A small Xserve deployment

Hello everyone,
I am currently planning out a deployment for a medium sized business. We are a start-up, and therefore want to avoid having a full-time IT person, and our requirements are mostly along the lines of file-sharing between groups. The maximum number of people will most likely be 150, but currently is only 20.
Everyone will be using MacBooks as their workstations, so we are planning on doing the whole plant wirelessly.
Essentially the server needs to do the following: Wikis, SharePoints (group folders), and the occasional podcast producer item. Mail and Calendaring are 'in the cloud' via Google Apps, so we aren't worrying about this.
What my proposed set up is is the following (essentially copying the workgroup cluster model):
- 1 Xserve, 16 GB RAM, 3 x 1 TB drives in RAID 5, and the RAID controller.
- UPS and Server rack (a la workgroup cluster)
- A Drobo for backup - 4 x 2 TB WD drives.
From this I will use Open Directory to manage RADIUS authentication and share points and Wiki's.
Is this adequate? I've never set up anything like this, but I have done some reading on OS X Server.
Thanks in advance!

I have seen that there are issues in using VPN and Open Directory
I guess that depends on which VPN server you use. If you're planning on using the VPN server built-into Mac OS X then there isn't an issue - it uses the Open Directory system by default.
Other VPN servers may have issue plugging into Open Directory but without knowing which VPN you're planning to use it's impossible to say.
what can a user access if they're on the VPN? Will the share points and Wiki's be available?
Once on VPN the user can access any resources that are on the private network. The only issue is that the auto-discovery protocols won't reach out over the VPN (so you can't use the Finder to browse the VPN network, for example). For the most part this isn't an issue as long as you have a good DNS setup.

Similar Messages

Smaller TestStand Deployment?

Hi,
Can a smaller/faster deployment be built which contains only changes?
Currently I am generating a deployment every time I need to modify or add to my LabVIEW libraries, and then install it over the original deployment on my test PC. Instead of creating a new ~1GB complete deployment containing the modified code and unchanged elements can I instead create an incremental deployment build. The core TestStand and LabVIEW system remains unchanged along with many other libraries and dependencies so is it possible these can somehow be ignored in the deployment build - preferably with the tool sensing via comparison with a previous build no changes have occurred.
Maybe there is a document which provides guidance/best practices on optimal deployments?
Kind Regards,

Hi,
You can build two deployments, one for core and one for test sequences. Install the core once and then the test sequences many. You'll have to experiment with it yourself to get the right mix, but I'd put the test sequences and all the VIs in one, and the TestStand engine, process model, ini files in the core. Others use this approach. In general, you should always do a build and not replace individual VIs.
NOTE: I believe in TestStand 2010, if you use LabVIEW 2010 Packed Libraries, you can safely replace individual VIs on your runtime machine, in essence creating a "patch" for your installations. This is new and I haven't thoroughly checked it out.
cc

Small Business Deployment on End-User-Owned iPads/iPhones

Background:
6 employees (6 iPads, 5 iPhones) at an architectural firm
Need cloud-syncing of large PDF libraries of drawings and CAD files for use in the field
Office server is Windows Server 2003 (currently locked-down to local LAN & wi-fi only, no external access (boss' fear of security))
Office workstations are Windows-only (I've got Macs at home)
We have just been gifted iPads (personally-owned) to use for work/play and I received the tasks of setup & deployment. I would prefer to implement a controlled system to limit my headaches, this isn't entirely possible because these iPads were given as personal gifts and the employees would raise ****. After downloading/researching the Apple Configurator, it seems like it creates too much of a locked system - so end-users couldn't have their personal Apple IDs on their devices and buy/configure their personal stuff while being managed (work-purchased apps and PDF/CAD deployment) by me. Correct me if I'm wrong.
MDM implementations seem a little overboard. $$$
I have come across the iAnnotate app, which syncs with Dropbox (I can convince my boss to pay for Dropbox for Teams). While I would rather rely on an Apple-developed app (iBooks) to manage our library of PDFs, iAnnotate seems stable through user reviews. AutoCAD WS would be used for CAD files, with Autodesk's built-in syncing service.
1. My headaches aside, should I have each user buy/download/update apps with their personal AppleID and reimburse them (or gift them through iTunes from our corporate account)?
2. How could I control the wiping of a lost iPad/iPhone?
3. Any other recommendations for deployment in my situation?

