VCS on multiple cluster

Similar Questions

• VCS EXPRESS WAY CLUSTER

  Hello

  We intend to deploy VCS Expressway Cluster for a company with two sites. Each site will have a VCS Expressway. Dual interface and NAT will be used. I would like to know the list of ports is used for cluster communications. Annex 3 of the Guide to creation and maintenance of Vcs Cluster located in the below mentioned website has details of the port. But this is not not in our Setup. It would be great if someone can provide me with the list of ports to be opened.

  http://www.Cisco.com/en/us/products/ps11337/products_installation_and_configuration_guides_list.html

  Krishna.

  Adding to Guy, note that all VCS must deploy within the period of round-trip 30ms for clustering.

  VCS cluster communicate among themselves and to replicate/share information between them.

• Multiple cluster type def not updated

  I combined group A and Group B with beam by name in Group C.  Group C is spent in a state machine in a shift register.   In some States, there are subvis with Group C as input.  All clusters are of type-def,

  When I change from Group A, I would expect Cluster C to update as well, but it became broken.  Why is this?

  
  

• You can edit vmx files on multiple virtual machines in a cluster to allow changes of setting delay start via RCLI or Toolkit VI?

  You can edit vmx files on multiple virtual machines in a cluster to allow changes of setting delay start via RCLI or Toolkit VI? How?

  SAs desire a lead time of 10 seconds through the virtual company composed of several VCs with multiple data centers and clusters.

  Known: we can edit the .vmx file and add: bios.bootDelay = "10000" this adds stunted ms 10000 (10 seconds) when starting

  Known: we may have to re-register or some other mechanism to apply the changes to the start on the VMS settings

  Has anyone tried to make a change of large company to bootDelay on VMS? Any help?

  The script below will do what you want. Have fun.

  -

• State of VCS cluster?

  can someone explain the status of vcs on a cluster configuration?

  My question is if the condition of the assets must be crossed between peers?, or there must be a single vcs active on the cluster?

  Please your help

  PS: version 7.1 of the vcs

  Peer 1 master:

  

  counterpart 2

  

  Thank you very much

  Hello Cristian,

  This looks ok. On the master, it will show the peer as active whice means that the peer is accessible... and even homologous way it shows control as active who once again even an indication that she is available.

  If you restart any server, and then refresh the page... Then it will show uknown or unreachanble or failed.

  imprtant part lies on the State of the cluster when you scroll down... It should show success and show you the next time both resync.

  Thank you

  Alok

• Replace the master VCS counterpart cluster (hardware failure)

  We need to replace a faulty Cisco VCS in a cluster. The peer was the master before failing.

  I if I understand cluster documentation we need to remove the VCs of the cluster and add it back in.

  My question is this will incur any interruption of on-line service on the cluster itself? Must the existing direct VCs or replacement be restarted etc.?

  the cluster is running X7.1

  Thank you

  David

  Hi David,

  No, the cluster will be always operational, just with a less VCS in its composition.

  Obviously, all calls that are active on the VCS that you delete will be affected, but the whole of the activities of the VCS will not (the other but many other you have in the cluster devices will be continuous without any required restart).

  Check out creation of Cluster VCS and service guide for complete instructions.

  Wayne
  --
  Remember the frequency responses and mark your question as answered as appropriate.

• 2 MCU and VCS Expressway, routing problem

  Hi all

  We have a design with a group of control VCS (2 members), cluster VCS Expressway (2 members), and a couple of microcontrollers (registred H.323 on VCS control cluster with the same prefix: 90).

  Each highway has a public IP address and incoming calls from outside can only be routed to microcontrollers:

  [email protected]/ * / for MCU_1 and [email protected]/ * / for MCU_2 (we don't have external DNS resolution).

  I put a conversion into motorway of VCS to change [email protected]/ * / to [email protected] / * / and [email protected]/ * / to [email protected] / * /.

  The problem is when someone calls [email protected]/ * / sometimes (randomly) the call is routed to MCU_2 (instead of MCU_1) and if the appellant see the auto attendant.

  The occcurs even then of the appeal [email protected]/ * / (MCU_2), sometimes the call is routed to MCU_1.

  Any idea what can cause this device or a way to make it work well?

  Thanks for help.

