Exception when using tnsnames AND ezconnect resolution methods with managed Oracle ODP 12 c rel.3

We are looking at upgrading to a web service to connect to Oracle using the unmanaged to the ODP ODP managed (12 c module 3).

We use tnsnames and have the following configuration in the sqlnet.ora file: NAMES. DIRECTORY_PATH = (TNSNAMES, EZConnect).

The resolver EZConnect is necessary because we use an OracleDependency object for database update notifications.

When moving to the managed driver, a SocketException exception is thrown when you try to open a new connection. Looking at the call stack, I can see that the managed Oracle driver, actually a 'OracleInternal.Network.EZConnect' object, calls System.Net.Dns.GetAddrInfo to get an IP address, but with the alias tns as a parameter. Of course, it fails with this exception "unknown host".

So it seems to me that there is a problem combining tnsnames, ezconnect manned ODP managed!

Someone else who has experienced this and found a solution, or it could be an error in the driver?

This problem seems to have been fixed in the version of 12 c 4.

Tags: Database

Similar Questions

  • My slider controls when using Excel and Publisher became too sensitive

    My slider controls when using Excel and Publisher have become too sensitive and more often, don't do what it is that I'm willing to do.  For example: while in Publisher if I click on a photo, that I've inserted, and try to move it, it will automatically open the Format Picture option.  She does it with a number of other features as well.  If I'm in Excel and I'm trying to highlight a phrase, it will begin to highlight the requested sentence, then the middle of it will stop after the sense that I'm going with my cursor and it will pass at the end of the sentence.  I guess you could say this kind of "jump" and the mouse became too sensitive with the controls.  It's really frustrating and it feels like it has a mind of its own.  At first, I thought maybe a low battery mouse, because she was the way she does when the battery is dying.  Changed the battery and no change.  I also tried to set my mouse speed and still nothing.

    Any ideas?

    Oh and btw this just started to happen during the first 4 to 6 months.

    Hi H.Morton,

    1. the problem occurs with excel and publisher?
    2. what operating system do you use?
    3. did you uninstall and reinstall the mouse drivers?
    4. have you tried with another mouse and check?
    5. do the mouse has a selection of sensitivity on the mouse itself?

    I suggest you to update the mouse drivers and see if that helps.
    Updated a hardware driver that is not working properly
    http://Windows.Microsoft.com/en-us/Windows7/update-a-driver-for-hardware-that-isn ' t-work correctly

  • Failback when using Etherchannel and FCoE

    I have an environment that I am able to dig in for the purposes of learning, but I can't it.  Since I can't make any changes to the environment and experience, I wanted to ask a few questions here.  I'm trying to figure out what that the failback option of virtual exchanges must be defined on when using FCoE and etherchannel.

    The environment consists of:

    • FCoE
    • 2 cards of CNA
    • EtherChannel
    • vPLEX
    • Cluster of metro
    • Nexus 5000


    Each host connects to a single vDS and each portgroup is configured to use:

    • Load balance policy: route based on the hash of the IP
    • Active uplinks: two 10 GbE CNAs
    • Eve & uplinks used: no
    • Physical switch is configured for etherchannel.


    The vSphere networking 5.5 guide says the following for failback on vDS trade option:


    Failback Select Yes or no to disable or enable failback. This option determines how a physical card is returned to active duty after recovering from a failure. If failback is set to Yes (the default), the adapter is returned to active duty as soon as their recovery, moving the daughtercard, which resumed its accommodation, as appropriate. If failback is set to no, an adapter failing has been idle even after the recovery until another active adapter fails, requiring a replacement.


    This evokes the assets & Eve links and since using etherchannel requires that all active links with no standby connection, this means that the restore option does not apply to etherchannel configurations?  Or applies to the way in which the load is re -distributed when a link comes back after having been conducted? 


