Controlling which node OWB connects to in a RAC environment

We have OWB (10g R2) deployed on a two node RAC (10gR2) database (DB name ZIAP, instances = ZIAP1 ZIAP2).
We would like to have client session and OWB sessions connect to instance ZIP1 consistently when we connect to ZIAP.
We have changed the tnsnames.ora file to switch load balancing off and to connect to ZIAP1 1st).
This works for client sessions.
When OWB sessions run (i.e. when we execute OWB jobs) we see them start on ZIAP2.
The question is how does OWB decide which instance to connect to and how can we control this?
Thks
John

Hi John,
look this
Re: Registering RAC location results in ORA-29532
We have changed the tnsnames.ora file to switch load balancing offto switch off load balancing it not enough to modify tnsnames.ora on client side,
also you need to modify instances parameters (for more detail look at forum thread which I specified)
Regards,
Oleg

Similar Messages

Specifying which node a task runs on.

Hello all,
Since upgrading to IdM 8.1 Patch 11, I'm noticing that idm is running scheduled tasks or bulk actions on any server that it feels like in the cluster. In prior versions these tasks would run on either the host the bulk update was launched from, or from the host specified in the tasks schedule. How do I control which node a task executes on? We keep some nodes reserved for end user access, and some nodes reserved for back end processing tasks...
Thanks,
Jim Beard
University of Oregon

Jim, this is easily managed by the use of the environment-target.properties. Change the task definition to the following:
<Hosts>
<List>
<String>%%NODE_TO_RUN_TASK%%</String>
</List>
</Hosts>
and add the following to each of the environment-target.properties file:
%%NODE_TO_RUN_TASK%%=nodeToRunTask
where the nodeToRunTask is different depending on the environment.Note that this macro-replacement is a feature of the Java plugin for IDM development (for Netbeans or Eclipse). It is not part of IdM itself.

Controlling which devices Facetime rings

Hi Forum,
we now have 3 iOS and 4OSX devices in the house, under one Apple ID and 2 iCloud IDs. 2 iClouds IDs for me and my wife, one Apple Id for purchases across all the dvices and so we only pay once.
With this set up, how do i control what calls what when i call from one of my devices with my iCloud Id to my wifes device and her iCloud ID? i dont even think the iCloud id matters, i just threw it in for set up clarity...

Hi,
If you force users to use RD Gateway to connect then you could configure the RD Connection Authorization Policy (RD CAP) to only allow computers that are a member of certain groups. For non domain-joined devices there isn't really a built-in method
to control which devices can connect.
-TP

Is there any way I can control which specific access point I connect (and stay connected) to from amongst a set of access points with the same SSID?

I'm working from a boat in a harbor in which the ISP has deployed numerous access points around the periphery. All the access points share the same SSID and each is configured to use either channel 1, 6 or 11. From my location, there are over a dozen of these access points "visible" (based on the the output of WiFi Scanner) with a range of RSSI and S/N values that vary over time.
The ISP has told me that the quality of my connection should be "perfectly fine" for any access point with an RSSI value better than -75, but I know from experience that my connection quality is miserable (i.e. < 50Kbps download) for almost all of these, including those with RSSI values better than -75. There is at least one exception, however, which gives me on the order of 2Mbps download, which is "great" in this context.
I've tried using a more powerful USB antenna plugged into my MacBook Air (mid 2011), but as far as I can tell, it really doesn't make much difference. Neither does my location within the boat. The overriding factor seems to be which access point I happen to connect up to.
I should point out that the closest access points are about 75 yards away, with many of them being several hundred yards away or more. I'm guessing that even though the signal strength of some of the distant access points is causing them to get "chosen" some times, the results are unacceptable due to the distance.
I'm hoping that I can determine, through experimentation, which access point(s) provide(s) acceptable performance and then configure my Mac to limit my connection to those points through whatever mechanism I need to use (e.g. channel, MAC id, etc.).

