Local Cache Visibility from the Cluster

Hi, can you give me an explanation for the following Coherence issue, please ?
I found in the documentation that the Coherence local cache is just that: a cache that is local to (completely contained within) a particular cluster node and is accessible from a single JVM.
On te other hand, I also found the following statement:
“ Clustered caches are accessible from multiple JVMs (any cluster node running the same cache service). The cache service provides the capability to access local caches from other cluster nodes.”
My questions are:
If I have local off-heap NIO memory cache or NIO File Manager cache on the one Coherence node, can it be visible from other Coherence nodes as a clustered cache ?
Also, if I have NIO File Manager cache on a shared disk, is it possible to configure all nodes to work with that cache ?
Best Regards,
Tomislav Milinovic

Tomislav,
I will answer your questions on top of your statements, OK?
"Coherence local cache is just that: a cache that is local to (completely contained within) a particular cluster node and is accessible from a single JVM"
Considering the partitioned (distributed) scheme, Coherence is a truly peer-to-peer technology in which data is spread across a cluster of nodes, the primary data is stored in a local JVM of one node, and its backup is stored in another node, preferably in another site, cluster or rack.
"Clustered caches are accessible from multiple JVMs (any cluster node running the same cache service). The cache service provides the capability to access local caches from other cluster nodes"
Yes, no matter if the data is stored locally in a single node of the cluster, but when you access that data through its key, Coherence automatically finds that data in the cluster and brings to you. Its transparently for the developer the location of data, but one thing is certain: you have a global view of caches, meaning that from every single member, you have access to all data stored. This is one of the magic that the Coherence protocol (called TCMP) does for you.
"If I have local off-heap NIO memory cache or NIO File Manager cache on the one Coherence node, can it be visible from other Coherence nodes as a clustered cache ?"
As I said earlier, yes, you can access all the data stored from any node of the cluster. The way in which each node store its data (called as backing map scheme) can differ. One node can use an elastic data as backing map scheme, and another node can use Off-Heap NIO Memory Manager as backing map. This is just the way about each node store its data. For the architectural point of view, its a nice choice to use the same backing map scheme across multiple nodes, because each backing map scheme can have different behaviors when you read and/or write data. One could be faster and another could be slower.
"Also, if I have NIO File Manager cache on a shared disk, is it possible to configure all nodes to work with that cache ?"
There is no need for that, since data is available to all cluster nodes without any effort. Having said that, this would be a bad strategy choice. Coherence is a shared-nothing technology which uses that model to scale and give you predictable latency. If you start using a shared-disk as storage for data, you will lose the essence of shared-nothing benefits, and create a huge bottleneck in the data mgmt layer, since will occur dispute per I/O in each read/write.
Cheers,
Ricardo Ferreira

Similar Messages

IPod Classic 160GB film library sync problem - trying to sync films that have converted from purchased DVD to mp4 format locally, not purchases from the online store - all other libraries sync perfect - why?

Hello Apple and the iTunes Windows PC users community.
I am trying to sync films that I have converted from purchased DVD to mp4 format locally, not purchases from the online store. The mp4s all appear and play successfully in my iTunes application but will not sync across to the iPod film library folder.
For your information: I am using iTunes 11.1.3.8 on a Windows 7 64bit machine with 500GB hard disk and 8GB of Ram in the UK.
I have restored the iPod classic 160GB three times now to see if it was a hardware problem with no joy. Each time all the music restores properly as do the podcasts and all the items in TV programmes all appear to sync and work fine.
I have also tried to copy films into the TV Programmes to get around it with no joy. They always go to the films section to start with. It is just the Films library that does not sync - all others work perfectly. As a last resort I have uninstalled and re-installed iTunes with no joy either.
I am technically savvy and have gone through the itunes and ipod settings but nothing appears to make a difference - This is the first time I have had to post here as I can usually solve the majority of the ipod anomilies but this one has me flummoxed.
Has the film encoding type changed in the newest itunes update? - Has this happened to anybody else and is it a hardware, software, or operating system problem.

Having uninstalled the current version 11.1.3.8 and loaded and older version of iTunes 10.7.0.21 I can now categorically confirm that the newest update seems to be causing the problem as the films and TV Programmes are syncing perfectly on this older version.
If you are going to do this please dont forget to remove the ' iTunes Library.itl' file as this stops the older versions from running as I've just found out

Node removed from the cluster and cannot joined to clustr again

Hi
I have problem with 2 node cluster. we are running windows 2008 R2 enterprise edition. Suddenly one node had some issue with network connnectivity and it was removed from the cluster. on the console its shows node1 offline. connectivity between the
server public and private nic is fine.
Node 'node1' failed to form a cluster. This was because the witness was not accessible. Please ensure that the witness resource is online and available.
Cluster resource 'File Share Witness' in clustered service or application 'Cluster Group' failed.
'File Share Witness is accessible to both the nodes. from the cluster log i can see failed to attempt lock on the FSW.
I have restarted the failure node. still the problem exist.
Thanks in advance to throw some valuable answers to my issue.

