Multi node two web servers?

Is it possible to have TWO web servers, on a multi node install. I am not really concerned about load balancing. I am mostly concerned with installing a second web server in a DMZ for external users to access.

There is a document in Metalink for this. We have implemented DMZ setup in multinode. Note:287176.1
-Thiru Pandian

Similar Messages

Two web servers cant use same port forwarding rule???

I have two web servers, each configured to respond to http reqests on port 80. I use NoIP service to map a domain name to my router. I create a port forwarding assignment for Server A by selecting the WebServer rule in the port forwarding rule table. All is fine. For Server B, I select the same Webserver rule and the router says there is a conflict.
I don't understand why. I think the router has enough information to route a http request for Server B by knowing its MAC address, which is different than Server A MAC address.
What am I missing here???
I was able to do this on my previous router with DD-WRT.....
So it would appear the way to resolve this is to assign a different port address to Server B???
grrrrrrrrrr I hate this router

You shouldn't have been able to do that on any router. You have to change the listening port of server 2, to 8080 or something like that, or make one a secure server and run it over 443
That is a known issue with port forwarding
Problem #8:
Same Port in Multiple Rules
There are some routers such as Linksys, D-Link and many others that do not do any checks if a port is already in another port forwarding rule. A port can only be forwarded to one Computer/IP at a time. So when there are multiples of the same port number the port forwarding rule will not work.
Here is an example.
As you can see port 2350 is in 2 rules. The 1st one points to a different IP than that of the 2nd rule. So the router will honor the 1st rule and the 2nd port forwarding rule to port 2350 fails.
By removing the 1st rule the 2nd one will now work.

Two Web Servers

Hey everyone. I'm stumped!
Here's my set up.
I have two 10.6 Servers. One's the OD Master, running DNS, DHCP, SMB and a few other services. The other's a replica running client fun time services. It's also hosing all the users "Public" info via AFP and SMB. Most importantly it's running the school's web server. The router forwards all requests on port 80 to that server.
I've set up setting up a 10.7 Server that's going to host all of the school's iPad configurations.
It's set up, users can authenticate and download profiles etc INSIDE the school. But not outside.
I've set up the name server info already so outside the school, 10.7server.server.com points at my IP address, but when you do that, it just dials up the 10.6 web server.
My question is this. How do I set up the router so that requests go to 10.7server.server.com when that's the address entered by users?
OR, is there a way make the DNS server or the Web server recognize when someone types 10.7server.server.com from outside the school?
-Graham

IP routers know from and deal with IP addresses, and not with DNS names.
If you want to implement a scheme where you have two external references into your network, then you would want and use two public IP addresses and the associated forwarding, or a much smart(er) router (and those aren't cheap), or you could configure your web server as a reverse proxy (which is more complex).
With a gateway-firewall-router with decent server-grade capabilities, I'd likely move the public-facing web server stuff to a box in a DMZ (as that "bastion" server is the most likely to be breached, particularly if it's running a web content management system or other services) and configure the gateway-firewall-router as a VPN server end-point. I'd expect the "bastion host" box would get breached, but the DMZ would keep it contained. The VPN is the path into your internal network, and avoids exposing more protocols and services than necessary; more paths for attacks.
If you want to configure DNS within your network to reference a different IP address than your public DNS, here's some reading. (I don't know how much you know about this networking area, so some of this may well be little more than a review for you.) With internal DNS from your server, you can serve the internal IP address to your users and to users on the VPN, while your external DNS can translate to a different address. That article links to another on configuring external services, as well as to a write-up on IP networks, firewalls and related details. (It's also possible to have a mid-grade router correctly process the external IP address and "reflect" it back into your network.)
And if you want to obfuscate your domain, use "example.com", "example.org" or "example.net". These domains are reserved for this use. The domain name you've specified is a real and registered domain.