Establishing a wireless connection with a client computer is left to the access point for various reasons. One reason that your Mac may not connect to the strongest access point is that it may have reached a limit of the number of clients it can serve, leaving it unable to accept a connection with another. The limit may not be very large.
Suppose that happens, and your Mac establishes a connection with a more distant access point having a weaker signal. Then, suppose a client drops off the network. Doesn't this mean your Mac will switch to the stronger access point? Not necessarily. The throughput delivered to and from your Mac would have to drop below a threshold specified in the AP for it to drop the client, leaving your Mac free to connect with another one. The reason for this is to prevent rapid switching from one AP to another in an area in which two signals are of approximately equal quality. If that were to occur the frequent and repetitive handshaking between the two devices would slow throughput to zero.
In an environment in which several access points are broadcasting the same SSID, Apple provides no insight as to how it determines which access point to choose. This is the reason I suspect this "choice" is a function of the router, or access point. The connection originates with it, not the Mac.
Now, what would solve your dilemma would be to determine a way to control the access point with which your Mac connects, by specifying the access point's unique MAC address for example. In this happy circumstance, you could maintain an editable "whitelist" or "blacklist" of the harbor's access points and be able to choose which among them you prefer.
I do not believe OS X maintains such a record of MAC addresses though, only those of the routers it uses. If I am correct about that, such a solution is unlikely to exist. Don't let that discourage you from searching for one though... I would concentrate on something like "selecting access point by specific MAC address".
I did find this patent application though:
Roaming Network Stations Using A Mac Address Identifier To Select New Access Point
Perhaps it's a start

HT4260 Can i control which wireless clients i like to betrol which wireless clients i like to be connected on my airport base station?

Can i control which wireless clients i like to betrol which wireless clients i like to be connected on my airport base station?

Yes. Use AirPort Utility's MAC Address Access Control.
AirPort Utility > select your base station > Edit button > Network tab > "Enable Access Control" > click the Timed Access Control button.
You need to specify the clients you want to control by their MAC address, which is usually printed on the device's exterior case.

Which user connected from which node in RAC

Hi All,
I have installed 2-node Oracle 10g R2 RAC on RHEL4. It is working fine. Now i want to know which user connected from which node, how can i know this details?
Thanks,
Praveen.

You can use the gv$ views which actually submit the query ran against them to all instances of Oracle for the RAC database.
The documentation for the gv$ views is the v$ view entry in the Reference manual.
As far as I know when you run RAC every v$ view is built on a gv$ view.
HTH -- Mark D Powell --

Controlling which WDS base station a machine connects to... how?

I'm trying to isolate a problem I've got with a simple WDS configuration. Configuration is:
Airport Extreme (n) Fast Ethernet - main WDS
Airport Extreme (Snow) - relay
Airport Express - remote
Due to configuration, machines that connect to the Express can also 'just see' the other two machines. One of them (a CoreDuo iMac) is reporting very poor Airport signal strength - even when the Airport Express is plugged in within 2 feet of the machine. We think that it is connecting to one of the other base stations - but can't work out how to find out if this is so, and if so how to stop it.
The iMac has recently returned from Applecare having had a new screen installed - is it possible that the airport antenna could have been disconnected as part of this? This might explain problem - but again I can't work out how to test this.
Suggestions anyone?

Airport Extreme (Snow) - relay
I know that you mean round AirPort Extreme base station (AEBS) but I want to clarify something.
There is not a snow AEBS. Referring to it that way is confusing because there IS an 802.11b Dual-Ethenret (aka "snow") AirPort base station (ABS). The snow ABS does not do WDS and does not have a USB port.
The iMac has recently returned from Applecare having had a new screen installed - is it possible that the airport antenna could have been disconnected as part of this?
Quite possibly.
There isn't any way to control which base station it connects to BUT you can see which one it is connected to. If the iMac is running OS 10.5, hold down the OPTION key while selecting the network name on the AirPort menu. You will see the MAC address of the base station which the iMac is connected to.

How to tell which node a session connects to in 9i RAC

I have a 6 nodes 9i RAC database. While running a process, how to tell which node it is connecting to?

Hi,
SQL> select inst_id, sid, username from gv$session ;
inst_id is the instance number a session is running on.
select from the v$instance view to know to instance number you are actually connect to
SQL> select instance_number from v$instance ;
INSTANCE_NUMBER
1
note the the "g" before the view "v$session", that's for GLOBAL (all instances of the cluster) : check the views gv$parameter, etc...
Fred

How do I access Firefox when Norton 360 firewall seems to be blocking access but all links on Norton 360 default to Mozilla page which can't connect?

