Client TCP extend starts a new cluster

Similar Questions

• Multiple TCP * extend Proxy

  Hello
  
  I'm developing a C++ Client for consistency and using the TCP protocol * Extend to connect to the Cache.
  
  I am able to run multiple nodes on the same machine.
  
  My goal is to add more machines to this configuration (is another machine called cluster or node?).
  
  
  I would like to have a configuration as explained below:
  
  NamedCache: AQRCache
  Cache Type: Close cover (Local and distributed/remote)
  
  HOST_A:
  TCP * extend A
  Node 1
  Node 2
  
  C++customera performer on Host_A talk to TCP * Extend_A.
  
  
  HOST_B:
  TCP * extend_B
  Node 3
  Node 4
  
  C++clientrunning on Host_B B talk to TCP * Extend_B.
  
  
  Host_C:
  TCP * extend_C
  Node 5
  Node 6
  
  C++clientC running on Host_C talk to TCP * Extend_C.
  
  
  Other Clients in C++:
  
  C++clientrunning on Host_D talk to TCP * Extend_A.
  C++clientE running on Host_E talk to TCP * Extend_A.
  C++clientF running on Host_F talk to TCP * Extend_A.
  
  
  Issues related to the:
  
  1. to add more machines, if I am running an instance on another machine with the same configuration file, it will work?
  
  2. is it possible to run separate TCP * Proxy extend on each host and it will be part of the same cluster?
  
  3 or should all customers C++ must talk to only TCP * extend Proxy?
  
  
  Thank you
  NS

  1. to add more machines, if I am running an instance on another machine with the same configuration file, it will work?

  Yes, you can add other machines/nodes using the same configuration file.

  2. is it possible to run separate TCP * Proxy extend on each host and it will be part of the same cluster?

  Yes, you can run several proxy nodes in a single cluster. Just about every production cluster has several proxy nodes.

  3 or should all customers C++ must talk to only TCP * extend Proxy?

  No, there is no reason to force all C++ clients to connect to a single proxy. In fact you might consider to configure each client with the complete list of proxy servers. Each customer is going to randomly select a server to connect, ensuring that a single proxy is not overloaded.

  Also take a look at this document: http://coherence.oracle.com/display/COH35UG/Best+Practices+for+Coherence+Extend

  Thank you
  Patrick

• Connection failed SQL State 08001 SQL Server Error 10061 SQL Native Client TCP provider no. could be made because the target machine actively refused it

  Wandering SQL Native Client Datasource - apparently, it's a pretty common problem and I see several resolutions to it.  This problem started for our long company until I started working here.  We do not allow remote connections.  JDE Enterprise Server and SQL server are on the same subnet.  Users are on a different subnet.  SQL Server gets this message tries to connect to the Server Enterprise JDE:

  Connection failed: State 08001 ' SQL': SQL Server Error 10061: Microsoft SQL Native Client TCP provider: no connection could be made because the target machine actively refused it.

  I tried several suggested solutions and do not get the case. Any help is appreciated. Thank you, Sophie

  Hello

  The issue of Windows XP, you have posted is better suited for the IT Pro TechNet public. Please ask your question in the SQL Serversupport for assistance.

  Hope the helps of information.

  Concerning
  Joel S
  Microsoft Answers Support Engineer
  Visit our Microsoft answers feedback Forum and let us know what you think.

• Start a new Thread in the activate() method

  Hello

  When my application is started or resumed after a phone call etc... I would like to perform some tasks in the activate() method. First of all, I would like to start a new Thread to call web services a bit, and when it's done, I would like to update the user interface.

  I tried to do. I created a new class that extends the Thread in the run() method, I placed all the relevant web service calling and code in the method activate() my main thread that I have done:

  MyThread myThread = new MyThread();

  myThread.start ();

  But I noticed that nothing seems to be done. So I simplified my code by removing all web service of the run() method calls and replaced by a wait method (500), followed by a method call to update the user interface (in this case it creates just a new ButtonField, just for testing purposes).

  But ButtonField never gets created.

  The new thread doesn't seem to be from all.

  Any suggestions?

  Thank you very much.

  Have you tried setting stop and monitoring of execution points using println?

  First of all, is called same activate()?

  Secondly, run() is reached?