Best way to transfer file between two web apps

I'm looking for some architectural suggestions. Seems like a simple problem, but I'm really struggling.
-- Problem:
We have two web servers, both running apps in JBoss. I need to pass a file of roughly 20MB from one server to the other, and get the other server to process it as soon as it receives it.
-- Details:
Seems simple enough... My plan was to write a web service to pass it. Our company is all onboard the SOA train. I wrote a service that encoded the file as a byte[] and passed it. Works amazing for small files, but throws an Axis OutOfMemory error when trying to base64 encode the file. I then looked at using attachments for web services, but dismissed that. It creates malformed XML since it treats the service as a multipart MIME message wraps the webservice inside. It quickly became messy and I couldn't get it working. Seemed to defeat the purpose of an XML service too.
I've considered just doing a HTTP POST of the file to the other server, but that's poorly documented and a bit of a hack way to do it. I also considered just dropping the file in a web accessible folder, then using a web service to pass a URL and get the other server to retrieve it via HTTP. Again, not a very nice solution.
Anyone have any suggestions? Thanks.

Kimos2 wrote:
I don't have access to run/configure services on these machines, so FTP is out. It has to be JVM to JVM. Even if I did, I'd have to worry about triggering JBoss to process the file right away after it had been dropped there. I know i could poll the directory or send a message to trigger the process, there'd be synchronization issues etc to work through and seems to introduce unneeded complication.Yes, I agree with that. Having been through the polling directories and grabbing partially uploaded files scenario ad nauseam. It's much worse with large files too.
Barring any suggestions I haven't thought of, I'll probably end up settling on the HTTP POST solution. My problem is that it's providing a service without a specification and is not self-describing. No WSDL file to pass to clients that plan on using it. I would do HTTP POST as well. I didn't understand the part about "without a specification". You do have a specification, don't you? Wasn't this it: "I need to pass a file of roughly 20MB from one server to the other, and get the other server to process it as soon as it receives it." Informal, yes, but I don't have a problem with that. Post it to a URL on the receiving server and set up the receiving server to do whatever it's supposed to do with the file.

Azure VM Web Servers, Cloud Service and SSL?

I am deploying two web servers in a cloud service for load balancing. I have tested and I have that working with no problems. My question is how do I setup for SSL? Do I setup a certificate on the cloud service? Do I setup on each server? Any help would
be greatly appreciated!!

Hi,
You could use the following link to configure SSL for your Cloud Service:
http://azure.microsoft.com/en-us/documentation/articles/cloud-services-configure-ssl-certificate/
Regards,
Malar.

Financial Analyzer Web access after multi-node setup

hey guys
any idea on the follwoing
Financial Analyzer Web access after multi-node setup
any hints will be helpfull
fadi

Thank you Fadi for this quick idea.
I will try to apply it in our environment.
If we couldn't do it, I think that we have other two options; first options is to reinstall OFA in the Web Server Node where there is a Metalink note states how to move OFA from server to server. The second option is changing the configuration files of OFA to reference the Web Server in the new node.
I will update you with the results.
Regards,
M.Muhtadi

CF Licensing on two front end web servers

I'm currently running CF 9 Standard on a since server that runs as my web and db server. I use Rackspace Cloud btw.
I've been thinking about changing my infrustructure around so it would look like this:
- 1 Load Balancer
- 2 Web Servers
- 1 Database Server
My question is about licensing on the 2 web servers.
- Is it possible to use a single CF Standard license to cover both web servers (since they are just syncing each other, they are basically the same server but split into two separate virtual machines), or would I need to purchase 2 individual licenses to make this work?
Any help or advice is appreciated. Thanks!

