Oracle bug?

Similar Questions

• Humungo Oracle bug. Verified by multiple sources

  http://StackOverflow.com/questions/2959130/Oracle-minus-query-how-can-i-get-a-result-with-NULLs-if-the-top-SQL-and-the-Bott
  
  If you are interested to see it reproduced while data lasts, answer...

  If you believe you have encountered a bug, why tell us?

  Log a TAR/SR with Oracle's Support and report. Maybe there is already a patch for it.

• Oracle patch in general

  I have the following questions regarding the oracle patch

  1. I learned from my s/n superior, saying that the application of the fix can be another bug? is this true? Can you share your experience, if any?

  2. is it good idea to apply any patches for a specific version (for example oracle 11 g r1) without consulting the Oracle support

  It is recommended?

  3. the question of whether the patch script will vary from instance ASM and no - ASM? or would be the same?

  4. how the Oracle bug id is different from patch set? In some cases, the Group of hotfixes is same as the id of the bug? is it possible to distinguish these two terms?

  5. is it possible to identify the list of patches and bugs corrected by querying the dictionary of oracle database?

  6. do make sense to apply a lot of patchsets in go - one rather than one. I mean apply a patch and wait 1-2 weeks and if we don't face any problem apply the second patch and so on.

  7. do we need to apply the patch (db clustered) instance instance or just the database?

  It would be great if you someone erases my queries listen on fix oracle.

  Thanks a lot for sharing the knowledge.

  NOT if there are scripts to fix POST. You will find in the correction of document when you apply a patch. Patching docs have step step of the instructions.

• Oracle 10g Clause WHERE properly executed.

  Enterprise Edition Release 10.2.0.4.0 - 64 bit seems to be a bug.
  
  If SQL "first_insert_date = to_date ('2010-09-07', 'YYYY-MM-DD')" paragraph, he must show 7 tuples. "." but it shows me tuples 0 sometimes.
  Please find below the log.
  Where is the site of Oracle bug report?
  
  
  SQL * more: Release 10.2.0.1.0 - Production Wed Sep 8 07:18:48 2010
  
  Copyright (c) 1982, 2005, Oracle. All rights reserved.
  
  
  Connected to:
  Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - Production 64-bit
  With partitioning, OLAP, Data Mining and Real Application Testing options
  
  SQL > delete from SCENARIO_SUSPICION_AUDIT_MGT where first_insert_date = to_date (' 2010-09-07', ' YYYY-")
  MM - DD') and scenario_seq in (select scenario_seq from the scenario_transaction where scenario_cd = ' MA00)
  011' and not scenario_ins_id1 (select max (scenario_ins_id1) in the scenario_transaction where scena)
  rio_cd = "MA00011" group of ss1));
  
  0 rows deleted.
  
  SQL > select count (*) in the SCENARIO_SUSPICION_AUDIT_MGT where scenario_seq in (select scenario_seq en
  OM scenario_transaction where scenario_cd = 'MA00011' and scenario_ins_id1 not in (select max (scenar
  io_ins_id1) from scenario_transaction where scenario_cd = "MA00011" group of ss1));
  
  COUNT (*)
  ----------
  20
  
  SQL > select first_insert_date, count (first_insert_date) in the SCENARIO_SUSPICION_AUDIT_MGT where scena
  rio_seq in (select scenario_seq from the scenario_transaction where scenario_cd = 'MA00011' and scenario
  insid1 not in (select max (scenario_ins_id1) in the scenario_transaction where scenario_cd = ' MA00011)
  "Group of ss1)) first_insert_date group;
  
  FIRST_INSERT COUNT (FIRST_INSERT_DATE)
  ------------ ------------------------
  AUG 30, 10 1
  JULY 12, 10 1
  JULY 5, 10 1
  AUG 2, 10 1
  MAY 31, 10 1
  AUGUST 16, 10 2
  AUGUST 16, 10 2
  05-SEPT 3.-10
  7-07.-10
  06-SEPT 1.-10
  
  10 selected lines.
  
  SQL > delete from SCENARIO_SUSPICION_AUDIT_MGT where first_insert_date = ' 07 - SEVEN.-10' and scenario_seq
  in (select scenario_seq from the scenario_transaction where scenario_cd = 'MA00011' and scenario_ins_id
  1 not in (select max (scenario_ins_id1) in the scenario_transaction where scenario_cd = "MA00011" group
  by ss1));
  
  0 rows deleted.
  