    For example, that we have a portgroup with two uplinks (uplink1 & uplink2), load balancing policy, the hash value of the IP and physical switch ports are in an etherchannel configuration.  During the normal operation of the IP of the hash calculations (http://blogs.vmware.com/kb/2013/03/troubleshooting-network-teaming-problems-with-ip-hash.html) are made and the load is distributed across the two uplinks.  Let's say the physical switch that connects uplink2 in goes down.  The ESXi hosts connected to that vDS immediately come into uplink2 equation which means uplink1 would only result in the calculation of hash IP?  What is the impact of the automatic restore option when uplink2 is back?  If failback is set to no, uplink2 still wouldn't be in the equation?  If so, how you would get it back in the equation?  Or when the link is up is immediately shot down there regardless of the restore option?  It feels like I'm missing something here. 

    Also, I read that attributing no failback is recommended when using IP storage, because it can lead to port heartbeat that can cause performance issues with the storage.  But I also read that traffic to an etherchannel appears as a single MAC/connection address.  This recommendation apply in an etherchannel configuration?  Are port and MAC stir the same thing?  Or is beating port just at the moment where the link goes up and down and MAC beat is when the MAC's switch between two ports?

    Sorry if these are fundamental questions, but not being a networking guy and not being is not able to experiment makes it hard.  Thanks for any help.

    The Member State of the binding not arise unless the link is established on both sides.

    I agree with the statement portfast. All the links go to an ESXi server do not participate in spanning tree protocols. The goal is to transition to a State of transfer immediately on the link turning upward. This must be paid regardless of the use of an OFFSET.

    LACP will dynamically build the GAL and validate the configuration on both sides is correct. I prefer LACP (dynamic LAG) on any static sort of LAG, but this requires a switch on the side of vSphere distributed. If you have the VDS, using LACP. However, LACP does not solve the problem of covering tree (you still want to avoid using tree covering weight on the SHIFT going the ESXi host).

  • Names using TNS and SQLcl resolution problems


    Hello

    I checked, but I don't see any answer to that.

    I use version 4.1 of EA of SQLcl.  In the video, Jeff says that SQLcl works using TNS, and I understand

    It's all jdbc.  But I was able to resolve names of database using TNS.

    I'm on Windows, and I'm using JRE 1.8.  I tried to set the TNS_ADMIN variable directly, and I tried settting it

    on the java command-line option - Doracle.net.tns_admin =.  But neither makes no difference.

    Here are the lines that I put in the SQL. BAT file.

    set ORACLE_HOME = C:\ProgramData\Programs\Oracle\instantclient_11_2-x64
    Set TNS_ADMIN = C:\ProgramData\Oracle\Network\Admin
    Set JAVA_HOME=c:\programdata\programs\java\jre_1.8


    SET SQL_HOME = % ~ dp0...
    REM the VALUE DEBUG = - agentlib:jdwp = transport = dt_socket, server = y, suspend = o, address = 8000

    Internal use of REM simple X for awt in
    SET STD_ARGS = - Djava.awt.headless = true
    SET JARFILE=%SQL_HOME%\lib\oracle.sqldeveloper.sqlcl.jar
    REM CUSTOM_JDBC="-Xbootclasspath/a:$ORACLE_HOME%\jdbc\lib\ojdbc6.jar.
    SET MYCLASSPATH = % MYCLASSPATH % REM; C:\drivers\3rdparty\sqlserver\jtds-1.2.jar
    SET MYCLASSPATH = % MYCLASSPATH % REM; C:\drivers\mysql\mysql-connector-Java-5.1.24-bin.jar

    %JAVA_HOME%\bin\java.exe % CUSTOM_JDBC % JAVA_OPTS % STD_ARGS %-Doracle.net.tns_admin=%TNS_ADMIN% % DEBUG % - jar '% c %' %. *
    REM - jar Java % c % *.