  When the thread does the work of the user interface, if you have that in a synchronized block try block invokeLater rather.

• Hi, I bought a new laptop and I would like to install Photoshop 5.5 Extended for the new laptop computer. Something to consider before you begin to install?

  Hi, I bought a new laptop and I would like to install Photoshop 5.5 Extended for the new laptop computer. Something to consider before you begin to install?

  The reason is to avoid any inconvenience. What do I need to uninstall Photoshop beforehand?

  Thanks for the tips

  The mobile fact have separate start-up and data readers?  If so and the boot drive has limited space, you may want to consider relocation of things like bridges on the data disk cache.  I can't think another thing to consider from the start, however.

• Installation of lightroom 5, later inst CC lightroom while lightroom free trial isn't over yet, I can't start this new (8 days remain :-(). Can someone help here?

  Installation of lightroom 5, later inst CC lightroom while lightroom free trial isn't over yet, I can't start this new (8 days remain :-(). Can someone help here?

  again, nothing you can do on this computer to extend your trial period except reformat your hard drive, software Adobe to trial has expired at the beginning

• Need help: vSphere 4.1 old cluster to a migration 5.1 new cluster

  Need help!

  Well, we need to replace the old vmware 4.1 (3 former hosts in cluster) with the most recent output 5.1 (3 new hosts in cluster).

  We have 3 DAS old_hosts connected to the SAN HP P2000 G3 FC (we have only a single SAN who has 2 active FC ctrl with FC port 2 each - if only one FC port is free).

  As the new material is very different (1 Xeon E5335 2 GHz for every former host) and 2 Xeon E5 - 2630 2.30 GHz for each new host I'm not so sure be allowed to add new_host to the old cluster, not you?

  If this is the case, that is the best strategy to run a migration as possible interruption less?

  Anyone can suggest me what logical steps, we need to achieve our goal?

  (b) create new cluster with EVC mode enable (based on the old family)

  so, I don't do any CVS operation on the old cluster?    Y/N

  Nothing changes in the old cluster and other Word is YES! operation of EVC.

  (d) to migrate all your virtual machines in esx old new host esx (with sufficient resources)

  When I see in my inventory vcenter the old and the new cluster in order to migrate a virtual machine between them that I have to drag and drop the virtual machine on the label of the new cluster name? It is enough, or there is another way to do this?

  Yes, you can drag and drop again cluster or you may use the Migration Wizard to migrate the virtual machines. Both way can be works.

  When I have all the virtual machines in the new cluster, the situation is:

  -cluster hosts are running in VCA mode

  -VMs are run in EVC mode

  so if I now turn off the CVS mode on the cluster hosts automatically will work with all the features of the CPU or they should be stop/start?

  VCA mode cannot be disabled for virtual machines operating in EVC mode. If you want to disable, you must stop your virtual machines before first VCA mode off.

  FYI, the VCA mode is for mixed with different hosts of the family of processors, but in your case, the VCA mode is you help to minimize downtime and the planning of your downtime for virtual machines. If you don't want to use the old family processor, then you must turn off your VM and migrate to new hosts. So in this case, you have no need to allow the VCA. This CVS is you help migrate live Webcam virtual machine to the new family of processor host without downtime. Because virtual machines live runs on the cpuID of processor mode the lowest.

  If you want to disable the VCA mode on the cluster, you need to do that there is no VMs is under tension. Because to the VCA ID CPU mode will change.

  Just my cents

• To a new cluster of vcenter 4.1

  I have two vmware 4.1 with separate storage clusters.  The two groups have no direct access to the storage of each other.  I am all of my VM migration to the new cluster.  The latest VM on my list is vcenter himself.

  I use vswitches distributed on my new cluster.

  I'm trying to think of a simple way of migration and drop my vcenter server from one cluster to another without falling on many problems.  I can't stop it and do a server/storage vmotion because I need vcenter to vmotion.

  One idea I had was to use vmware converter to move it to the new cluster.  It power off on the old cluster and on the new power.  Everyone thinks it would work ok?