Are you using Volume Manager Disk Group. Please check the below link if you are using it.
http://www.symantec.com/business/support/index?page=content&id=TECH127796
Thanks,
Umesh.S.K

No Connection to Integration Builder (only local data visible): from R/3

Hi,
I am also facing the same problem. I maintained the RFC destinations in R/3 properly. I did Test Connection and for RFC destination its working fine. I maintained proper values in SLDAPICUST. When I run SLDCHECK I got the following error.
Calling function LCR_LIST_BUSINESS_SYSTEMS
Retrieving data from the SLD server...
Function call returned exception code 4
=> Check whether the SLD is running!
Summary: Connection to SLD does not work
=> Check SLD function and configurations
Otherwise access to XI Profile works fine. But then when I run SPROXY I am getting the same message No Connection to Integration Builder (only local data visible)
I also checked the following report =>Check with report SPROX_CHECK_IFR_RESPONSE (3rd option) its giving informational message Integration Builder data not understood, for the 4th report ==>Check with report SPROX_CHECK_IFR_CONNECTION, it is also showing the same above information message.
Looking forward for the response,
Appreciate your help,
Thanks and Regards,
Jagadish

Hi,
I had the same problem and I have already solved it.
You have just to apply the following sap notes :
1232805 - Saint: Queue calculation for an add-on with CRT´s fail.
769120 - Support Package for APPINT 200_620
The first one is to update tha SPAM/SAINT to v 30 and the second one allow you use the pi/xi objects on application system.
There is another sap note 689847, but you just apply this if the sap notes above described did not work.
Paulo Correa

No shared disks visible in the Cluster Configuration Storage dialog

When installing the Oracle 10g clusterware the "Cluster Configuration Storage" dialog shows no shared disks.
We are using:
Windows 2003 Server
HP Eva 4400

Hello,
all disks in cluster are visible from all nodes (2 of them).
We tested it with unpartioned and partioned disks (primary and extended). No way to make them visible for the OUI.
Automount is enabled in Windows like required from Oracle.
Besides, we are using Standard Edition. Therefore we have to work with ASM.
Any more information needed.
Thanx in advance.

Help required for temporary removing node 2 from the cluster

Hi Support,
There are some hardware issues on node 2 of 11g RAC cluster and vendor going to fix it;
We want to make node 2 not available until they fix it;
The user will continue with node 1 of the cluster;
Following are the current status of the site and I'm considering to perform the folowing task
while the folllowing command from node 1, i got the following output
D:\srvctl status database –d EPVAL
Instance epval1 is running on node corskmaps001
Instance epval2 is running on node corskmaps001
D:\srvctl status database –d EPPROD
Instance epprod1 is running on node corskmaps001
Instance epprod2 is running on node corskmaps001
I made the following steps in chorological order; I appreciate if someone can correct me if I made any mistake or add/amend/update as per requirement.
Assuming corskmaps002 is our passive mode which will be taken away by HP for hardware checking;
srvctl stop instance –d EPPROD -i EPPROD2
srvctl stop instance –d EPVAL -i EPVAL2
srvctl stop listerner_corskmaps002 –n corskmaps002
srvctl stop nodeapps -n corskmaps002
Do we require stopping service also once we stop the instance? Following are the commands if your answer is yes;
srvctl stop service -d EPROD -i EPPROD2
srvctl stop service -d EPVAL -i EPVAL2
srvctl disable service -d EPPROD -s EPPROD2
srvctl disable service -d EPVAL -s EPVAL2
I appreciate help
Thanks

Hi Bjoern,
I agree with you, very valid point, thanks for that;
Just for clarification, do you thing the sequence should be like this?
********disable command to start with***********
srvctl disable service -d EPPROD -s EPPROD2
srvctl disable service -d EPVAL -s EPVAL2
*********than instance stop commad**********
srvctl stop instance –d EPPROD -i EPPROD2
srvctl stop instance –d EPVAL -i EPVAL2
************than listener stopping command********
srvctl stop listerner_corskmaps002 –n corskmaps002
*************than node stopping command*********
srvctl stop nodeapps -n corskmaps002
*********do we also require stopping of the service ?>***************************
Do we require stopping service also once we stop the instance? Following are the commands if your answer is yes;
srvctl stop service -d EPROD -i EPPROD2
srvctl stop service -d EPVAL -i EPVAL2
Can you please eleborate further
Then, you also might want to add '-o transactional' or "transactional local" to the
instance stop command so that running transactions are not affected by this.
Thanks