Thanks vishu,
I can read the license agreement, but it's the interpretation of legal wording that I'm having troubles with. Basically, I'm looking for a real world YES or NO to my question. I'm sure others have had this question, that's why I came to this forum.
2.1 General Use. You may install and use one copy of the Software on up to the Permitted Number of your compatible Computers as long as, when required by the Software, you present a valid serial number for each copy; and
2.2 Distribution from Server. You may copy an image of the Software onto file server(s) within your Internal Network for the purpose of downloading and installing the Software onto Computers within the same Internal Network for use as permitted by Section 2.1; and
2.3 Server Use. You may install the Software on Computer file server(s) within your Internal Network only for use of the Software initiated by an individual from a Computer within the same Internal Network as permitted by Section 2.1. The total number of users (not the concurrent number of users) able to use the Software on such Computer file server(s) may not exceed the Permitted Numbe

What is the recommended ratio of publish nodes to dispatcher web servers for CQ5.5?

What is the recommended ratio of publish nodes to dispatcher web servers for CQ5.5?
We are getting conflicting opinions on this form the field. I realize this is a loaded question. I'd like to see how Adobe presents this opinion. In the past scaling at the webserver level was only loosely tied to the publishing layer. Now it seems as dynamic functionality increases the ratio is getting closer to 1:1.
Also,I'd like to know in what use cases this load on publisher increases dramatically.

Hello, this is an evergreen. Just call setPickOnBounds(false) on the CustomHexagon.
An issue tracking this problem is open here: https://javafx-jira.kenai.com/browse/RT-17024

NLB nodes remain in a converging state indefinitely? Server 2008 Standard R2 x64 used as web servers, IIS 7.5.

I'm a newbie to NLB. We have just two identical servers in the NLB, a pair of nics, and unicast mode, being used as web servers. Let me know if you need additional info and what it may be.
What I'm doing is adding ip addresses to the cluster using the NLB Manager, one ip add for each site. It doesn't happen every time, but many times when I add an ip add one of the servers just spins converging.
I have KB 978943 which says there's a hotfix available for the problem. What I don't have is the Event ID 119 in the Event logs mentioned in the KB. I have nlb.sys v6.1.7600.16385 dated 13July09.
How can I fix this?

I had similar issue with 2 virtual web servers configured in NLB Windows 2008 R2.
The setup was working fine until recently and in bizarre way one of the machines went into converging state and never stopped doing that.
I tried:
- reinstalling NLB
- rebuilding the machine which kept on converging
- changed the IP addresses on both the machines; but with no luck. Even with new machine it kept on converging when I introduced it to the NLB
Resolution I found:
1. The issue was not with the machine which kept on converging; but the machine which remained converged all the time
2. I removed the machines from the domain. Changed their DNS names. Reintroduced them to the domain and BINGO; they bothe converged successfully in NLB after doing this action. In some awkward way the domain schema had a old foot print of the this NLB configuration
which got resolved after changing the DNS names of both the web servers.
Let me know if it works.
note: I have not used KB 978943

2 Logical Host, 2 Web Servers, Big Problem?

I am setting up a sun HA cluster using 2 E4500 servers. I have created 2 logical hosts, each one needs to host a Netscape iPlanet 4.0 web server sitting at port 80. Each logical host is serving up web applications for different clients.
If I need to fail over one of the logical hosts so that they are both running on the same system, the newly imported instance of the web server will fail because port 80 is already in use by the logical host that is on that physical host.
At first this seemed totally wrong. Each logical host should be able to run applications on any port they need to. Then someone who has a lot more time on Solaris told me that this was not the case, and each logical host had to steer clear of using the same ports as other logical hosts in the same cluster.
Can someone clue me into what is reality?
Any good documentation that tells how to set this stuff up?
Thanks!
Bruce

Hi,
The best way to resolve this would be to try implementing the same.
Instead of going through an entire cluster install/configuration
process, you may want to try setting up two different web servers
on a single node. You may want to set up a virtual interface (like
le0:1, hme0:1 etc) for this. You coud then try connecting to
individual web servers on port 80. If this works, then the two
webserver/two node cluster implementation should also work.
Hope this helps.
Thanks,
Gopinath.

Has anyone installed XMLP in a multi-node appsTiers environment?