I'm writing this via a Windows Explorer browser window which should tell you the problem, or at least that I can't connect through Mozilla Firefox.
Mozilla has been my default for years and I've always had Norton, 360 for the last few years. I've been through the traditional update problems with Toolbar/ID Safe and both of those are working fine.
Both were updated to absolute latest version within the last week due to an Update/Restart problem I was having with Windows 8.
There were some periodic problems with initializing the Norton problems which generally led to a Firefox not responding situation.
But I was able to run both quick and complete scans yesterdays and the Norton Autofix tells me no problems with their program/app.
I just downloaded newest Mozilla 25.0.1 last night again as last resort because I can't access Firefox beyond the homepage.
Spotify also told me there appeared to be a firewall blocking access. I also can't access iTunes.
I went to Windows 8 Action Center and it listed Norton 360 as controlling the firewall.
But I went to the allow apps through Windows Firewall tab in System anyway and didn't see any listing for Mozilla or Firefox there (maybe you have a less obvious filename).
My own network connection is working fine, obviously I'm on Explorer and no problems with my other computer.
I'm going to contact Norton to deal with this problem, too.
But my real problem now is that any attempt to contact the Symantec/Norton help/support site is automatically sent through a Firefox page which can't connect.
Right now I can't figure out how to do a tech chat or something to work at it from their end.
So that's why I'm here asking if you've run into this or know any way around it.
Obviously, the quicker a response the better.
Thanks for your time.
Ciao 4 now,
Don Snowden

Hello J Scher,
I don't know if this answer solves it, but I did contact Norton Support through IE and the tech had me download Firefox again...just like I did the night before on my own. But for the moment, for whatever reason, all those issues have gone away, knock on wood three million times. I can access Norton support pages, I can get on Spotify and I can reach the iTunes store.
I will freely admit to not being a great computer person but I have no idea why my machine did this to me all weekend and why it cleared up now, unless the Norton tech did some behind-the-scenes tweakery. But for now, I'm okay and things are working, knock on wood four million more times (figure I better up the ante).
Thanks for your help.
Don Snowden

Generation of Control file in OWB Mapping.

Hi All,
OWB config details is as follows:
OWB 9i :9.2.0.2.8
OWB Repository: 9.2.0.2.0
OWB 9i Client OS: Windows XP Environment
Oracle DB 9i Enterprise Edition 9.2.0.1.0-64 bit Production on IBM AIX machine.
I developed mapping which involved a source file (CSV) & oracle table as target.
Validated & Generated mapping. No error displayed.
Configured mapping set Data File Location, Bad File Location, Log File Location (All these parameters were set to point to different directories in unix.)
Registered all the above location in Deployment Manager.
Deployed the mapping & executed mapping.
I was expecting control file generated by mapping to create on unix partition, however many times it did save on my local machine on Windows XP rather than on unix. The control file created in the local directory where OWB client is installed.
When the same mapping is executed from another machine still the file was created on my local machine.
This was observed after I created file location for Bad file & Log file from OWB.
Why does control file is created on local machine rather than Unix Server?
Can someone please let me know more on this?
Thanks in Advance.
Regards,
Vidyanand

Hi Vidyanand
You may try to register physical file location under one logical location. Then register the location with physical datafile lication. It worked for us.
Cheers, Bana

I need a control which allows a user to choose from specific values (which I specify), NOT 0, 1, 2 etc. [I KNOW I can let him enter 0, 1,2 fro

m a listbox, and then change the value on the block diagram, but what I want is that on the subvi node itself he can enter the actual values, say 1, 10, 200. Does anyone know how to do this, creatively, since currently a control like this is not available.I need a control in Labview which allows a user to choose from specific values (which I specify), NOT 0, 1, 2 etc. [I KNOW I can let him enter 0, 1,2 from a listbox, and then change the value on the block diagram, but what I want is that on the subvi node itself he can enter the actual values, say 1, 10, 200. Does anyone know how to do this, creatively, since currently a con
trol like this is not available.

m a listbox, and then change the value on the block diagram, but what I want is that on the subvi node itself he can enter the actual values, say 1, 10, 200. Does anyone know how to do this, creatively, since currently a control like this is not available.> I need a control which allows a user to choose from specific values
> (which I specify), NOT 0, 1, 2 etc. [I KNOW I can let him enter 0, 1,2
> from a listbox, and then change the value on the block diagram, but
> what I want is that on the subvi node itself he can enter the actual
> values, say 1, 10, 200. Does anyone know how to do this, creatively,
> since currently a control like this is not available.
>
One way of doing this is to use either a ring or enum where the strings
presented to the user are "1", "10", or whatever you like. On the
diagram, the values will be 0, 1, 2, in a compact counting sequence.
Then do an array index into a constant. If this is done more than one
place, make the control a typedef so you can keep them consistent an
d
make the index be done in a subVI. Then whenever you need this, you
drop the control and the subVI, out comes the integer, float, or
whatever it is you want to index.
Greg McKaskle