    But when I connect using SQLcl, it won't resolve the name of the database:

    Error message = IO error: could not resolve the 'DB1' connection identifier

    Needless to say I have 4.1 Developer SQL on the same Windows machine, and that resolves very well. So there is nothing wrong with the

    tnsnames.ora himself.  Someone at - it ideas?

    concerning

    Paul Stuart

    Paul,

    have you tried to add the parameter - optl? It should show you more info.

    So if you use - OIC (and libraries right in your path) sqlcl uses thick/BRC instead of thin.

    Can you give the complete error message with the parameter - optl, please?

    Martin

  • getting exception when using JavaFX SDK/JRE bundled in Netbeans

    When I try to put my JavaFX SDK/JRE to use Java SE and run my application I get an exception ClassLoader (java.lang.NoClassDefFoundError: javafx/application/application).

    I still have to get my Netbeans to use properly the bundled JavaFX SDK/JRE that come with JRE/SDK to the Java SE. I am referring to Java for JavaFX platform properties (in a project-> properties-> library-> manage-> tab JavaFX platforms).

    Here's what I have my JavaFX platform settings the value:
    JavaFX SDK--> C:\Program Files (x86)\Java\jdk1.7.0_10
    JRE JavaFX--> C:\Program Files (x86)\Java\jdk1.7.0_10\jre

    I am currently using 7u10 7.2 Netbeans and Java (the problem also occurs with 7u7 and 7u6).

    For those who using Java7u6 and later, can you tell me what your platform setting manage is the JavaFX tab?

    Thank you
    Jose

    OK its fixed. I noticed that the word hellow JavaFX netbeans 7.2 contains the following pots as part of its compilation libraries, jfxrt, deploy, plugin and javaws. In my project of problem I just add them to my library of compilation and it worked.

    Thanks for the help Igor.
    Jose

  • make exception when using BUlk collect

    Hi all

    version of DB 10 g

    I've written a procedure as below
    create or replace procedure test_proc2 is
    
    type v_emp_rec is record 
    (empno number,
    ename varchar2 (200));
    
    type v_emp_tab is table of v_emp_rec index by binary_integer;
    
    v_tab v_emp_tab;
    
    begin
    
    select empno,ename
    into v_tab(0)
    from emp
    where empno = 7369;
    
    dbms_output.put_line (v_tab(0).ename);
    
    exception 
    when no_data_found then
    dbms_output.put_line ('I am in no data found exception');
    end;
    now, I tried to perform the procedure below
    begin
    test_proc2;
    end;
    the foregoing is works well, but I want to use the functionality of collection in bulk so I use as below
    create or replace procedure test_proc2 is
    
    type v_emp_rec is record 
    (empno number,
    ename varchar2 (200));
    
    type v_emp_tab is table of v_emp_rec index by binary_integer;
    
    v_tab v_emp_tab;
    
    begin
    
    select empno,ename bulk collect
    into v_tab
    from emp
    where empno = 7369;
    
    dbms_output.put_line (v_tab(0).ename);
    
    exception 
    when no_data_found then
    dbms_output.put_line ('I am in no data found exception');
    end;
    the foregoing gives me exception, please someone can explain why it is not in the exception block.

    Thanks in advance.

    The obvious answer is because an exception no_data_found is raised.
    The question is there.

    It's either select it because there is no data where empno = 7369.

    Or is the .ename of v_tab (0), because there is no element 0. Automatic collection of indexing starts at 1.

    for example

    SQL>  declare
      2    type t1 is table of pls_integer index by pls_integer;
      3    v1 t1;
      4   begin
      5    v1(1) := 1;
      6    dbms_output.put_line(v1(0));
      7   end;
      8  /
     declare
    *
    ERROR at line 1:
    ORA-01403: no data found
    ORA-06512: at line 6
    
    SQL> 
    

    If you catch the exceptions and do not spread them properly, then hide you all information kind of useful as a line number, etc..