We are experiencing a problem with installing XMLP 4.5.
Our EBS 11i(11.5.9) configuration: dbTier is on HP-UX, admin/CM of appsTier is on HP-UX and web/forms of appsTier is on Windows, i.e two nodes appsTier.
Applying the major patch 3554613 (XMLP 4.5) created $APPL_TOP/xdo/11.5.0 on both UNIX and Windows nodes. But the $JAVA_TOP/oracle/apps/xdo is only created on UNIX node.
When we reported the problem of XMLP in relation to OA Framework (see my other posting on this forum) to Oracle, the support engineer said the patch should create %JAVA_TOP%\oracle\apps\xdo on Windows too. But he is not sure as he only has a single node environment.
We tried a couple of times of reapplying the patch without any luck.
We start wondering if the XMLP 4.5 can be successfully installed in a multi-node appsTiers environment.
So my question is:
Has anyone successfully installed XMLP in a multi-node appsTiers environment?
Thanks.

Hi
XMLP is installed on the apps middle tier so it should be no problem. The only thing, once you have it installed is if you want to use our cfg file you'll need to have the cfg file installed on all nodes.
Whats the TAR number ?
Tim

How to setup multi node installation

Hi all,
I have successfully installed Oracle E-Business Suite 11.5.10 in a single node.
But, I would like to get some suggestions on How to perform the multinode installation in which all the application tier servers in independent nodes?
If the nodes are placed in different places geographically, how can I move all these nodes in to a single network? Is this possible?
Please guide me.
Thanking in advance
Rameshkumar

1. Is it right if I assumed that we must check that
box if we want to install apps tier in more than one
server so that if we only want to install apps tier
in only one server then we don't need to check that
option? Or we better check it even no matter how many
server we installed apps tier into? You only need to check that box if you intend to used a shared application tier filesystem for a multi-node installation. If you are only going to have one app tier server, you don't need to check the box. If you plan on multiple apps tiers, each with its own copy of the application tier file system, you also don't need to check that box.
2. If I still uncheck that option will there be any
problem such as in backup & restore, clone, testing
activity, etc?No, should not be any problem at all.
3. When we created user for each db tier and apps
tier, what will be better, create the same user for
both tiers or differentiate user for each db tier and
apps tier? I'm thinking of using same user for both
tiers to avoid privilege problem later on.
I'd recommend the same user for application tier files on all nodes in the system, and a different user for database-tier files. For example, in our system the database software (RDBMS ORACLE_HOME) and database datafiles are owned by 'oracle', and the EBS application files (APPL_TOP, forms and web server, etc) are owned by a user called 'applprod' (though of course you should choose a user name that makes sense in your environment.
4. Can we start clone process in db tier and apps tier >paralel in multinode environment or still start clone >process in db tier first then apps tier?Still need to do the db tier first, since 'adcfgclone.pl appsTier' will want to talk to a database. :-)
Regards,
John P.
Message was edited by:
jpiwowar (forgot to answer the 4th question the first time, whoops)

Two node - Two site Active/Active DAG Configuration

Information
Both sites exist in a single AD domain
Site A
1 Exchange 2013 multi-role server (CAS + MBX)
Server will contain mailboxes for individuals who work in Site A
Server will contain database copies of databases in Site B
Site B
1 Exchange 2013 multi-role server (CAS + MBX)
Server will contain mailboxes for individuals who work in Site B
Server will contain database copies of databases in Site A
Goal
To create a configuration where users who work in Site A have their mailboxes active in Site A and users who work in Site B have their mailboxes active in Site B.
Questions
Since this DAG has an even number of members, where do I place the witness server? Do I need two witness servers, one in each site?
It seems to me as long as both sites can still communicate, I only need one witness server to maintain quorum.
To my current understanding, if Site A has a witness server, and Site A looses its WAN connection, Site B's databases are going to go offline because it has lost quorum.
I'm assuming if I placed a witness server in each site, and one of the sites lost its WAN connection, then I'd be placed in a split-brain scenario as both sites would still think it had quorum. Is this accurate?
The company has given me a limited budget hence the reason for only having two exchange servers in this setup. Is my only recourse to have four nodes (two nodes per DAG) in this active\active scenario? If the company says they absolutely cannot spend money
on two more servers, could I use the current setup in an active\passive configuration?
Thanks!