  SQL > select count (*) in the SCENARIO_SUSPICION_AUDIT_MGT where first_insert_date = to_date ('07-09-2010
  'yyyy-mm-dd') and scenario_seq in (select scenario_seq from the scenario_transaction where scenario_c)
  d = "MA00011" and not in scenario_ins_id1 (select max (scenario_ins_id1) in the wh scenario_transaction
  Scenario_cd = "MA00011" AGE group of ss1));
  
  COUNT (*)
  ----------
  0
  
  SQL > select count (*) in the SCENARIO_SUSPICION_AUDIT_MGT where first_insert_date > to_date ('2010-09-06
  'yyyy-mm-dd') and scenario_seq in (select scenario_seq from the scenario_transaction where scenario_c)
  d = "MA00011" and not in scenario_ins_id1 (select max (scenario_ins_id1) in the wh scenario_transaction
  Scenario_cd = "MA00011" AGE group of ss1));
  
  COUNT (*)
  ----------
  7
  
  SQL > delete from SCENARIO_SUSPICION_AUDIT_MGT where first_insert_date > to_date (' 2010-09-06', ' yyyy-")
  mm - dd') and scenario_seq in (select scenario_seq from the scenario_transaction where scenario_cd = ' MA00)
  011' and not scenario_ins_id1 (select max (scenario_ins_id1) in the scenario_transaction where scena)
  rio_cd = "MA00011" group of ss1));
  
  7 deleted rows.
  
  SQL > commit;
  
  Validation complete.
  
  SQL >

  As first_insert_date is a DATE data type, its data probably contains also the hours, minutes, and seconds.

  Therfore your WHERE clause... .or first_insert_date = to_date ('2010-09-07', 'YYYY-MM-DD') will always return 0 rows.

  It returns the correct number of lines, your WHERE clause must be... where trunc (first_insert_date) = to_date ('2010-09-07', 'YYYY-MM-DD')

• Problems with utl_https, Oracle Wallet and firewall

  Hi all
  We experiment utl_http and Oracle wallet and try to make a transfer of https, but we are facing some problems. I'd appreciate your help greatly if you can advise on what could be wrong. We are on db version 10.2.0.1 and Unix HP - UX and operate since in a firewall. The intention is a https url ping command and get a 200 simple answer. Future development would include documents XML get/post to this url and other interesting things. I understand that utl_http with Oracle wallet can be used for this purpose.
  
  The portfolio was created and the ewallet.p12 exists. We downloaded the url Web site SSL certificate and downloaded in the portfolio. Everything works if I put in a url with http simple but then things work with a HTTP * S * url. Is what is called HTTPS TUNNELING required because we have a firewall? I don't know what it is or how it can be done.
  I tried https with a URL internal breast of the firewall. But again, no luck. -So probably not just a firewall problem.
  With HTTPS when I run the code below, I get the following error with https internal or external sites. Yet once, greatly appreciate your time and help because it is the first time us use Oracle wallet manager and don't know where to go from here.
  
  ORA-29273: HTTP request failed
  ORA-06512: at "SYS." UTL_HTTP", line 1029
  ORA-29268: error of the HTTP client
  
  declare
  URL varchar2 (225);
  Utl_http.req req;
  resp utl_http.resp;
  my_proxy BOOLEAN;
  name varchar2 (2000);
  value varchar2 (2000);
  V_proxy VARCHAR2 (2000);
  v_n_proxy varchar2 (2000);
  v_msg varchar2 (100);
  v_len PLS_INTEGER: = 1000;
  BEGIN
  -Disable the verification of the status code.
  Utl_http.set_response_error_check (false);
  -Set proxy server
  Utl_http.set_Proxy ('my-proxy');
  Utl_http.set_wallet ("file: < Unix full path to the wallet on DB server > ',' wallet998'");
  Req: = utl_http.begin_request ('https://service.ariba.com/service/transaction/cxml.asp');
  -Authentication proxy set
  Utl_http.set_authentication (req, 'myproxyid', 'myproxypswd', 'Basic', TRUE); -Use HTTP Basic
  
  resp: = utl_http.get_response (req);
  
  FOR i IN 1.utl_http.get_header_count (resp) LOOP
  Utl_http.get_header (RESP, i, name, value);
  dbms_output.put_line(Name ||) ': ' || (value);
  END LOOP;
  Utl_http.end_response (resp);
  exception
  while others then
  dbms_output.put_line (SQLERRM);
  END;

  user11992646 wrote:

  The difference between your method and this new method is that you spend the user_name: pswd as part of the url, so here it is passed as a parameter of the set_proxy.

  Using the user name and password in the proxy URL is often the easiest way to deal with authentication of the proxy - even supported under Linux with the environment variable http_proxy (that you can set for s/w to detect which is the proxy configuration, when access to the web).

  Would be in him passing a part of set_proxy also be considered as in "clear text"?

  Don't know exactly what it generates http traffic. Likely basic auth?

  It may seem that some form of "+ encryption +" is done with basic authentication. The user name and password are strung with a colon separator and then encoded in base64. So, the resulting string seems encrypted to the human eye, but can easily be decoded again (no secret necessary to perform decoding).