  • Strange behavior when using units and controls the ring

    Hello

    Something unexpected occurs when you use a control of the ring with units:

    1. Create the control of the ring,
    2. change the representation of double
    3. display unit label,
    4. the value of mV as a unit,
    5. edit items
      1. disable the sequential values,
      2. Add an element with a value other than zero for example 123,
      3. Ok
    6. Select the value created,
    7. create a KPI DBL with V as unit,
    8. wire from the control to the indicator,
    9. run

    What I see in the indicator is 123 V instead of the expected 0,123 V

    You guys are able to reproduce this?

    Noticed on LabVIEW 2013 and 2014, 32 and 64 bit on Windows 7 and 8.1

    Thank you

    Boris

    Bring up the digital display of the command of the ring.  You will see the units you enter under edit items menu is indeed volts.  Any prefix to units only acts on the screen.

  • exceptions when using dynamic instructions

    Friends

    In my monthly script, I need to drop 24 months old partitions and create the new partition for the coming months. I wrote the following procedure for the same thing:
    DECLARE
       v_msg_type    VARCHAR2 (100);
       v_msg_desc    VARCHAR2 (100);
       v_sql1        VARCHAR2 (1000);
       v_sql2        VARCHAR2 (1000);
       except_acct   EXCEPTION;
       except_vntg   EXCEPTION;
    BEGIN
    
    /* Remove 24 months old partition */ 
       SELECT    'alter table table1 drop partition p_acct_'
              || TO_CHAR (ADD_MONTHS (SYSDATE, -24), 'mmyyyy')
              || ' update global indexes'
         INTO v_sql1
         FROM DUAL;
    
       EXECUTE IMMEDIATE v_sql1;
    
       v_msg_desc := 'SUCCESS: Old partitions removed from table1';
    
        package.procedure('information',v_msg_desc);
    
       /* Add new partition */
       SELECT    'alter table table1 add partition p_acct_'
              || TO_CHAR (ADD_MONTHS (SYSDATE, 1), 'mmyyyy')
              || CHR (10)
              || ' values less than (to_date('''
              || TO_CHAR (LAST_DAY (ADD_MONTHS (SYSDATE, 1)) + 1, 'yyyy-mm-dd')
              || ''' ,'''
              || 'YYYY-MM-DD'
              || ''')) '
         INTO v_sql2
         FROM DUAL;
    
       EXECUTE IMMEDIATE v_sql2;
    
       v_msg_desc := 'SUCCESS: New parition added in table1';
        package.procedure('information',v_msg_desc);
    EXCEPTION
       WHEN OTHERS
       THEN
          DBMS_OUTPUT.put_line (SQLERRM);
    END;
    In this script, how can I add exception given for each EXECUTE IMMEDIATE statement?

    You wrap them with begin / end

    BEGIN
       EXECUTE IMMEDIATE --stuff and or things
    EXCEPTION WHEN --something
    THEN
       --more stuff
    END;
    
  • Failure of the assertion when using Labview ActiveX functions to communicate with the dll

    Hello

    I'm looking for suggestions and tips about the following error that I get when I use .NET Labview functions to communicate with the dll written in .NET environment. I just need a starting pointer on where can I search for the origin of this error. I am a newbie in. NET.

    I am also attaching the .vi I use, if this is of any help. All suggestions are welcome!  Thank you.

    The origin of the error found after studying the examples provided by the OR .NET. The callback function had to be moved out of the loop, which solved the problem.

  • Possible to use Lan and Wifi laptop computer with VMWare?

    I have a laptop Dell D630 running Windows 7 as the host operating system.  Normally if you use wifi and you plug a cable lan wifi disconnects because the laptop uses the connection with higher bandwidth.

    Is it possible to have the laptop direct is the entrance the guest operating system on lan and wifi running for the host operating system?  In this scenario, the guest operating system could have total control on the lan input and it would not need to be accessible by the host operating system.

    Thank you!