  José

  I think still that separate prefix would work, but here are a few ideas:

  Are incoming calls which do not possibly using SIP that is being interoperability H323 or incoming calls all certainly the H323?

  If you want to keep pure H323, you could perhaps just have a search rule/turn on your VCS-E who changed [email protected] to an E164 e.g. 90... and had a search on your VCS - C rule that says 90... stop at the local area.

  I also noticed that you direct calls to [number]@MCU-IP - have you tried to direct all calls to [number]@VCS-C IP instead?  If the VCS is the holder of a registration for a number, it should be able to deliver accordingly.

• VCS Virtual X 8 / Root access

  Hello world

  Just a quick question.

  Since the last update of x7.2 to x8.1, virtual VCS in a cluster is not accessible with root login.

  The strange thing is that a fast track (device), within the same infrastructure is accessible via the root.

  It does even more on the vmware console.

  Any suspicion or have you seen this problem yet?

  Bravo and thank you for your answers,

  See you soon,.

  Jul

  I can't really interpret your last message.

  You said 'connections series' where do you use it on a virtual serial ports of vcs - c?

  Do you have pwrec command to run and reset the password?

  Please see the guide to the administrator for more entries!

  The needs of VCS be restarted just to re - define the password password through pwrec, please follow the Administrator's guide.

  pwrec should be done via the virtual console in ESXi.

  (guess a virtual serial port might also work, but I've never tried and also why make it more complicated).

  Try something simple, like 8 characters containing A - X characters and 0-9 to rule out the possibility of keyboard problems.

  BTW, if you have the key to the strengthening of security, root access is disabled.

  Please remember useful frequency responses and identify useful or correct answers.

• TRAP on several CM / integrations CME

  Hello:

  Play with several cluster integrations. I understand THAT MWI is sent to all clusters. TRAP, works, however, based on the ports of TRAP activated from the bottom up. So, if I incorporate my box of MAN as the integration of third party, all connections of TRAP go first.

  He works most of the time because I have complete dialplan for connectivity between the systems, but he raised an interesting question:

  Is it possible to control which ports of TRAP are used depending on number of destination? As a plan of numbering within the unit for him say what ports to use?

  I suppose that, if not, then I guess I'll just disable the TRAP for all remote clusters and force it to send these calls to the local cluster and let it route like this...

  Suggestions?

  Thank you

  Vance

  No, there is no way to get the unit to use a particular set of ports active TRAP according to the phone number or the ID of the Subscriber or similar. Like a lot of granularity that you get is by switch - switch-based is smart enough to know which ports of TRAP to catch the user is associated with the database. But in the case of a multiple cluster as this configuration, there is that one ID switch in the unit so it just goes to catch the next available trap port activated when the request arrives.

• How to obtain a list of recipients

  Hi, I'm trying to get ot the list of voltage mV in my environment but I don't need VMS cluster like laboratory, vdi etc. How can I do this?

  my code

  Get - vm | WHERE-object {$_.} PowerState - eq "Receptor"} | measure-object

  The cmd above download the whole environment, including clusters of laboratories list.

  I think that where-object should do the trick, but I'm not able to add multiple cluster name in where-object. It will be nice to see that the cmd above can be executed with the where-object cmdlet

  Thank you

  Of course, the where clause used to filter the objects that are passed through the pipeline.

  In the block of code in the where clause, you provide a condition that should produce a result that is $true or $false.

  In the example above, the results of the GET-Cluster went through a where clause.

  The State uses the NotContains operator, which takes the left operan, which is an array of names in this case and check if the right operand does not appear in this table. In this case, the right operand is the name of the cluster.

  Do not forget that the variable $_ represents the object that comes from the pipeline and we deal specifically with the Name property of the object.

  So if the NOMCLUSTER is present in the table on the left, the object is not passed (we use operator NotContains).

  Then in the construction, we now run the Get - VM cmdlet against each cluster that passed the where clause.

  There is a much broader explanation of the where clause in Tobias free ebook.

• 500 + sessions is normal?

  The number of sessions specified in the default - .log out seems high, but I don't really have a frame of reference... during the day, he floats between 200 and 520 sessions. Is this good number? I want to just make sure that we're not close (or hit) ceilings.
  