How does RAC determine which instance number to run on which node

Good afternoon
I am trying to script a process and I need to identify how RAC decides which node runs which instance. Is it the first one of the pair added to the cluster that runs node 1 or is there some other piece of information that I need to work this out
This is on Redhat Linux with 10.2.0.3 for CRS, ASM and databases
Cheers
Peter

Before "you" start an instance you set env variable ORACLE_SID. This identifies an instance. When you go into sqlplus and issue STARTUP, Oracle starts the instance named by the sid. Thus the instance running on the server is controlled by you. This changes as noted below.
If you were using a non-Oracle tool to start instances, such as Veritas, then you would see it start the instance you coded into the tool. It would not randomly pick an instance. It looks in the Veritas config file and sees that you always want instance 1 on this node and instance 2 on that node.
That said, you can make Oracle more random or "grid" like. 10g RAC done Oracle's way likes to bounce around between primary and secondary nodes. To see which instances are running on a node you can "ps -ef | grep pmon". Alternatively, use sqlplus to look in the database: view gv$instance gives you each instance name paired with the name of the host it is currently running on. There is one line of output per instance currently running.
-Mark

How to control each node mx:tree

Hello all,
I use an mx:tree. The issue is that I wantto control each node for creating lazy load mode tree. I want to see node as an clickable node which has children. But I'll call and add the childrens to dprovider when I need it, on click itemOpen event.The mx:tree watches dataprovider so show us the node if has children or not.If a node hasn't children in dataprovider, it will seem like an item,not a bin with itemOpen event.
Can you suggest me to fix this problem ?
footnote: Each node Object has Type property if it's Bin or Video etc. I want eachone if object.Type==bin, show me a bin with itemOpen event.
It's called Disclosure Triangle, yeah, that I want to control this guy
Thanks in advance

DefaultDataDescriptor class is defined for mx:tree as default descriptor. You may manipulate it by creating your customDataDescriptor class and you can override isBranch , hasChildren and getChildren functions.
For more details read about DefaultDataDescriptor class from http://help.adobe.com/en_US/FlashPlatform/reference/actionscript/3/mx/controls/treeClasses /DefaultDataDescriptor.html .

OWB connection to teradata through ODBC driver

Hi:
I am seeking answers to my OWB connection settings. I have installed OWB 10.2.0.4 on Windows 2003 server. The Oracle 10g database is on the same server. Right now, I want to configure ODBC connection for OWB to source database (all on teradata) on my windows 2003 server. However, I found from control panel
/Adminstrative tool/Data source (ODBC), there is no teradata driver in ODBC administrator. (It is available on windows XP). How can I solve this problem?
OWB perform ETL workflow through OWB server and source database. The connection between OWB server and source database must be set up. If no teradata ODBC driver, is there any other way to work around. Or should I update windows 2003 server to patch teradata ODBC driver to it? Please give me your idea and your thoughts. Thanks in advance.

Hi,
look at this thread
SQLServer access from AIX Warehouse builder
What is the problem with installing Teradata ODBC driver to your Windows2003 server?
Regards,
Oleg

Tab Control Property Node gives a build error when used in a VI for LabVIEW PDA

Hello,
    Iam new to LabVIEW development & need help regarding this:
    We are trying to port a LabVIEW application from desktop to a Windows Mobile PDA. Please note that the LabVIEW application works fine on the desktop. In one of the VIs, a Tab control property node is used which gives a build error with LabVIEW PDA project. A sample VI which shows the problem & the error list snapshot is attached with this post. Can anyone please suggest an alternate approach to achieve the same functionality with LabVIEW PDA? Waiting for an early response from anyone...
Thanks & Regards,
Subhashini
Attachments:
VI_Error.zip ‏31 KB

Hello Ton,
    Am glad that i got a response, thanks. The following links indicate that Tab control is supported on PDA target:
http://digital.ni.com/public.nsf/allkb/CBA335641E27B858862571490015EA88?OpenDocument
http://forums.ni.com/ni/board/message?board.id=170&message.id=221507&requireLogin=False
    I feel that the Tab control property node that has been used in the VI (which is attached earlier) may not be supported on the PDA target. Can u (or any LabVIEW expert) suggest an alternate approach to achieve the same goal with LabVIEW PDA (changing colour of Tab control pages programatically)??
    I had missed out some details earlier: We are using LabVIEW 8.5 PDA module on a Windows Mobile PDA.
Thanks & Regards,
Subhashini

Controlling which node OWB connects to in a RAC environment

Similar Messages

Maybe you are looking for