boss would rather not have iTunes
The name is rather unfortunate in your situation. iTunes came out as a way buying music for the iPad. Apple had iTunes around, so they added features over time. It's now a device manager for all Apple i devices. It's more than a way to purchase music.
Do you take the iPad on customer visits. iTunes can load presentation on iPad.
If you sync the iPad on multiple computers, check this out.
http://support.apple.com/kb/HT1202
Robert
ps. Windows makes you paranoid.
If it is not broke, do not fix it.

Why font of menu is very small when deploy form to form server?

The font of menu is small When I used browser to run form in form server.
The size of font is correct when I browsed from windows 2000 But it is uncorrect when I browsed from windows 98 thai edition.
How can I adjust the size of font?

Try playing with the font settings in the /form60/java/oracle/forms/registry/Registry.dat
This will control what default fonts are used.

Is TMSPE a requirement for a Small/Regular Deployment?

Hi,
i am totally new to TMS and this is my first time installing. I have a small system with only 25 licenses.
Do i need to install TMSPE? and Does TMSPE need a license if I need to install it?
Thanks,
Surendra

Hello Surendra -
You only need to install TMS Provisioning Extension if you want to have any of the following features within TMS:
Smart Scheduler - Simple, Intuitive Scheduling of Single and Recurrent Meetings
Collaboration Meeting Rooms - Easily create and configure personal meeting spaces or conferences for users with there own dial-in address. Requires TelePresence Conductor.
Provision Endpoints to Users - Such as Jabber Video for TelePresence or EX Series endpoints. Requires TelePresence Video Communication Server.
It does not require any license to install. However, in order to provision endpoints to users (option 3 above), that does require provisioning licenses to be installed on your TMS.
More information can be found in the TME Provisioning Extension data sheet.