  Here's a piece of the newspaper
  (ignore this busy/total value - most of these threads is blocked by an error that I work in another post. It is a low to medium traffic site. We rarely see more than 3 concurrent requests beyond peak hours)
  
  07/31 11:57:36 metric threads (busy/total) Web: 12/22 Sessions: 514 total memory = 144508 = free 74834
  07/31 11:57:41 metric threads (busy/total) Web: 12/22 Sessions: 513 total memory = 144508 = free 78448
  07/31 11:57:46 metric threads (busy/total) Web: 15/22 Sessions: 512 total memory = 144508 = free 84773
  07/31 11:57:51 metric threads (busy/total) Web: 15/22 Sessions: 512 total memory = 144508 = free 88089
  07/31 11:57:56 metric threads (busy/total) Web: 13/22 Sessions: 512 total memory = 144508 = free 75082
  07/31 11:58:01 metric threads (busy/total) Web: 12/22 Sessions: 514 total memory = 144508 = free 71981
  07/31 11:58:06 metric threads (busy/total) Web: 12/22 Sessions: 514 total memory = 144508 = free 75503
  

  The number of active sessions is not really a matter of concern. The number of requests per unit time is more important.

  Sessions will be by default the last 20 minutes AFTER the last page of the user request. Each user probably makes less than 10 or 20 applications page per minute (+/-depending on what type of site you have), probably less. Lets say that each user made 12 requests for page per minute * 500 users. It is 6000 pages requested per minute, which is equal to 100 pages requested per second.

  Assuming that the ColdFusion server has 1 CPU and is configured for 5 concurrent requests, and assuming that each page request takes 50ms to fill, then in 1 second, the server can complete approximately 100 applications page.

  In the example scenario, the server is located near its maximum flow rate based on its existing configuration since the inflow of 100 reqs/s is about the same as the rate of flow.

  With some trial and measurement, you can use a way of calculating the rate of speed for your server with your similar app config. You may find that you can also make some changes to the application like caching that could cut the page execution time in half and some other parameters like light cache trust to get an another boost of 10%.

  Once exhaust you the options, server performance if the load is still higher that flow, your next option would be to scale the application in multiple cluster servers each running your application on ColdFusion.

• VCS - C cluster cluster VCS-E

  Hello

  I have a VCS - C cluster and a bunch of VCS - E. Once VCS-C/VCS-E is in place and the second VCS-C/VCS-E is at a different location. What is recommended for the creation of area courses - what is multiple course areas or several address pairs in the same area of course.

  Thank you!

  George,

  an even multiple would be the right approach. but you have to consider opening the ports between highways and control for both servers.

  concerning

  Alok

• Jitter of RT! Can multiple readings to a variable / cluster cause a deadlock condition?

  Howdy do.

  While gradually develop and test an application on a crio9068 (RT linux) I started to see 'finished later?' in my main timed loop indicator flicker. Start pulling my hair out trying to figure out how to prevent this. I did a 'max hold' vi and can see the longest execution time for each image.

  The application runs fine at about 75% of CPU with the Panel open load, and the majority of the iterations run at the time. Sometimes I'll have a 'pic' at run time, all four managers in the timed loop take significantly longer than normal to run and the indicator 'late' says.

  A few questions I've had accumulate while chasing this:

  -If I use local varables to pass data between the loops, but only write to the variable in one place, can I still cause a condition of blocking/jitter by competing readings of this memory space?

  -If I use a node FPGA read/write to pass data between the FPGA and the timed loop, I should expect this causes a problem? I selectively disabled a lot of my code, and it seems that is where some of the delay occurs. What me stem, it's that these parts of the code have not changed in the recent development and the thing never used to run to the end.

  -On the theme of the node FPGA read/write, I assumed before that I should not write to the same FPGA FP element in different subvis. However, the code is set up so that there are several parallel calls to the read/write node, just with different selected items. Is that BAD?

  -Similarly, if I ungroup and read the same item in a cluster in a "parallel" mode control, can this cause a deadlock situation, or is it the same as unbundling and wiring away in several places?

  -I am thanks to the recently renamed "software calibration and management toolkit NI (SCM)," formerly Drivven CalView, to manage the communication between the RT and a windows host. He is also careful fault management stuff. Anyone else using it, and is it possible that I am getting jitter by having too many calpoints in my deterministic loop?