    I have no problem to use Ethernet and WiFi adapters at the same time on any of my notebooks including Dell.

    If you want to isolate virtual machines to be filled to the Ethernet adapter on the host computer and do not have the host use simultaneously then proceed as follows:

    In the value the target adapter VMnet0 virtual network Editor.

    Then on the dialog box properties of the adapter to the target on the host computer uncheck everything except VMware Bridge Protocol.

    Now with the guest on Bridged they connect to the physical LAN while the host will be able to use the WiFi adapter to communicate with the physical LAN.

  • Use: OLD and: NEWS in the triggers without oracle for each line

    I use Oracle 10 g.
    I want to insert a row in the table of the newspaper by each Witch of the query is executed on a table.
    And I want to insert on the line at a table of detailed log for each row changed in a query.
    I want to add modification date and time for each line in the two tables of the newspaper and I want it to be similar.

    So, I'm curious about using old and new variables without using for each row on the creation of the trigger. Is this possible?
    Or maybe I can score some variables static for this query (so sysdate will be the same and some of the Pavilion for the insertion of small newspapers)?
    Or maybe is there other ways to do it?

    LeopoldStoch wrote:
    But if I use sysdate in each trigger and it will update something as my sysdate 100000 lines will be different for different lines but I want it to be time and on the same date.

    Hi Leopold,

    Then today is your lucky day :-): sysdate is the same for all the DML. Even if it takes more than one second:

    SQL> create table rob
      2  ( id int
      3  , creation_date date
      4  )
      5  /
    
    Table created.
    
    SQL> create trigger t
      2  before insert on rob
      3  for each row
      4  begin
      5          :new.creation_date := sysdate;
      6  end;
      7  /
    
    Trigger created.
    
    SQL> set timing on
    SQL> insert into rob (id)
      2   select level
      3     from dual
      4  connect by level <= 10000
      5  /
    
    10000 rows created.
    
    Elapsed: 00:00:00.37
    SQL> set timing off
    SQL> select min(creation_date)
      2       , max(creation_date)
      3    from rob
      4  /
    
    MIN(CREATION_DATE)  MAX(CREATION_DATE)
    ------------------- -------------------
    07-05-2010 09:29:24 07-05-2010 09:29:24
    
    1 row selected.
    

    You can read more here: http://rwijk.blogspot.com/2008/07/sysdate.html

    Kind regards
    Rob.

  • Exception when using setWhereClause()

    Hello

    I use JDeveloper 11.1.2.3

    I need to get a row in the Dattabse table.

    To do this, I put the WHERE clause in the VO using the setWhereClause (String) method.

    Where clause looks almost like

    setWhereClause("Companyid = '"+getCompanyId()+"' and CustomerGroupName like 'Elite' ");
    

    while I try to run the application, his throw and error as:

    oracle.jbo.SQLStmtException: Houston-29114 ADFContext is not configured to process this exception messages. Use the code of exception stack trace and error to investigate the root cause of this exception. Root cause error code is Houston-27122. Are parameters of error message {0 = SELECT * FROM (SELECT * FROM OFFERRULE) QRSLT WHERE (Companyid = "01616C8F-4430-4F11-9257-B23626C50E99") and CustomerGroupName as 'élite'}

    Why is this happening?

    Kind regards

    Nigel.

    Nigel,

    the error you got is only the analytical error. There is an error detailed in the journal a place that tells you exactly what the problem is (for example table or view does not exist). The summary shows you the statement sql has a problem, so an action you should do is copy the statement and run it from a sql inside jdev worksheet. See is the instruction

    SELECT * FROM (SELECT * FROM OFFERRULE) QRSLT WHERE (Companyid = "01616C8F-4430-4F11-9257-B23626C50E99") and CustomerGroupName as 'élite'

    runs without problem.

    Next step is that you must not build your own where clause. To do this, you must use a ViewCriteria and apply the criteria of the vo, when you need to get the data.