You would need to place the File Share Witness in a 3rd datacenter for automatic failover, otherwise depending which DC the FSW is ( and you can only have one FSW per DAG active a time), then in case of a WAN or DC failure, the databases will be active in
the DC that has quorum.
If the FSW is in Site A and Site A goes down, then quorum is lost and Site B will need to be manually swithed over using this procedure:
http://technet.microsoft.com/en-us/library/dd351049(v=exchg.150).aspx
To be truly active/active in both DCs, then you would need to 2 DAGs with each respective FSW in each DC that has the preferred user base.
Twitter!: Please Note: My Posts are provided “AS IS” without warranty of any kind, either expressed or implied.

Wlw-runtime-config.xml & two managed servers

I have a problem with WSDL generation and WebLogic workshop in a
cluster.
I'm using a wlw-runtime-config.xml file to ensure that my web service
generates the correct url for it's endpoint.
The problem I have is that I have two managed servers on one physical
box, one listening on port 7010, the other on port 7030.
The wlw-runtime-config.xml file is placed in the domain root, and I
only seem to be able to specify a single port/dns name in the config
file.
My web service request will come in through the IIS plugin, hit one of
the servers, generate the WSDL containing the port information, which
may or may not be correct for the WLS instance that it has actually
been proxied to.
If that is correct, how do I handle the fact that the web service is
deployed across all cluster members, two of which are hosted on one
single machine.
Am I missing something?
Pete

Hi Pete,
I'm assuming you have seen the "how do i" on this topic [3], plus the
information on how to setup a cluster [1].
Your missing link may be the way your DNS entry is configured for the
cluster. See the docs [2] on this topic, but you will need to specify
the multiple nodes something like this:
DNSName1:port1,DNSName1:port2,DNSName1:port3
Note the folks in our super support group will be best suited to assist
your efforts.
Regards,
Bruce
[1]
http://e-docs.bea.com/wls/docs81/cluster/setup.html#684345
[2]
http://e-docs.bea.com/wls/docs81/cluster/setup.html#714310
[3]
http://edocs.bea.com/workshop/docs81/doc/en/workshop/guide/deployment/howClusterDeployment.html
Pete wrote:
>
I have a problem with WSDL generation and WebLogic workshop in a
cluster.
I'm using a wlw-runtime-config.xml file to ensure that my web service
generates the correct url for it's endpoint.
The problem I have is that I have two managed servers on one physical
box, one listening on port 7010, the other on port 7030.
The wlw-runtime-config.xml file is placed in the domain root, and I
only seem to be able to specify a single port/dns name in the config
file.
My web service request will come in through the IIS plugin, hit one of
the servers, generate the WSDL containing the port information, which
may or may not be correct for the WLS instance that it has actually
been proxied to.
If that is correct, how do I handle the fact that the web service is
deployed across all cluster members, two of which are hosted on one
single machine.
Am I missing something?
Pete

11.5.10 multi-node instalation

Hi,
I'm trying to install 11i on Oracle Linux in multi-node mode. I have two servers (fin-01 and fin-02) and decided to use fin-02 for DB and CM, fin-01 for the rest. Installation starts on fin-01. But System Check Status showed in 'Database Install Check', that "RW-40000: Warring: Unable to establish a Database connection. Please make sure the database is up before continuing with this install. Connection string used - jdbc: oracle:thin: @fin-02.domain:1521:VIS"
Is that means, that I need to install DB before?
regards,
Krystian

Hi;
You need to start installation from db tier.Please see:
multinode installation steps
multinode patch
Regard
Helios

Multi node two web servers?

Similar Messages

Maybe you are looking for