Deploying small JSF app to OC4J 10.13.10

Hi all,
I'm having a small problem deploying my JSF application on oas 10.13.10. Deploying actually works fine but when I try to surf to my app it gives me a HTTP 500 - Internal server error.
When I check the logs in the entreprise manager I see the following error:
07/10/25 08:46:32.203 10.1.3.1.0 Started
07/10/25 08:46:32.296 localvalidation: Error preloading servlet
javax.servlet.ServletException: Error instantiating servlet 'Faces Servlet'. Servlet class javax.faces.webapp.FacesServlet not found
in web-application localvalidation
at com.evermind.server.http.HttpApplication.servletClassNotFound(HttpApplication.java:4866)
at com.evermind.server.http.HttpApplication.findServlet(HttpApplication.java:4832)
at com.evermind.server.http.HttpApplication.findServlet(HttpApplication.java:4734)
at com.evermind.server.http.HttpApplication.initPreloadServlets(HttpApplication.java:4922)
at com.evermind.server.http.HttpApplication.initDynamic(HttpApplication.java:1134)
at com.evermind.server.http.HttpApplication.<init>(HttpApplication.java:738)
at com.evermind.server.ApplicationStateRunning.getHttpApplication(ApplicationStateRunning.java:414)
at com.evermind.server.Application.getHttpApplication(Application.java:545)
at com.evermind.server.http.HttpSite$HttpApplicationRunTimeReference.createHttpApplicationFromReference(HttpSite.java:1990)
at com.evermind.server.http.HttpSite$HttpApplicationRunTimeReference.<init>(HttpSite.java:1909)
at com.evermind.server.http.HttpSite.addHttpApplication(HttpSite.java:1606)
at oracle.oc4j.admin.internal.WebApplicationBinder.bindWebApp(WebApplicationBinder.java:238)
at oracle.oc4j.admin.internal.WebApplicationBinder.bindWebApp(WebApplicationBinder.java:99)
at oracle.oc4j.admin.internal.ApplicationDeployer.bindWebApp(ApplicationDeployer.java:547)
at oracle.oc4j.admin.internal.ApplicationDeployer.doDeploy(ApplicationDeployer.java:202)
at oracle.oc4j.admin.internal.DeployerBase.execute(DeployerBase.java:93)
at oracle.oc4j.admin.jmx.server.mbeans.deploy.OC4JDeployerRunnable.doRun(OC4JDeployerRunnable.java:52)
at oracle.oc4j.admin.jmx.server.mbeans.deploy.DeployerRunnable.run(DeployerRunnable.java:81)
at com.evermind.util.ReleasableResourcePooledExecutor$MyWorker.run(ReleasableResourcePooledExecutor.java:303)
at java.lang.Thread.run(Thread.java:595)
my web.xml looks like this:
<servlet>
          <servlet-name>Faces Servlet</servlet-name>
          <servlet-class>javax.faces.webapp.FacesServlet</servlet-class>
          <load-on-startup>1</load-on-startup>
     </servlet>
     <servlet-mapping>
          <servlet-name>Faces Servlet</servlet-name>
          <url-pattern>/faces/*</url-pattern>
     </servlet-mapping>
     <servlet-mapping>
          <servlet-name>Faces Servlet</servlet-name>
          <url-pattern>*.faces</url-pattern>
     </servlet-mapping>
     <session-config>
<session-timeout>35</session-timeout>
</session-config>
<mime-mapping>
<extension>html</extension>
<mime-type>text/html</mime-type>
</mime-mapping>
<mime-mapping>
<extension>txt</extension>
<mime-type>text/plain</mime-type>
</mime-mapping>
     <welcome-file-list>
          <welcome-file>Index.html</welcome-file>
     </welcome-file-list>
     <error-page>
          <error-code>500</error-code>
          <location>/ErrorPage.jsp</location>
     </error-page>
The Index.html, which is the welcome page, I can see fine. That page redirects to a page called Validation.jsp (but i redirect to Validation.faces ofcourse) and then the error comes in. Could this have anything to do with my shared libraries? I added the jsf 1.2 libraries to the shared libraries and then I got a different error saying I still miss something else, a javax.faces.FacesFactory (don't know the exact name of the missing class).
Could someone please help me on this? I've been trying several stuff now for 3 days and haven't found any solutions on the net so far.
Thank You!

Hi Frank,
I don't use /faces/ I use the .faces extension but I don't think that it matters wich servlet mapping you use, does it?
About the jsf-impl.jar, as soon as I put any libraries in my .war file (in the WEB-INF/lib directory) my application won't even deploy .. I even tried freshly installing a newer version (10.1.3.3.0) oc4j j2ee containers and tried to deploy both with libraries and without them but still got the same errors.
Isn't there some guide to follow for deploying jsf applications to oc4j j2ee containers? One which describes all the steps? Or maybe some people here that have already deployed such an application (even if it's as small as hello world) could describe how they got it working step by step?
Thanks for your help!
Tom.

New Xserve: Which Rack to Buy?

Hey guys,
We just bought a new Xserve Quad-Core w/ a Raid Card and 3 2TB HDs. I'm trying to figure out the best small rack to buy for it. The only thing that we'll be mounting in it for the time being is the Xserve.
First, I bought this rack: http://www.amazon.com/Solid-Versatile-Mounting-Servers-Network/dp/B000NMDIPE/ref =cmcmu_pg_header
The Xserve mount kit (square hole) fit on the inside flanges, but obviously didn't work because the Xserve could not slide in all the way. Then I ordered another rack:
http://www.tripplite.com/en/products/model.cfm?txtModelID=4025
Same situation. It's obviously because of the fact that the uprights have a "C" shape and not an "L" shape. So I need a small, Xserve compatible rack recommendation. Preferably open and not an enclosure, and under $500. Any suggestions?