Splitting the app traffic from the cluster and admin traffic

Hi,
We currently have a 10.3.2.0 setup where an admin server is behind a firewall and is running on the Administration Port and is connected to two managed servers in front of a first firewall in the DMZ.
On each managed server there is two network interfaces one for management and one for apps/DB connections.
So I was wanting to know how to you get the managed server to split the traffic??
If I set the managed servers listening address to the management interface then it starts up fine as that what the admin server behind the firewall can see, BUT it means app and JDBC Connections dont work and the server goes to ADMIN state first before you have to manually resume it.
If I set the managed servers listening address to the app interface then it can't start up and the app interfaces address is block from the admin server to the managed server via the firewall.
If I set the managed server to have no listening address and leave it blank in the interface field, it starts up and listens on all interfaces, BUT can't find a route to the admin server.
So what is the answer, can you do something with Network Channels?? Or is it the case you just can't do it and just to have one interface and one listening address as the admin traffic is split by the default administration channel anyway.
Would be get to know.
Alistair.

Are you using cellular data? If yes, try to use Wi-Fi and see if the app works better by using a faster data connection. If you have no problem using Wi-Fi and maps, see if you can have a faster cellular connection by switching to 3G, 4G to LTE service in Settings/Cellular.

How to open local system folder from the browser

Hi All,
I'm working on some stuff, wherein the user downloads the files from internet to his local machine. I'm able to catch the path of the folder where user wants to download the file from SaveAs prompt. Now my requirement is , i want to provide a link button or
a button on the web page and assign this is folder path dynamically so that when the user clicks on it, it should open the local system folder where he has downloaded the files.
string myDocspath
=Environment.GetFolderPath(Environment.SpecialFolder.MyDocuments);
string windir
= Environment.GetEnvironmentVariable("WINDIR");
System.Diagnostics.Process prc
= new System.Diagnostics.Process();
prc.StartInfo.FileName = windir + @"\explorer.exe";
prc.StartInfo.Arguments = myDocspath;
prc.Start();
Work fine with local system, if i host the application in IIS , the above code snippet wont work
Thanks in advance,
Suraj Kumar
Software Engineer

Hello,
local resources is disabled in all modern browsers due to security restrictions.
if the reply help you mark it as your answer.
Free No OLE C#
Word, PDF, Excel, PowerPoint Component(Create,
Modify, Convert & Print)

How to get the data from a cluster table to BW

Dear All,
I want to extract the data from R/3 to BW by using 2 tables and one Cluster B2.
Actually my report contains some fields from PA2001, PA2002 and one cluster table B2 (Table ZES). Can I create View by using these 3 tables? If it is not possible how can I get the data from the cluster? Can I create generic datasource by using cluster tables directly?
In SE11 Transaction the Cluster (table ZES) is showing invalid table.
I referred some Forums, but no use.
Can any body tell me procedure to get the data from a cluster (table ZES) ?
Waiting for you results.
Thanks and regards
Rajesh

HI Siggi,
Thank you for your reply..
I am also planning to do FM to get the data. But it is saying that the Cluster table ZES does not exist (ZES is the the standard table, in SE11 also).
How can I use the Fields from the that table.?
What can I do now, can you please explain me about this point.
Waiting for your reply.
Thanks and Regards
Rajesh
Message was edited by:
rajesh

Is it possible to move the local CD database from Windows to Mac?

I have/had iTunes installed on a Windows box. Windows knows all my CD and track titles (some customised from those in the online CDDB). There are more than 250 CDs, so it is a lot of data (if I have to retype it all...)
I have recently moved to a Mac. The ripped library files copied across just fine (even kept the album artwork).
However, the Mac doesn't know any CD/track titles, so when I insert a CD, there are no names. The Mac is not on the net, and probably never will be. I want to move the copy of the local CD database from the windows machine to the Mac. Is this possible? How?
Dual G4 Tower Mac OS X (10.3.9) Also Windows XP
Dual G4 Tower Mac OS X (10.3.9) Also Windows XP

I imagine you will need to network your PC with your Mac to transfer or transfer your Library from your PC to an external drive then hook this up to your Mac and transfer.

