's your suggestions?
please note that i know specific answer will depend on NFR (user load, performance etc) but i am looking for general guidance and reference around same like -
how many managed server recommended?
what kind of topology recommended for web center application?
etc
thanks.

jiri.machotka wrote:
I'd recommend you to start with this document: http://docs.oracle.com/cd/E11035_01/wls100/cluster/overview.html as a lot of this question is related to Weblogic clustering than WebCenter.
Note that whilst clustering has many pros, it has also few cons; among those:
- you might require additional hardware/software, or used components (like filesystem) will have to have features that you don't need in a non-clustered environment
- though small, scaling does consume some resources (I've heard that WebCenter Content takes 1-2% of Solaris, but up-to 25% on Windows Servers!)
- it's always at least a bit more expensive
Therefore, you should not go for a cluster, because it's cool, but because you have damn good reasons for that (that you didn't mention!), stemming from SLAs, or requirements for Disaster Recovery.
You write that you have just one physical server, so you already have one potential single point of failure.
I think I've heard that Weblogic Server itself is capable of self-recovery; this means, that a server may go down (you will lose all the current sessions), but is capable of auto-recovery. If your users survive that from time to time they will lose uncommitted work, then maybe a need for clustering (HA) won't be that strong (you might also go just for tuning your Weblogic Server).
On the other hand, where clustering might be necessary, if you need to scale, especially, if you need cloud-like increase/decrease of resources. I doubt it is possible to say up-front how many servers is recommended, but you may tune this later - whilst going from 1 node to 2 is relatively a big leap, going from 2 to 3, or 3 to 4 is then quite easy.
Either way, I'd recommend you to be in sync with your app server admin.thanks jiri. this was helpful.
while i agree that any architectural decision should be justifiable to requirements but as far as my experience goes clustering is basic thing in any production environment... in case 1 managed server goes down, other can still serve requests...
i'll read link provided by you and ask specific questions if any...
thanks.

How to move Oracle APPS 12 setup data to new APPS 12 production environment

Dear All
I need to know , how we can move Oracle APPS 12 setup data to new APPS 12 production environment?? What is the easy and secure way, we dont want to do the new installation of APPS 12 and re-setup the things....
cheers

Hi,
Can you please explain more regarding to move setup from DEV to PROD instance.The steps are documented in the iSetup manual, so have you gone through the documentation?
Also we would install PROD with multi tier nodes, and HA cluster. So in that case what would be the change to move the setups from DEV to PROD, as DEV is single node TEST APPS 12 installation.iSetup is helpful in moving the functional setup between the two instances. If you want to create an identical copy of your DEV instance to PROD instance (provided that both instances contain no data), then you can use "Rapid Clone" -- See (Note: 799735.1 - Rapid Clone Documentation Resources, Release 11i and 12).
Regards,
Hussein

Unable to import the Internet servivce into Production Environment

Hi,
We put the internet service and its HTML templates in a custom package for integrated ITSMobile. We got everything but the internet service ZXXX in the production. Has anyone come across this before? We were able to transport to the quality system. But it did not create the internet service in the production environment. Is it authorization issue? Your input will be highly appreciated.
Regards,
Subbu

Hi Subbu,
I suggest you to as your basis guys how are familiar with transports between systems. They will find out what the reason is. Without being able to look into transport protocols it is unlikeliy that someone here can advise you.
Regards,
Klaus

Issue in production environment

Hi,
We got an issue in the production environment on last Friday and yet not able to find out the root cause of the issue.
Environment
GUI : .Net
Server : Java
Application Server: Jboss
Database: Sqlserver
The Java application is deployed into Jboss application server which connects to SQL server as back end. GUI makes the web service call to server for data communication
Issue
Friday at 2 PM users reported the slowness of the application (Not getting the response from server in GUI), all the request from GUI were getting timed out.
Restart of Jboss didn't help
Restarted the Jboss and sqlserver for the second time and than the environment became stable
Analysis
1. From the thread dump of Jboss log we see that there are many threads waiting on a socket for database connection(according to database team all the connections are open and available at that particular time).
2. The size of transaction log is almost doubled during this period (when the issue was reported).
We couldn't find a reason why this issue is happened. Is it a database issue or something else? Please suggest...
Thanks,
Manoj

Hi Manoj,
According to your description, when running the web application, many threads are waiting for the socket from database which cause requests from front GUI time out. Right?
In this scenario, user can access the front GUI, which means the application server is working. Since the connections are all open at that particular time, it should not be issue on your JDBC. It seems to be a deadlock issue which cause other threads
waiting and hang. I recommend check and optimize your code. On database side, I suggest you open the SQL profiler, select DeadLock Graph, it will record when deadlock occurs. Please refer to links below:
Detecting and Ending Deadlocks
Analyze Deadlocks with SQL Server Profiler
If you have any question, please feel free to ask.
Simon Hou
TechNet Community Support

