Mavericks Caching Server and PAT

Our network has a PAT configuration, with multiple public IPs. How effective will the Mavericks Caching Server (CS) be in our environment?
A PAT configuration would seem to be more likely in a medium to large sized network, which is where a CS would be most effective. Documents we've found so far indicate that the Apple update servers track the public IP of the CS and redirect clients using the same IP to the local CS. If that is correct, we'll probably see minimal benefit: with 32 public IPs in our PAT range, there is only about a 3% chance a given client has the same IP as the CS.
Although a firewall rule to route all Apple update-related traffic out via a single IP might work, that would require knowledge of all the addresses in use. A better solution might be to provide a way in the CS configuration to specify the pubic PAT range shared by the local private networks, with some mechanism for verifying that the CS can utilize all of those public IPs.

Apple have the entire 17.0.0.0 block of addresses so any Apple traffic will be in that block.
I also suggest you go to the following webpage on both your Caching Server and your clients several times over a period of time to see if they do all show the same single public IP address in which case you will not have a problem.
See http://www.whatsmyip.org

Similar Messages

Mavericks Caching Server very slow on gigabit Microsoft Network

Specs. OS X 10.9.5 / Mac mini / 8 GB Ram - Connected to Gigabit ethernet, 2 TB Hard Drives, 2.3 Ghz i7
Only a few days old. I have done all available updates that Apple released.
I had some difficulty getting the Mavericks caching server to work in the first place on our Windows Server 2012 network. Caching would work for the Server itself but none of my iPads or other Windows machines would cache or pull from the cache.
My work around was to manually create a DNS Host Name and PTR on the Windows Server so that we could actually nslookup and ping the Mavericks Server by hostname.domain.local This immediately fixed the issue and all hosts can cache and pull from the cache.
Speed is an issue. I currently cannot get any transfer speeds over 2 MB/s
Using the caching server to deploy an app (I verified that the internet connection was not being used and that it was coming from the caching server)
Connecting to a file share on the Windows Server and manually Copy/Paste files.
I verified the slowness using Activity Monitor on the Mavericks Server.
I have researched and found many similar complaints for the SMB2 protocol. I did not find a good enough Terminal Command to paste that I felt comfortable doing and at a later date undoing when Apple fixes the slowness issue.
Any help is greatly appreciated.
Does anybody know the protocol that Mavericks uses by default to send out the cached updates etc (SMB)?
If you know the correct command to paste into Terminal to force Mavericks to use SMB1 and also the command to undo this change I am willing to try that as well.
Thank you in advance,
TechJeff

I tweaked the DNS a little bit more...Under DHCP IPv4 Properties --> DNS Tab. I selected "Dynamically update DNS A and PTR records for DHCP clients that do not request updates". This allowed our DNS to register the Macintosh server and ipads etc. My SMB2 speed is up to 300 mbps from 16 mbps.
My caching server transfers are still maxing out at 2.1 MBps. I have switched out my wireless unit to a 150 mbps unit that the iPad is connected to. My next troubleshooting step will be to download an app in iTunes on a computer that is plugged in to our gigabit ethernet on the same switch as the Mac Server to elliminate any potential bottlenecks or wireless issues.
Still looking for ideas if anybody has troubleshooting ideas.

Mavericks Caching server 2 for all Apple IDs?

Good evening,
At Mavericks server 10.9.1, does caching server 2 on Mavericks server caches for any Apple ID that is ont the local network?
Francois.

The application and the authorization are independent. The app/book are just data. When the first system requests it, the file is transferred to the caching server and held there as it is passed to the client that requested it. iTunes, Appstore, or the iOS device will then talk to the authorization server and assign the asset to the users account. When the next user requests the same file, it can now be vended from the caching server but authorization is handled by Apple's servers. So there is still an outbound request but this is trivial compared to the size of the payload.
One to of the best ways to see this in action is with a large OS update. Update one device. Then when complete, review your caching server to see that the data is there. Then update another system. It downloads in a blink of the eye.
R-
Apple Consultants Network
Apple Professional Services
Author "Mavericks Server – Foundation Services" :: Exclusively available in Apple's iBooks Store

Maverick osx server and dnssec for dns master/slave services

Hi,
I have some questions to the maverick osx server.
Is this possible to configure dnssec keys for the maverick osx dns server?
Where are the configuration files from the dns server located?
Thomas

Okay - shortly after posting I found the checkbox "allow zone transfer" in the zone settings but there seems to be no way to control which secondary servers are allowed.
Anyhow I would like to know, where the config and zone files ar on Yosemite Server.