  No indication on any of the above would be greatly appreciated. If I don't make sense on any of the above points I do excerpts from example to describe.

  Hi Matt,

  What about under the name of multithreading - the same thread can run on multiple cores, but everything in a timed loop will always hold in a single thread.  For reference (see 'Set the execution order for screws of real-time synchronization'):

  LabVIEW Help: Synchronization of deterministic Applications (real time Module)

  http://zone.NI.com/reference/en-XX/help/370622L-01/lvrtconcepts/timing_control_loops/

  To cache the runtime data, I didn't have a specific method for mind, rather I generally try to catch at least one instance of the hiccups by capturing variables of interest in the shift registers or a FGV, just to see if there is any kind of difference.  This may not be useful in your case, as I suspect it's interactive execution, what causes the problem.

  Regarding the interactive performance and determinism, there is really no way of knowing what will happen, but it will often cause the kind of behavior that you see now. Because you are more or less continuous between the target and the host data, everything hangs in the network or on the computer host can set the RT system, especially if you read or write data on the front panel. Given that all of your curls are delayed at the same time, it seems the most likely scenario, unless you have a VI reentrant or similar shared between all the loops - I have seen that a few times with managers error or stop sharing.  Given that these spots are quite often (it resembles something like 1/100 iterations, I would recommend put in some shared variable RT FIFO benchmarking and the deployment of the application as a stand-alone RTEXE.)  It should be fairly easy at that time see if instability persists.

  Best regards

• VCS CLUSTER PROBLEM

  Need an idea to fix a few alarms on my cluster configuration two VCS.

  VCS are running X8.5.1

  Had five alarms on the my resume

  1. the replication is no longer available.  Systems have 4ms between them.

  2 lost all Admin user accounts less?   Strange... Newspapers reported that no one was in the VCS

  3. admin VCS Master on the master account was returned to the TANDBERG default?   (Alarm has stated that this was the case).

  Changed the admin password and recreated my admin on the master account.

  4 approves. it 2 (slave) lost all VCS user account, but the admin password reset not and does not have that I have listed.

  My partner has been able to connect to the peer 2 VCS, even if his account is not listed in the list of users?

  ------------------------------------------------------------------------------

  After the captain restart now, I have three alarms

  a. unexpected software error has been detected in ClusterDB: unknown reason on the slave?

  b. There is a port conflict between the media (36000-47999) port range and the range of ports UDP ephermeral (40000-49999) on Master

  c. Core dump mode has been changed, but a reboot is required before this takes effect on Master

  1. any ideas what could have caused the user accounts to disappear on the page?  I think that they are still in the VCS configured just cannot see them.

  2 will be restarting on the master clear alarms b and c?

  3. a reboot on the peer 2 (slave) would open the ClusterDB alarm?

  I can't restart until later tonight that calls are up on both.

  Oh finally where one or even any changes a core dump mode?

  Thank you

  Connect to the VCS via Telnet/SSH/series and log in as 'root '.

  use the command 'selectsw' to see what partition software he's using now, then, if image1 is active, allows the command "selectsw image2" switch to the other, or vice versa, and then restart your computer.

  Wayne
  --
  Remember the frequency responses and mark your question as answered as appropriate.

• VCS Cluster configuration

  I'll put up a cluster between VCS - 02c in version 8.x and one of them will be the captain. I will put all endpoints in the VCS - C Master points, what would happen if the master of VCS - C problems? My end points would be automatically transferred to another VCS - C (backup)?

  Hello

  Setup is a cluster of VCS VCS a set up a Master and others as a slave.

  Your points of termination/Codec must register to the VCS cluster (you need to update your DNS with VCS cluster name i.e. clustervcs.domain.com with the ip addresses of two VCS.  You can use SRV records to do a primary and the other back).  If your CODEC does not support the SRV record then you can put the IP of master VCS and the registration, the VCS Master will provide the idec with its ip address and ip address of the slave VCS.

  A good link:

  http://www.netcraftsmen.NET/blogs/entry/Cisco-VCs-clustering-configuration.html

  I hope this helps.

  Kind regards

  Ahmed