Development and product environment

Dear all,
I want to setup two databases in seperated server. One for development purpose and another for product environment.
Which good method/tool in oracle can transfer data (like pl/sql,tables,fields) from
development to product environment?
Thanks you

I would strongly suggest cloning the database from Development to Production, using RMAN. You get a proper copy that way - and it isn't hard to do. Howard Rogers has all the answers here: http://www.dizwell.com/prod/node/9

APO-DP - Change in Calculation Type of Key Figure in Production Environment

Hi Team,
I wish to know how to make a change in Calculation Type of Key Figure in Production Environment. I wish to change calculation type from "P" to "S" and also wish to enter Disaggregated Key Figure.
Regards,
Tarun Jha

HI,
The TR will not move the changes as you are expecting here.
Any changes to planning area, below 5.0 need to de-intialize the planning area.
just like
1. hold the latest back up
2. de-intialize planning area
3. make the required changes to planning area
4. initialize planning area and do conisistency checks
5. load back the data to planning area at all CVC level
all the steps need to perfom, in Dev and Quality and Production servers,
Plan the activity for a weekend so that users will not get disturb*
All the best
ARUN R Y

Security Issue httpOnlyCookies="true" is not working with production environment

Dear All,
I have tired to make set cookie value as httpOnly
The development environment is working fine and give me an output like that
Set-Cookie
WSS_KeepSessionAuthenticated={3644d93c-d1d3-46cd-845f-42c01640ab21};
path=/;
HttpOnly
But when implement the same changes in web.config production environment its not working
Set-Cookie NLSessionCdomainweb=a98q9jnsy0vXk5+RHeHnlKmM+HnneA9KmhIAR6g1bJiwTs8sD6d7dfV1gBffc8HiJXBowxdO8LhZAiIEKiFY6PzNWySyRs5rvgCfPu8XIFnqKcN4XQ4UL9PN3JI3f4E6;path=/;domain=.domain.com
I am using sharepoint 2010, under web.config i made the following changes
Add following tags under system.web
<httpRuntime maxRequestLength="2097151" enableVersionHeader="false"
/>
<httpCookies httpOnlyCookies="true"/>
Add following tags under <system.webServer>
  <httpProtocol>
      <customHeaders>
               <remove name="X-Powered-By" />
             <remove name="MicrosoftSharePointTeamServices" />
      </customHeaders>
    </httpProtocol>
Can any one tell me why it's happening i have checked all possible reasons from my side but no success
Regards
Rashid Bilgrami
RB

Hi,
From your description, I know you want to set cookie value as httpOnly but it is not work.
Please try to add codes below to your global.asax file:
<SCRIPT language="C#" runat="server">
protected void Session_Start(Object sender, EventArgs e)
try
if (Request.IsSecureConnection == true)
                Response.Cookies ["ASP.NET_SessionId"].Secure = true;
catch (Exception)
</SCRIPT>
Please refer to this article:
https://social.technet.microsoft.com/Forums/sharepoint/en-US/0fe55c13-3911-422e-af17-cb5c1ab2abd7/missing-secure-httponly-flags-on-sharepoint-2010-cookie?forum=sharepointadminprevious.
Best Regards
Vincent Han
TechNet Community Support

Problem on developing SSRS report in a production environment

Any problem could be caused if have to develop SSRS report in a production environment ?
If so, how it can be prevented?
Thanks!

Hi ,
As per my understanding, Normally our environment setup contains Development,testing,preproduction,production and its varies from company to company.
We develop SSRS report in development environment and after testing we deploy it to production report server.
Development environment use development database for creating report like your database IP and credential will use development environment
setup. and production environment use production database for creating report like your database IP and credential will use production environment setup.
Thanks
Neha Prajapati Please Mark This As Answer or vote for Helpful Post if this helps you to solve your question/problem.

How to port database changes from development to a production environment

How do I port database changes from the development to the production environment?
I am using v8 and have always had to redo everything using the schema manager all over in the production environment. Is there an easy way to generate a script, for example to dump the database changes on the development machine to be executed later on the production machine?

This should already be a clearly defined change control process. Once a procedure, function, package, trigger, or whatever completes the testing rounds, it should be promoted to production.
Forgive me if it seems I'm trivializing, but I don't see the problem, just copy the object(s) from your software library (or development) into production using whatever tool works best or has been chosen. If you are doing data copies then you have various options again including good old export/import.

Redeploy in cluster in production environment

Similar Messages

Maybe you are looking for