When using the OS X caching server, after the first update is made, do the rest of the computers have to enter the Apple ID to download the cached update

Hello,
I have a question that I had not found an answer to yet. I would like to use the OS X caching server, and I know that it caches a software update as soon as the first computer downloads it. If an application requires authentication using an Apple ID, and the first computer authenticates and downloads that update, do the rest of the computers have to authenticate as well, even though the update is downloaded from the caching server?
I am looking for a solution in that regard. For the company that I work, I would like a solution for users to be able to download software updates (OS or Application), without having to authenticate with the company Apple ID, for security reasons.
Would anybody be able to help?

Yes. You still need to authenticate on the subsequent machines if you are interactively applying the updates. Caching server stores the application data, not the App Store authorization information.
You have some alternatives.
1: If you are using a single "corporate" Apple ID to claim your free apps (iMovie, iPhoto, etc), then you can download them on your master machine and then user ARD, JAMF, or other tool to push the full application packages out to all the clients. They will already be tagged to the Apple ID and since the push is happening in the background, as long as the user is not using the application, they will not know you are updating. Remember, App Store updates are the entire application, not update packages. So pushing the entire .app bundle is effective.
2: If you are looking to distribute purchased apps, you really should look at the VPP (volume purchase plan) from Apple (http://www.apple.com/business/vpp/) (there is also one for education but you mentioned company). This allows you as the organization to purchase the correct number of copies and then control the distribution of the apps to your end users. They can use their own Apple ID to claim the apps but you can reclaim the license should the user leave the organization.
Reid
Apple Consultants Network
Apple Professional Services
Author "Mavericks Server – Foundation Services" :: Exclusively available in Apple's iBooks Store

Cache server getcert http request issue

Hi,
We are using third party repository for documents archiving and storage. And we have separate application for connecting third party repository to the SAP.
We have installed content server and cache server on the system where SAP was installed.
Now we want to run getcert HTTP request from our application so that we will get the certificate from SAP cache server. What we need to do in order to achieve the same? Or in other words how we can make the connectivity between cache server and the third party repository?
We have verified the following URLs
http://10.224.1.37:1090/ContentServer/ContentServer.dll?serverInfo
http://10.224.1.37:1095/Cache/CSProxyCache.dll?serverInfo
(Where 10.224.1.37 is the IP address of the system where SAP is installed)
for content server and cache server respectively.
The URL for cache server was giving the correct server information but URL for content server is not at all showing any server information even after running for long time.
Could you please tell me step by step by procedure and configuration steps for running get cert request to cache server? We want to know how to send the getcert request to third party content management system from cache server.
We have given the following URL in http get functional module but we are getting a 400 bad request response
http://10.224.1.37:1095/Cache/CSProxyCache.dll?getcert&pversion=0046&conrep=RH
Where RH is pointing to the third party content server (through transaction oac0).
Thanks,
Ravi

Hi All,
I am also facing the same problem, Please help us out to solve this issue.
Thanks in advance
Regards
Harshavardhan.G

Deski Cache Server Security switch

Hi;
I would like to know if there a security switch for Deski cache Server as it exists for CR Cache Server. We have problem with delay in right change in Deski. A similar case has been identified, and the solution was to add the (-securityCacheTimeOut 0) switch to the start command line of CR Cache Server and that has helped resolving the problem. I am asking if there is a similar switch for Deski Cache Server.
Thanks in advance

Hi!
First of all you can enable security constraints for +/mapviewer/*+ pages in you app server.