How to determine client locale from the mail?

How to determine client locale(language/country) from the mail(header)?
I investigate that some mails have not got any content-language header...
Also, If mail is in multipart , I cannot get it throught the contentType....
e.g some header contain header like this: content-Type: "text/html; charset=Big5"
but some are : content-Type: multipart/mixed; boundary......";
So, how can I know the client locale?
Thanks

You should be able to get the charset from the content type, as u said. the reason some types don't have a charset, such as multipart/* types, is because they don't contain any content on their own! they only contain nested body parts, each of which should have a charset specified. you're gonna have to get the charset by following the mail structure recursively.

Inaccessible IronPort c670 appliance from both GUI and CLI. Can I tunnel-in from other c670 appliance in the cluster?

I am unable to access one of the 6 IronPort appliances. Also, got an error that the appliance has got disconnected from the cluster. Is there any CLI command I can use to possibly tunnel-into the faulty appliance from another appliance in the cluster to reboot?

No. Normally, from 'clustermode' you can then access machine level on the different appliance(s) in cluster. But, if this is disconnected, then that command is not going to work.
If you cannot SSH/telnet to the appliance in question, and do not have some form of remote console or access pre-configured, you will need to connect directly to the appliance, or hard boot the appliance in order to attempt to regain connectivity.
Please see the C670 quick start guide for assistance:
http://www.cisco.com/c/dam/en/us/td/docs/security/esa/hw/C670_QSG.pdf
Setup and Management
•• For access by Ethernet™, connect to the Management Network
Port. Use a browser to access the web-based interface on the
default IP address 192.168.42.42. You can also access the
command line interface by SSH or terminal emulation software
on the same IP address. (The netmask is /24.)
•• Or, for Serial access, connect to the Serial Port. Access the command
line interface by a terminal emulator using 9600 bits, 8 bits, no parity,
1 stop bit (9600, 8, N, 1), flowcontrol = Hardware.
I hope this helps!
-Robert
(*If you have received the answer to your original question, and found this helpful/correct - please mark the question as answered, and be sure to leave a rating to reflect!)

Clean way to De-register/De-commission the database with EBS from RAC/Cluster.

Hi,
We have a database with EBS on RAC. We are looking for Clean way to decommissioning the database from the cluster. We tried earlier, but when we again installed a fresh database on that server some how we had issues with the user oracle in cluster configuration files.
Early response is highly appreciated.
Thanks,
Rajashekar

Hi Hussein,
We have two node db and two node application server in a cluster environment.
We want to Remove/Decommission JUST DATABASE, APPLICATION related all files from the cluster. Cluster files should not be effected. After removing them we want install a fresh Db and Application on the same cluster.
Looking for any standard method for decommissioning.
Thanks,
Rajashekar

HTTP 400 Bad Request when accessing Oracle XE/Apex from the Internet

I have Oracle Express Edition upgraded to Apex 3.2. I can access everything (SQL prompt, Apex HTTP) from my local machine. I have configured the server for remote access on a non-default (not 8080) port.
I have executed: exec dbms_xdb.setListenerLocalAccess(false);
I can access the server using:
http://coyote:9977/apex (brings me to Application Express Administration Services login screen)
(where "coyote" is the local machine name of my Windows Vista box, where the Oracle/Apex server is installed.)
But when I try:
http://internet_server_name:9977/apex
(where "internet_server_name" is the internet DNS name visible of my machine visible from the internet.)
I get "HTTP 400 Bad Request"
I know the request is hitting the Oracle Listener (getting through firewalls, cable modem router, etc) since the http attempt causes the entry:
20-MAR-2009 20:36:53 * http * (ADDRESS=(PROTOCOL=tcp)(HOST=68.189.244.22)(PORT=52540)) * handoff * http * 0
to be made in the C:\oraclexe\app\oracle\product\10.2.0\server\network\log\listener.log
An attempt using the local (Windows) machine name makes a similar entry in the listener log:
20-MAR-2009 20:47:35 * http * (ADDRESS=(PROTOCOL=tcp)(HOST=192.168.1.7)(PORT=52811)) * handoff * http * 0
This seems like a security issue of some kind (is the server dropping a request from "outside" the local domain?)
I'm new to Oracle so I am not familiar with how to debug this sort of connectivity issue. In case it helps:
LSNRCTL> status
Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=IPC)(KEY=EXTPROC_FOR_XE)))
STATUS of the LISTENER
Alias LISTENER
Version TNSLSNR for 32-bit Windows: Version 10.2.0.1.0 - Production
Start Date 20-MAR-2009 19:27:00
Uptime 0 days 1 hr. 30 min. 59 sec
Trace Level off
Security ON: Local OS Authentication
SNMP OFF
Default Service XE
Listener Parameter File C:\oraclexe\app\oracle\product\10.2.0\server\network\admin\listener.ora
Listener Log File C:\oraclexe\app\oracle\product\10.2.0\server\network\log\listener.log
Listening Endpoints Summary...
(DESCRIPTION=(ADDRESS=(PROTOCOL=ipc)(PIPENAME=\\.\pipe\EXTPROC_FOR_XEipc)))
(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=coyote)(PORT=1521)))
(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=coyote)(PORT=9977))(Presentation=HTTP)(Session=RAW))
Services Summary...
Service "CLRExtProc" has 1 instance(s).
Instance "CLRExtProc", status UNKNOWN, has 1 handler(s) for this service...
Service "PLSExtProc" has 1 instance(s).
Instance "PLSExtProc", status UNKNOWN, has 1 handler(s) for this service...
Service "XEXDB" has 1 instance(s).
Instance "xe", status READY, has 1 handler(s) for this service...
Service "XE_XPT" has 1 instance(s).
Instance "xe", status READY, has 1 handler(s) for this service...
Service "xe" has 1 instance(s).
Instance "xe", status READY, has 1 handler(s) for this service...
The command completed successfully
Any help appreciated,
john