I personally think that for a 1u xserve you should get an xrack pro. There a little more expensive but are amazing i think. They pretty much sound proof the xserve and keep it very cool. They also come with a glass door that locks.
Just my opinion

Choosing xServe model

Hello,
I am planning on making a order for a xServe during next week. My goal is to find the value for money balance and not buying something too overkill for my server-needs. I hope you can help!
I have a small company with 4 employees who are all connected to our Gigabit network / Aiport Extreme (n-band).
We will be using it for our primary mail accounts, daily backup services, shared drives for photoshop and webdevelopmentfiles, calender and mobile services in 10.6 Server. It will also run some VPN services for our network.
I have been looking at the BASE 2.26 Quad with 6 gb Ram, 3 x 1TB drives for my primary RAID (backup and shared drive data) and a SSD drive just for the system and applications following advice from another thread here!
Will the small xserve be able to handle this and also if we expand with 5-10 people more and some 'friends' get calendars and mail accounts. I know xserves are build for much higher user numbers, but due to the many purposes for this xserve I was not sure if it was better or necessary to buy a 8 core model.
Also I have seen some 1. gen XEON xServes being sold cheaply second hand. They look like they are in mint condition, but the prices would allow me to buy 2 or maybe three for the same price. Is used xserves a no-go?
Thank you!

I am always amazed at how low expectations are for Xserves. You spec is way more than adequate for 4 people. It can easily handle 30 to 50 as long as you don't intend to use it in a computation grid. In the role of traditional file services, your system spec is adequate. RAM is more important than CPU. However, I am still not comfortable with the SSD drive concept. Once again, there is no redundancy.
I personally prefer an Xserve with three 1 TB drives and the hardware raid card. Yes, you loose capacity but you gain redundancy. If you partition an 80 GB boot volume you have 1.7 TB of space for data and the knowledge that if a drive dies, you are still up and running.
Hope this helps

CME 8.6 Call Disconnect Issue

Hi All -
My company has a small CME deployment in Australia and yesterday the local network resource made some changes to the network related to other work but following their changes, a very weird problem is occuring on the system.
The site has 4 POTS lines coming in on a VIC-4FXO card and when calls come in, the IP phones ring but as soon as they try to answer the call disconnects. I was suspecting a codec negotiation issue but since the call is coming from the PSTN, I'm not real sure because it should be G.711 which the IP phones should handle fine.
my ccapi debugs show the call disconnecting with the typical 16 cause code (normal clearing). I can provide any part of the config or debug output required for some ideas.....
I appreciate the help.
Tom

Hey Paolo,
I am facing the same issue but only with a single user.
Using ISDN2e (bri). She answers the call and it disconnects, can you please shine some light please
Please rate the post
Shibly Ibrahim

SharePoint 2013 Environment Setup with minimum resource

Hi,
Could you please advise considering below scenario
- What would be the minimum resources in order to setup SP2013 Enterprise environment?
- If we want to buy a single server from DELL as a host for all virtual machines as below, what would be the configuration for that server?
- How much RAM do we need for the host?
- How much ram / processor core allocation do we need for the VMs.
Requirments:
USER: 500 Users
Features : All OOB features / Services will be used in SharePoint 2013 Farm
Web Application : 3 Web Application, 10 Site Collection
Total Content : 100 GB
Virtual Box for entire SP2013 Environment Setup:
- SQL Server 2012 Sp1 Enterprise
- SQL Clustering / Mirroring
- SP 2013 Application Server1 (Enterprise)
- SP Application Server2
- SP Application Server3
- SP Application Server4
- SP WFE 1
- SP WFE 2
- BI Server 1
- BI Server 2
- OWA Server 1
- OWA Server 2
- Workflow Manager 1
- Workflow Manager 2
- TMG Firewall 2010
- Exchange Server 2010
- Microsoft Lync Server 2010
Thanks
srabon