  Billy

  If you have a mount LUNS or NFS temporary swing in clusters (or at least a host of each), you can do the following:

  1 put the temp data store online in both groups

  2 storage vMotion your vCenter for storage to the swing server

  3 vSphere client as root on the host where the vCenter

  4. free off the vCenter Server

  5 unregister it / remove inventory (don't delete disk of course)

  6. client vSphere as the root of the host target in the new cluster who sees the same storage

  7 browse the approriate time shared data store, right-click the .vmx and save it

  8. Select the port vDS group appropriate to change the settings, then turn on the computer, virtual

  8. give him the time to get online and update of the inventory of the hosts

  9. open client vSphere vCenter Server

  10. notes that the summary page of the vCenter VM specifies the correct host in the new cluster

  11. when the content, svMotion in the data store permanent desired

  As always, make sure you have good backups, or a clone of standby, you can switch to if you experience complications (unlikely).  I would also make sure the target VLAN has been tested on the new cluster.  If you have problems with vCenter get the update of the inventory (unlikely), you can restart the agents of management on the source and target hosts.

• New cluster or new pool of resources for tier 1 applications?

  Hello

  I have an existing cluster 5 node blade Esxi HP that I put in place for resource pools containing loads of pre-production and production work.  Now, we want to welcome applications from level 1 (Oracle DBs, Weblogic and middleware VMs servers).  Is there a best practice in the establishment of tier 1 workloads?  Should I add additional nodes to the cluster (until 8) and create a separate pool of resources for tier 1 applications, or should I implement a cluster 3 separate node (and generation) devoted entirely to the only level 1 workloads?  What are the advantages/disadvantages of each method?

  Thanks for your comments.

  M

  It depends on.

  (1) that you're going to develop this group even more tomorrow, or after you have added the hosts you have in mind who will be the end of the work?

  (2) how the negative risk is your boss or your client?

  (3) do you run your environment oversubscribed?

  (4) you understand the fundamentals of Resource Pools?

  Since I'm pretty risk adverse, I am tempted to create a new cluster with 4 nodes, where each node spans at least 2 or 4 chassis. Resource pools can resolve the problem of conflict of resources, but adds an extra layer of complexity to an environment that is not always welcome, and can make the diagnosis of difficult performance.

  Since you mention that you run ESXi on blades, please see the following article. I found it very informative regarding the problems of redundant N + 1 with blade configurations;

  http://longwhiteclouds.com/2012/05/13/when-should-N1-not-equal-1-host-with-VMware-HA/

  Kind regards

  Paul

• Problem migrating Vcenter server from the old cluster to the new cluster?

  Hey guys,.

  Just came across this situation, and I don't know what to do. I pass my vcenter server older glomerulus with hosts of esx4.0, to a more recent cluster with hosts of esxi4.1. The material of host machines are different generations, and vmotion don't migrate vcenter because of the cpu sets between cluster mismatch. If I could turn off the virtual machine and migrate it would work, but since it's my vcenter Server I can't do.

  Should I restore it on the new cluster?

  Thanks for the suggestions/tips

  Kevin

  Connect directly to the ESX host using your client vsphere, vCenter vm shutdown, vCenter remove inventory, then go to the new host, browse the data store on the vCenter vm is, and add the vCenter Server inventory.

  This assumes that the same storage is presented to both hosts.  If this isn't the case, you will need to use winscp or something similar to copy the files from the virtual machine.

  -KjB

• Could not start the new virtual machine or VM remove a data store

  Hi guys,.

  I'm not new with ESXi, but my knowledge with vCenter and Clustering must be improved, so here's my problem.

  We have a Cluster of 2 ESXi 4.1 hosts on a San with 2 VMFS.

  My first problem is when I created a new virtual machine and try to start it, I got an error message similar to the one included with this post.

  My Second problem is more picky. Since I am unable to start my new virtual machine, I had removed from drive and I had created a new. But when I took a peek in the data store where was deleted VM, the directory with all the files are still there. When I try to delete it, I got an error message telling me that it is impossible to proceed.

  Q please?

  Best regards

  Stéphane

  82% of failure can be other issues.  The article below

  http://KB.VMware.com/kb/1006052

  Also, I saw, where bad records DB for your ESXi hosts can cause weird problems as well.  The host with the problematic guest, can you do a right click on it and "disconnect"?  Try to get out of the bunch at the data center level.  If you get an error vCenter, that means return folder DB.  From there, while the host is in a disconnected state, remove the host to vCenter, then "add host". This will update the DB record.  Maybe worth it?