Hans Forbrich wrote:
CoyoteTech wrote:
I have tried running with the firewall (F-Secure) completely disabled, but it made no difference. I do see the request hit the listener, and the handoff appears to be successful, but it goes silent from there. I also have a cable modem router that has the port forwarding set, but maybe there are other handoff ports besides the initial one (e.g. 8080) that need to be forwarded?Please also check the Windows firewall is off as well. Yes, Windows Firewall is disabled as well. See my previous post to Tyler - I'm pretty sure the reqwuest makes it through the firewall/router layer, since it causes an entry in Listener.log
Is your problem through the router only? In other words, if you try from a second machine that is on the same side as your XE, are you working OK? (My home ISP blocks a bunch of ports. Including SMTP, FTP, Telnet. And 8080 since that is a traditional default for many Java 'servers'.)
Yes, it appears to be related to a non-local IP address. I'm sure it makes it through the router - I also have several other ports that make it through (e.g. port 80).
I have poked around the DBMS_EPG docs, and ran a few of the commands there (list DADS etc). DBMS_EPG was first 'released' to us in 10gR2. As far as I'm concerned, XE is the public 'beta' or 'release candidate'. (This is supported by Oracle's docs that state that Apex using DBMS_EPG is first supported using 10.2.0.3 or 11g.) There are known bugs and there is no way to patch XE to fix them. However, I do not know whether you are hitting those bugs.
http://www.astral-consultancy.co.uk/cgi-bin/hunbug/doco.cgi?11410 provides some good notes.
Interesting. I will check out these docs later today.
>>
What happens when the listener hands off to a registered handler?
http://download.oracle.com/docs/cd/B19306_01/network.102/b14212/concepts.htm#i1049597 gives a good description.
Thanks.
How do I pick up the [log] trail from there?I'm thniking that the listener's log is still the primary way to go. You may need to increase the log or trace level. I have not investigated how to gen or access logs inside EPG.I tried Tyler's trace but did not seem to add additional info to the Listener.log.

Local Cache containing all Distributed Cache entries

Hello all,
I am seeing what appears to be some sort of problem. I have 2 JVMS running, one for the application and the other serving as a coherence cache JVM (near-cache scheme).
When i stop the cache JVM - the local JVM displays all 1200 entries even if the <high-units> for that cache is set to 300.
Does the local JVM keep a copy of the Distributed Data?
Can anyone explain this?
Thanks

hi,
i have configured a near-cahe with frontscheme and back scheme.in the front scheme i have used local cache and in the back scheme i have used the distributed cache .my idea is to have a distributed cache on the coherence servers.
i have 01 jvm which has weblogic app server while i have a 02 jvm which has 4 coherence servers all forming the cluster.
Q1: where is the local cache data stored.? is it on the weblogic app server or on the coherence servers (SSI)..
Q2: although i have shutdown my 4 coherence servers..i am still able to get the data in the app.so have a feel that the data is also stored locally..on the 01 jvm which has weblogic server runnng...
q3: does both the client apps and coherence servers need to use the same coherence-cache-config.xml
can somebody help me with these questions.Appreciate your time..

Local Cache Visibility from the Cluster

Similar Messages

Maybe you are looking for