Installing Live Cache Server on SCM 4.1 system

Hi All,
I have installed an SCM 4.1 Abap + Java System and Live cache client was installed during SCM 4.1 installation.
I would like to install Live cache server and i have problems with it.
I have downloaded the following installation guide
SAP liveCache 7.5.0 For SAP SCM Server 4.1
Document Version 1.02 u2013 April 8th, 2005
Guide tells us to use installation master CD for installating Live cache Server but note 708118 tells us to use the CD 51031447
04 Nov 05 Windows AMD64: Start Installation from liveCache CD
Start the installation of the liveCache server or client from the liveCache
CD (number 51031447) and not from the installation master CD (number
51031443)
Upon checking the cd 51031447 i see SDBINST.exe and upon clicking it i get the following screen
Installation of SAP LiveCache Software
existing profiles
0: APO Livecache
1: Runtime for SAP AS
2: DB Analyzer
3: JDBC
4: Server
5: Loader
6: ODBC
7: All
8: None
Please enter profile id :
I select profile 0 and it comes back with another user input option
Im searching for a document which gives me the procedure for installing Live cache using SBDINST.exe
Regards,
Ershad Ahmed.

Hi All,
I went ahead with Option 0 and it asked me for Kernel file and i provided the files.
I checked logs in /sapdb/wrk directory and i see that Installation of SAP livecache finished successfully.
From LC10 i gave database connection name (LCA) and clicked on live cache monitoring and it comes back with an error
Name and Server : LCA - XXXXXXX (XXXXX = Server Name)
DBMRFC Function: DBM_EXECUTE
Command: dbm_version
Error: DBM Error
Return Code : -4
Errror Message: database instance not found.
Application Server: XXXXXX                     ( Windows NT )
1. Connect. test with "dbmcli db_state"
     Error! Connection failed to node XXXXXXX for database LCA: database instance not found
     External program terminated with exit code 2
2. Connect. test with command mode "dbmrfc db_state"
     Name and Server     : LCA - XXXXXXX
     DBMRFC Function     : DBM_EXECUTE
     Command             : db_state
     Error               : DBM Error
     Return Code         :         -4
     Error Message       : database instance not found
3. Connect. test with session mode "dbmrfc db_state"
     Name and Server     : LCA - XXXXXXXX
     DBMRFC Function     : DBM_CONNECT
     Error               : DBM Error
     Return Code         :         -4
     Error Message       : database instance not found
4. Connect. test with "native SQL" ( LCA )
     For detailed information, see the developer trace for work process:          1
From the OS level i gave command dbmcli db_enum and i get OK as reply.
Can someone please let me know how to resolve this issue.
Regards,
Ershad Ahmed.

Caching Server receives non-whitelisted requests

I run a OSX Caching Server on a 10.8.3 Mountain Lion System. While this update server solution works fine for most of the client computers, one client causes unusual requests:
2013/04/01 09:51:36:104 Request for non-whitelisted URL denied (http://192.168.42.3:53783/)
THis message is generated from each attempt to access a Mac Appstore Download (this one was for the Final Cut Pro X 10.0.8 update) and after using this URL, the client simply ignores the caching server and moves on downloading this assets from Apples source servers.
The client is a Protina MBP running the official 10.8.3 build.
Regards,

I have this exact problem too. Interestingly, I have this on my Mini server, then I've setup a caching server on my MacBook Pro and its the same. This log is from my MacBook Pro, its called badgerbookpro.local but its the same on the Mini server. Both are runing 10.8.3
Jun 1 01:34:22 badgerbookpro.local AssetCache[4811]: Caching server started
Jun 1 01:34:25 badgerbookpro.local AssetCache[4811]: Registration succeeded. Resuming server.
Jun 1 01:37:02 badgerbookpro.local AssetCache[4811]: Request for non-whitelisted URL denied (http://192.168.1.48:62249/)
Jun 1 01:37:02 badgerbookpro.local AssetCache[4811]: HTTP Server: Error 400 - Bad Request (/)
Jun 1 01:38:23 badgerbookpro.local AssetCache[4811]: Request for non-whitelisted URL denied (http://192.168.1.48:62249/)
Jun 1 01:38:23 badgerbookpro.local AssetCache[4811]: HTTP Server: Error 400 - Bad Request (/)
Jun 1 01:43:51 badgerbookpro.local AssetCache[4811]: HTTPConnection[0x7fdc04b1b900]: responseHasAvailableData: - Sender is not current httpResponse
Jun 1 01:44:46 badgerbookpro.local AssetCache[4811]: Request for non-whitelisted URL denied (http://192.168.1.48:62249/)
Jun 1 01:44:46 badgerbookpro.local AssetCache[4811]: HTTP Server: Error 400 - Bad Request (/)
Jun 1 01:45:05 badgerbookpro.local AssetCache[4811]: Request for non-whitelisted URL denied (http://192.168.1.48:62249/)
Jun 1 01:45:05 badgerbookpro.local AssetCache[4811]: HTTP Server: Error 400 - Bad Request (/)
Jun 1 02:51:24 badgerbookpro.local AssetCache[4811]: Request for non-whitelisted URL denied (http://192.168.1.48:62249/)
Jun 1 02:51:24 badgerbookpro.local AssetCache[4811]: HTTP Server: Error 400 - Bad Request (/)
Jun 1 03:55:04 badgerbookpro.local AssetCache[4811]: Server shutting down (15)
Jun 1 03:55:04 badgerbookpro com.apple.launchd[1] (com.apple.AssetCache[4811]): Exited with code: 15