• new cluster resource

  Hello
  
  I created a new cluster to stop/start agent resource. I realized that when I stop agent by using the command "emctl stop agent", agent is restarted automatically in 3 seconds. What is the reason for this?
  
  
  
  Here is the command I used to create a resource:
  

  crsctl add resource agent -type local_resource -attr " ACTION_SCRIPT=/oracle/ora11202/grid/crs/script/agent.sh,  
  DESCRIPTION=Agent, DEGREE=1, ENABLED=1, AUTO_START=restore, START_TIMEOUT=120, UPTIME_THRESHOLD=1h,  
  CHECK_INTERVAL=60, STOP_TIMEOUT=120, SCRIPT_TIMEOUT=120, RESTART_ATTEMPTS=3, OFFLINE_CHECK_INTERVAL=60, START_DEPENDENCIES=, STOP_DEPENDENCIES="  

  Here's the script:
  

  vi agent.sh
  
  #!/bin/sh
  export ORACLE_HOME=/oracle/agent11g
  case $1 in
  'start')
     /oracle/agent11g/bin/emctl start agent
     RET=$?
  ;;
  'stop')
     /oracle/agent11g/bin/emctl stop agent
     RET=$?
  ;;
  'check')
     /oracle/agent11g/bin/emctl status agent
     RET=$?
  ;;
  esac
  exit $RET

  SQL * more is now integrated with crs now. If you stop the database with SQL * Plus, crs does not interpret this as a failure of the database and does not attempt to restart the database. But if the instance falls down for a reason, the crs will try to show the instance automatically. For example, try to kill the pmon/smon process for this instance and see what happens.

  In the case of the resource of your agent, when you stop the agent manually the crs interprets this as a defective component and try that restarts automatically. If you stop the agent via the crsctl command, then crs considers this as a defective component and he won't try to make appear the resource.

• Why is - FCP so complex: How can I simply leave a project and start a new, a new destination?

  Why is not more intuitive FCP to create the new project and close one on the screen, and then start a new and let me choose where I want to create it?

  I need a 3rd party app to import video content miniDV external HARD disk. Stuff is there for you.

  Now, I open the PCF and try to start a new project. However, this old project still there let me possibility to do it in the same folder: of course, being a tidy person, I don't want.

  I want to create another folder on the external HARD disk.

  What I am doing wrong?

  Hello.

  Keep in mind that a project must reside in an event and an event must reside in a library.

  You can have as many libraries as you want a store them wherever you like (even if you don't save them on a volume of time Machine.

  So start a new project in a new library or in a new event in an existing library to keep things orderly.

  Not sure I understand the part you need a third-party for mini DV application. You shouldn't have anything other than the FCC.

  Russ

• How to start the new film w/o bundle?

  Hello

  I'm on a MacBook Pro. I noticed things may work more slowly Gets the biggest of my library. What I want to do, is to reduce what must read my ram.

  I would like to start a new movie without a bundle. How is that done?

  Other films allows this folder as a foundation without having to use everything in the folder as fcpx use as a library package?

  For me, this is a big drawback for a fab app. Where you cannot choose what film project has in its database. I wish I could take stuff 'offline' when I want to and able to link again as needed.

  Not to mention, because the sequences are called projects, it screws up to talk about the 'project' film, so there are a lot of ridiculous workarounds. Irritating.

  The best

  Elmer

  I think most of the users work with external supports. You can change where the library stores the media in the library properties.

• Hello. I installed the latest update of Firefox and when I start a new tab page displays not now my entry from google line.please help.tks

  Hello. I installed the latest update of Firefox and when I start a new tab page displays not now my google input line. I use Windows 10. Until the recent update (v4.1.01) each time I started a new tab the new page would show the google search entry line which is my default. Can help you to restore this feature? TKS

  Hi, 41 of Firefox, you must use the add-on tab override (browser.newtab.url replacement), to make changes to the page tab.

  Once installed, open the Add-ons (Ctrl + Shift + A) Manager > Extensions, and then look for the new tab override and click Options to set your preference.

  Notice, if you want only the new tab to be the same as your home page, you can use the Add on new tab homepage instead.

  You will then need to restart Firefox for all changes to take effect.

  If your question is resolved by this or another answer, please take a minute to let us know. Thank you.