It's all here:
http://technet.microsoft.com/en-us/library/cc262485%28v=office.15%29.aspx
Putting that many servers together for a 500 seat company is madness. Go for the text book small farm deployment at most, in reality you'd probably be better off just having three servers total (1 WFE, 1 App, 1 SQL).
I can't remember how much you should reserve for the hypervisor but if you allocate 1GB for that i think you're ok.
For a three server deployment you'd probably be ok with 32GB, 8GB for the WFE and App server, 7GB for SQL and one for the hypervisor. 64GB would give you a lot more comfort.
If you want more options like the enhanced workflows and OWA you'd need servers for each of those as well.
You can always overprovision CPUs on the hypervisor but more than 12 Cores would be wasted on a 3 server deployment for 500 users.

Can you use a self signed certificate on an external Edge Server interface?

Hi,
I have a small lab deployment for evaluation purposes. The Lync FE server works great for internal users. I have now added an Edge server. For the internal interface, I have a self signed certificate from our internal CA. (no problem there) For the external
interface, I have a self signed certificate from our own external CA. I have installed the cert on the client machine of the external user and installed it for trusted operation. I have used the RUCT and digicert tools to prove that the external self signed
cert is valid (root and intermediate have been checked for validity).
At first, when logging in from the Lync 2013 client on the external users machine, I would get an error from Lync about the cert being untrusted. I have now fixed that error by adding it as trusted. At this point, there are no errors or warnings in the Event
Viewer (in the application or system logs) However, I receive the following error from the Lync client, "Were having trouble connecting to the server... blah, blah".
Here is my question. Does the Microsoft Lync 2013 client and/or the "testconnectivity.microsoft.com" tool specifically prevent or forbid the use of self signed certificates on the external interface of an Edge server? They seem too.
I can tell if the certificate is my problem or something else. Any ideas on how to trouble shoot this?
Thx

Drago,
Thanks for all your help. I got it working.
My problem with the Lync client error, "Were having trouble connecting to the server... blah, blah", was NOT a certificate error. It was a problem with my Lync Server Topology. (My sip default domain needed to match my user login domain.)
Let me update everyone about self-signed certificates:
YES, you can self-sign a certificate on your external edge server. It is a pain, but possible.
I have a self signed certificate from our own external CA. I have installed the cert on the client machine of the external user for trusted operation. I have used the RUCT and digicert tools to prove that the external self signed cert is valid (root and
intermediate have been checked for validity).
Here are my notes:
Create/enable your own external Certificate Authority (CA) running on a server with internet access.
On the Lync Edge Server, run the "Lync Server 2013 - Development Wizard".
Click "Install or Update Lync Server System". (Lync will automatically determine its deployment state)
You should have already completed: Step1 and Step 2.
Run or Run Again "Step 3: Request, Install or Assign Certificates".
Install the "Edge internal" certificate.
Click "Request" button to run the "Certificate Request" wizard.
You use can "Send the request immediately to an online certificate authority" option to connect to your internal CA, and create the certificate.
Once the certificate has been created, use "Import Certificate" to import it.
Once imported, on the Edge Server, go to: (Control Panel -> Administrative Tools -> Internet Information Services (ISS) Manager -> Server Certificates -> Complete Certificate Request...
In the Lync deployment wizard - Certificate Wizard, "Assign the newly imported "edge internal" certificate.
Install the "Edge External" certificate (public Internet).
Click the "Request" button to run the "Certificate Request" wizard.
Press "next"
Select "Prepare the request now, but send it later (offline certificate request).
Supply the "Certificate Request File" name and location. (You will need the file later. It should have the file extension ".req").
Click next on the "Specify Alternate Certificate Template". (which means you are using the default options)
Give it a Friendly Name. Bit Length = 2048. I selected "Mark the certificate's private key as exportable" option.
Fill in the organization info.
Fill in the Geographical Information.
The wizard should automatically fill-in the "Subject name:" and "subject alternative name:' fields.
Select your "Configured SIP domains"
"Configure Additional Subject Alternative Names" if you want. Otherwise, next.
Verify the "certificate Request Summary". Click next.
Run the wizard script to "Complete". The wizard will create a file containing the certificate request with the file extension ".req". (Let's assume the file name is "myCert.req")
Move your myCert.req file to your external CA. Have your CA issue the cert (based on myCert.req) and export the new cert to a file. I save it as a P7B certificate. (Let's call it "ExternalCert.p7b")
In the Lync Deployment wizard - Certificate Wizard, click on "Import Certificate" for ExternalCert.p7b.
Once imported, on the Edge Server, go to: (Control Panel -> Administrative Tools -> Internet Information Services (ISS) Manager -> Server Certificates -> Complete Certificate Request... (assign it a friendly name. Let's say "EXTERNAL-EDGE")
For the "External Edge certificate (public Internet), click "Assign".
The "Certificate Assignment" wizard will run.
Click next.
From the list, select your cert "EXTERNAL-EDGE".
Finish the wizard to "complete".
You are finished on the server.
Move the "ExternalCert.p7b" file to the machine running the lync client. Install the cert via the "Certificate Import Wizard".
When installing it to a particular Certificate Store, select the "Place all certificates in the following store" option.
Browse
Select "Trusted Root Certification Authorities"
Finish the wizard.