Architectural Choice for accessing Coherence Cache Server

I am a newbie and have a coherence use-case question.
When accessing an independent coherence cache server from application code such as an EJB deployed in WLS, architecturally does one write up an entity which is then used as a sole point of
access to the resource (coherence cache server) for querying, adding, modifying entries or are the accesses to the coherence cache server split and spread among the application code.
For example,
1. I write an EJB (EJB 1) which recieves the request from other EJB's (EJB 2, EJB 3), EJB 1 runs requests from EJB2 and EJB 3 on the Coherence Cache Server and acts as the sole point of
contact to the resource.
2. EJB 2 and EJB 3 both run requests against the coherence cache server. No fixed entity in architecture repsonsible for interaction with Coherence Cache Server.
Which is more common ?

stevephe wrote:
Yes you could treat Coherence as a "pluggable" resource, just like a database. But that, just like in the case of a database, wouldn't boil it down to a single entity/interface. You'd treat Coherence as a "integration tier" resource that you'd "plumb in" just like you would a database, thus shielding your application's "domain" objects from integration-level concerns. That's how I've tiered our application, although we aren't inside a container like Weblogic/WebSphere/etc. The domain objects specify their persistence requirements via a multiplicity of interfaces; those interfaces have a number of implementations in the integration tier, one set of which just happens to be a Coherence set. You can use a "registry" approach to pick up the appropriate implementations (we use Spring injection.) Have a look at the Coherence book from Apress for more details.Apress? You mean Packt, don't you?
Best regards,
Robert

Customized Cache Config and Coherence Override files not Overrided.

Hi,
I am doing an example on coherence and was wondering if anyone can help with a couple of questions.
1. I am trying to override the coherence-cache-config.xml and tangosol-coherence-override.xml with my custom xmls. But when I start cache server from cache-server.cmd executable provided in bin folder of Coherence, it is not able to pick up the overrided files. How can I override both these files.
2. In most of the examples a new executable file is made for cache server start and the custom coherence-cache-config.xml is specified in this. The examples runs perfectly fine but I'm not able to see the same cache elements available when I run cache-server and cache-client present in the bin folder of Coherence. How can I integrate both of them.
3. When we use Coherence as a web service from server, are we suppose to use run cache-server.cmd for starting the cache server or make some custom file executable file made for cache server start.
I know some of the questions may sound silly to you but I'm not still able to figure the answer to these question.
Any help will be appreciated.
Cheers,
Varun