    Timo

  • The behavior of copy (.png files) screen when using TextEdit and printing

    I have long used the method to copy and paste texts and/or pictures to TextEdit and print files .rtf or .rtfd below:

    I use the command-control-shift-4 and the cursor to select things and paste into TextEdit.

    That stopped working in the Sierra.  I tried several methods (PDF format conversion, fall in Pages and Numbers, for example), but nothing worked: the Canon PIXMA MX922 printed on a white sheet and left the box of small characters with text "filter impossible."

    There is no new printer driver on the Canon site; However, there is a new driver IJ Scan Utility and I downloaded it already after that installation of Sierra and the scanner works fine, as also directly print something (e.g. on the Web).

    Now, after several days of testing and I was wondering, I tried this yesterday:

    Instead of the above command, I used (which I do when I want to save it to the desktop) command + shift + 4.  I copied the file .png from office, dropped in the TextEdit, printed, and... IT WORKED.

    Copy and leaving fall directly in the TextEdit and printing, is so not working, but let fall the same screenshot on the desktop and copy and paste into TextEdit and it print shops.

    I noticed several other annoying problems to Sierra, some I've reported here (communities), some didn't.  Business-as-usual: first version of an upgrade is full of bugs, which 'disappear' in future versions, until a new update is made available, the loop continues, continues, and...

    He works here.

  • Application crashes with the Exception when using javax.microedition.media.Player

    Hello

    I have an application that processes data from bluetooth and send it to the web service. Recently, there was a request to add sounds to the request. Now when the application processes lots of data and the player plays constantly after a few seconds I get exception "Application is not responding. And then the treatment is finished. In the newspapers, I see many connected after this exception ForcedStackTrace exception.

    The sounds are played in the separate thread. If the app is not play sounds or short sounds - everything works fine. Is it possible to avoid what happens an exception? Why this is happening?

    Thanks in advance,

    Viktor.

    Solved by implementing my own PlayerManager, which running in a separate thread would point the way to the queue rather then the number of threads by using the internal player application.

  • Question about TNSNAMES and EZCONNECT

    I have 10.2.0.4 EE, RAC + Data Guard. From an application server, I run the following:

    + $tnsping base1 +.

    AMT Ping Utility for Solaris: release 10.2.0.4.0 - Production on January 14, 2013 12:37:36

    Copyright (c) 1997, 2007, Oracle.  All rights reserved.

    Use settings files:
    Oracle/product/10.2.0.4/network/admin/SQLNET.ora

    TNSNAMES adapter used to resolve the alias
    Try to contact (DESCRIPTION = (ADDRESS = (PROTOCOL = TCP) (HOST = 172.19.246.142)(PORT = 1521)) (ADDRESS = (PROTOCOL = TCP) (HOST = 172.19.246.143)(PORT = 1521)) (LOAD_BALANCE = no) (CONNECT_DATA = (SERVER = DEDICATED) (SERVICE_NAME = DATABASE1)))
    OK (0 msec)


    He said that he had used the tnsnames, so I went to take a look:

    + $cd /oracle/product/10.2.0.4/network/admin/+

    + $ls +.
    samples shrept.lst sqlnet.log sqlnet.ora


    There is no tnsnames.ora. Oracle lied to me? :) It's the sqlnet.ora:

    + $cat + sqlnet.ora
    + sqlnet.ora # Network Configuration file: /oracle/product/10.2.0.4/network/admin/sqlnet.ora+
    + # Generated by Oracle configuration tools. +

    NAMES. DIRECTORY_PATH = (TNSNAMES, EZCONNECT)


    Thanks for your help.

    If your client does not find $ORACLE_HOME/network/admin/tnsnames.ora it will search tnsnames.ora in your current directory and /var/opt/oracle/tnsnames.ora and then (on some operating systems). Culd that be happening?

Maybe you are looking for