To Load Balance or Not to Load Balance? ISE and F5 Big IP

Currently my team is debating whether to put our two ISE appliances (PSN nodes) behind our F5 load balancing deployment.
Our network is relatively small in size (5K users) with a small wireless deployment (4 Cisco controllers with 300 Access points). Network growth should remain relatively minimal over the coming years.
We will be rolling out wired Dot1X, followed by posture assessment and remediation. (BYOD is not an option).
On one hand, the Big IP features could make it easier for us to perform load balancing, maintenance and troubleshooting.
On the other hand, the Big IP adds another element of complexity into an already complex deployment. We already have the capability to load balance from the switches themselves. Load balancing for wireless should not be an issue as our deployment is very small and I expect it to remain so. Given the size of my environment, there seems to be relatively little to gain for the additional effort and potential pitfalls.
Would anyone care to share their honest opinion on this issue?
Thanks,
Phill

Load balancers are elegant and do their job nicely when it comes to distributing the load between servers. You already have one so I would suggest using it if you have the technical expertise to configure it.
With that being said, if your team is not 100% comfortable with F5 then you should definitely skip it. Instead, you can configure your WLCs to use Node #1 as primary and Node #2 as secondary Radius server and then your Switches to use Node #2 as primary and Node#1 as secondary.
I hope this helps!
Thank you for rating helpful posts!

Question on Contract Manager Adminstrator roles

I have a question on IT roles around contract manager. What do people typically see as far as contract manager IT administrator roles ?
We are deploying contract manager to a user base of about 500 users and perhaps 100 concurrent users (eventually). Right now we have a small production deployment for 3-4 construction projects.
We are trying to figure out how to structure the contract manager admin role. One idea we have is to establish a 1st level of support, provisioning users and possibly setting up new projects. I could be involved in a 2nd level of support, dealing with more infrastructure problems, such as defining new security templates, configuring document attachment storage, troubleshooting problems with the server or with letters. We’d structure it this way so that the day-to-day work of dealing with users would fall to the 1st level of support and the more backend problems would fall to me. I have some other projects I’m involved in so the day-to-day dealing with CM users would not be as much of a burden on me.
My boss is concerned about clear cut roles. If his group is responsible for an enterprise system (contract manager) then only members of his team (me) have the power to do things like modify the configuration of CM, start/stop the server, reboot the machine. Other IT team members, not a member of his team (1st level support person) could have admin privileges to deal with user/project provisioning only. 1st level support person wouldn’t have full administrator privileges. Is this technically possible? Does the security model of CM support such a division of labor?
What else have people seen that works as far as roles? Is having just one person take on both user/project provisioning and the rest of the admin tasks doable? If a single person is managing contract manager, is this a part-time load or a full-time load? Recall that I’m slated to be involved in other projects such as UCM in general, I/PM (Oracle Imaging and Process Management), possibly some integration projects too.
Are there any guidelines for best practices around admin roles?
If anyone has any experience/opinions on Primavera P6 roles I’d like to hear about them also.