Hi Varun,
Yes you can use the property -Dtangosol.coherence.cacheconfig =<file_name> for specifying the override file. No creating a jar is not a bottleneck and you can use either jar or system property in Production Environment. It completely depends on how your environment is setup and is for the flexibility pruposes? Use whichever is handy ...
Ideally, you WS application should run in a seperate JVM than your Coherence Nodes. In this scenario, your WS application is set as storage-disabled and your Coherence nodes will be set as storage-enabled and both WS and Coherence Nodes will be part of the Coherence cluster. There will be no syncing between WS application and Coherence Nodes if you are using distributed cache; all the data requests from the WS application will be served by the Coherence Nodes. But, if you are using Near Cache then the data syncing will depend on the invalidation strategy used by you in your cache configuration. Read more about near-cache scheme here: http://wiki.tangosol.com/display/COH35UG/near-scheme.
In WLS 10.3.4, there is a integration of Coherence and you can create Coherence Cluster from the Admin Console and define various Coherence cluster nodes and the documentation is here (http://download.oracle.com/docs/cd/E17904_01/web.1111/e16517/coh_wls.htm). But if you are using any other version of Weblogic or any other container then you cannot start from Weblogic but you need to start them seperately. Also, you need to make sure that the Coherence Nodes with storage-enabled are started before the storage-disabled nodes.
Hope this helps!
Cheers,
NJ

Some Crystal Reports are not using the Cache Server

Hi...we are currently running BOE XI R2 SP2. I setup a server group with a Cache Server and a Page Server. In the CMC, I set this cache server to keep reports cached for 15 minutes. On the Process tab of some reports, I specifically set the report to use this new server group. On some of the reports this works as expected. For others, it does not appear to be since the cache folder's Date Modified date does not change. So far the only similarity between the reports that do not use cache is that these reports have sub-reports. Has anyone experienced this before? If so, how did you resolve it?
Thanks!

Hi Rich,
the connectivity on the Designer an don the BusinessObjects system is identical. Is region the only characteristic being used in the report ? which connectivity did you use in Crystal Reports ?
ingo

Caching Server - Caching Mavericks Installer

Is there a way to allow the caching server in Server 3.0 to cache Mavericks 10.9 to limit external bandwidth?

I tweaked the DNS a little bit more...Under DHCP IPv4 Properties --> DNS Tab. I selected "Dynamically update DNS A and PTR records for DHCP clients that do not request updates". This allowed our DNS to register the Macintosh server and ipads etc. My SMB2 speed is up to 300 mbps from 16 mbps.
My caching server transfers are still maxing out at 2.1 MBps. I have switched out my wireless unit to a 150 mbps unit that the iPad is connected to. My next troubleshooting step will be to download an app in iTunes on a computer that is plugged in to our gigabit ethernet on the same switch as the Mac Server to elliminate any potential bottlenecks or wireless issues.
Still looking for ideas if anybody has troubleshooting ideas.

Installed mavericks server and now some users don't have access to folders!? Why and how do I fix it???

I have just upgraded the server and os to mavericks and now users that used to be able to open and save or even log in are now locked out with the server saying wrong password. I reset their password and still nothing. I created a new user and they can't even log in? Why now? And how do I fix this. It seems that some users and groups now don't have access and all that was done was to let the group open a shared folder that they have access before. I reset everything and am now very frustrated!!!

Hi
Seem to be an issue with Maverick
After trying to solve the issue with all tips i found here and on the net
I decided to go back to Mountain Lion for my server using TimeMachine backup, and stay with Maverick for my imacs
Work well again
Until Apple will communicate on it or deliver a patch, i would suggest to do the same

How can I create a client console and work together with the Cache Server?

How can I edit the following Cache-Server.cmd file to create a client console and work together with the Cache Server?
The following is the cache server file: contacts-cache-server.cmd
@echo off
setlocal
if (%COHERENCE_HOME%)==() (
set COHERENCE_HOME=c:\coherence
set CONFIG=C:\home\oracle\coherence\Contacts
set COH_OPTS=%COH_OPTS% -server -cp %COHERENCE_HOME%\lib\coherence.jar;C:\home\oracle\
coherence\Contacts;C:\home\oracle\coherence\Contacts\classes;
set COH_OPTS=%COH_OPTS% -Dtangosol.coherence.cacheconfig=%CONFIG%\contacts-cache-config.xml
java %COH_OPTS% -Xms1g -Xmx1g -Xloggc: com.tangosol.net.DefaultCacheServer %2 %3 %4 %5 %6 %7
:exitEdited by: junez on 23-Oct-2009 09:20

Hi
To run the console, change DefaultCacheServer to CacheFactory
Paul

Mavericks Caching Server and PAT

Similar Messages

Maybe you are looking for