Yes, the arrangement you describe with you being the 2nd level support (server level functions) and another person being 1st level (user admin, etc) is certainly technically possible. You would have full admin rights to the server on which CM resides, and the 1st level person would only require access to the CM Admin web-based interface through an ID and password. From what you describe, that person would not need physical access to the server at all in order to handle the user account management.
Now, based on the number of users which you descibe, it really sounds like a full time job for at least two people. These numbers will vary based on how much the CM app will be customized and integrated with other systems. Beyond the IT support you describe, there will be a need to have "super-user" level users that can handle the application level requests such as how do I add a new contact, what value do I put in that field, etc.
Although it's becoming more common, 500 users is still a large implementation in the PCM world. I've seen organizations with up to 10 IT type people involved with implementations of that scale, but maybe not all full time.
Rick

Air for Android performance

Hi everyone,
I'm currently working on a game for iOS devices and I've decided to test it on android as well to see the performance difference.
My game works at 50-60 fps on iphone 3GS, almost constant 60 fps on iphone 4 and constant 60 fps on iPad2.
I own a Nexus 1 and from reading in different forums it seems like it should perform almost the same as iphone 3GS.
When I deploy my game to my Nexus using the same code and optimization techniques I used for the iphone (and of course using GPU mode) I get 17 fps at the most. Now I've done some testing and it seems that the game runs on 60 fps when nothing is moving around but when I start to blit one bitmap object it drops to 30fps right away.
Is 30fps the max you can get on devices like the Nexus 1 using Air or is it Air's limit for all android devices?
Thanks,
Oz

Ok, here is the code of a small test I did:
var bitmap:Bitmap = new Bitmap(new Square());
addChild(bitmap);
addEventListener(Event.ENTER_FRAME, function() { bitmap.y += 0.2; });
addChild(new FPSTimer()); // a class that shows the frame rate, I tried removing it to see if the speed of the square movement changes but it doesn't make a difference.
That's all the code I have on the first frame of the project. The Square is a 10x10 pixels bitmap with a blue solid color and no transparency at allt that I have in my library.
When I only add the bitmap to the stage without running the update event I get 60fps, once I start to run the update to move the Square it drops immediately to 30fps, for moving only this small bitmap.
deploying the same file using cpu I get arounf 55fps.
edit: I also tried changing the color depth back to 16-bit like it was in older versions of Air. It's now working on 19fps instead of 17 :/
edit 2: I decided to try and replace my sdk back to 2.7. doing that I also managed to increase the fps to 19-20 but not above that. I also tried to run the test I did and now it runs with 40fps on gpu and 57 fps on cpu. but it shouldn't even change the fps because I'm moving on small bitmap, I really don't get it.

CA Server and GET VPN Key Server

Hi,
Can I have an IOS CA Server and a GET VPN Key Server working in the same ISR G2?
Thanks
Emanuel

Emanuel,
No I would not necessarily call this a small scale deployment, although we do scale above 4000 GMs.
Please note that, at least as far as I am aware, there is no strict definition that a setup like this would not be supported for larger scale deployment. You may want to shoot your SE an email so they can discuss with business unit it they limit supportability of such setup somewhere.
Technically speaking, what you need to take into consideration:
- CPU utilization during registration (can be offloaded by using external CDP URL).
- Type of rekey.
- Amount of GM re-registrations. (i.e. stability of environment).
- KS COOP or not.
- KS platform of choice.
What you want to make sure is that PKI functions will not affect KS functions. (For example during multi spokes registering and performing CRL checks).
And make sure that KS is not a single point of failure for entire domain - that mean storing PKI data of the router.
M.

A small Xserve deployment

Similar Messages

Maybe you are looking for