  So this is about as insecure as passing user name and password in clear text - darkening bit added to the meter reading it naked.

  In the Windows world (where it is stuck inside and look at the free and open to the outside Standards), proxy servers often use NTLM (new technology/LAN Manager) authentication. Not this again as LanManager goes back to BACK 5 project of the 1980s. ;-)

  Anyway, NTLM is a proprietary protocol and not easily supported when you code using UTL_HTTP. Basic authentication will fail in this case, the proxy expects that NTLM authentication. The workaround is to add the user name and password to the URL of the proxy (as I did in this example code).

  My problem is why set_authentication does not require us to use methods "unsupported"? It seems an obvious Oracle bug that they must solve.

  I think that you also use NTLM authentication on the proxy server? This isn't a standard. It is the crass seller proprietrary. No real compelling reason for other providers to support - especially not when there are open standards available and used.

  We have a beautiful SR and you said that you have opened a SR before without success. I'm surprised Oracle is dropping the ball on such a useful tool.

  My SR has been essentially around get UTL_HTTP proxy access work with NTLM - where workaround has been as I described. I don't see this as a particular problem, Oracle have to solve.

  The question is a consequence of the decision to use proprietary protocols and "+ standards. If anyone is to blame, it's Microsoft and their attempts at blocking of clients in a complete solution of owner. Unfortunately, these attempts often succeed too.

• In the Index tracking problem

  We have allowed index followed for some indexes in our database. We can now verify which index is used or not, as well as the monitoring of State in $object_usage, but suddenly after one day followed by status in v$ object_usage turned automatically 'NO' without doing anything, is this oracle bug?

  Database version: 11.2.0.3 (RAC)

  Please help me.

  index rebuild will not affect index followed, but if there was no change in the structure of the index, then you will need to restart the monitoring for the index again.

• ORA-07445: taken exception [ACCESS_VIOLATION]

  Hello

  All of a sudden we get the error below.

  Thu Jul 02 12:30:23 2015

  Exception [type: ACCESS_VIOLATION, UNABLE_TO_READ] [ADDR:0 x 100] [PC:0x7FF459D1395, kglpin () + 501]

  Thu Jul 02 12:30:24 2015

  Errors in the fichier...\devdb\devdb\cdump\devdbcore.log

  ORA-07445: taken exception [ACCESS_VIOLATION] to [kglpin () + 501] [0x000007FF459D1395]

  DBA said that after you disable a WORK Planner, this is error is no longer displayed.

  What should I check? Can someone please clarify?

  BANNER CON_ID

  Database Oracle 12 c Enterprise Edition Release 12.1.0.1.0 - 64 bit Production 0

  PL/SQL Release 12.1.0.1.0 - Production 0

  CORE 12.1.0.1.0 Production 0

  AMT for 64-bit Windows: Version 12.1.0.1.0 - Production 0

  NLSRTL Version 12.1.0.1.0 - Production 0

  ORA-00600/ORA-07445/ORA-03113 = Oracle bug-online research on Metalink or contact Oracle support

• JDBC: thin gives error mystery

  Hi all

  I'm using Oracle 11 G and I have a java clause that is used to connect the other DB and extract the data. Here is my Echappementjdbc

  Public Shared Sub query (string usr, String pssw, String hostname, String, String DB, String InitialDate port) survey of exceptions
  {
  Class.forName ("oracle.jdbc.OracleDriver");
  String url = "jdbc:oracle:thin:@//" + name + ":" + port + "/" + DB;
  Jacot connection = DriverManager.getConnection (url, usr, pssw);
  try {}
  Statement stmt = connIW.createStatement ();

  Now I get

  ORA-29532:
  Appeal of Java completed by eception Java exception: java.sql.SQLException:
  ORA-01017: name of user and password invalid. connection refused

  But I can't able to connetc with Sql developer, using the same user name, password, HOST, port, and DB Service name .

  Could help me solve the problem.

  Kind regards

  San.

  We have fixed the problem. This
  is an Oracle Bug, please see the Oracle note for reference below.

  Connection to the external database
  Oracle Jvm using the Jdbc thin driver server-side throws Ora-01017 (Doc ID
  1303712.1)

  Applies to: Version - JDBC: 11.2.0.1 and later [Release: 11.2 and later] Information in this document applies to any platform. Symptoms A Java stored procedure using the JDBC/thin driver side server to connect to another database. When Java code is deployed on a pre - 11 g of the side database server JDBC/thin connection works very well. Once the Java code is deployed on a 11.2 the database-side JDBC/thin server failed to connect with ORA-01017: name of user and password invalid. connection refused If the Java code is executed on an external JAVA virtual machine connection works fine. Despite the error ORA-01017 it was determined that the user name and password are correct, and that it is not a matter of case sensitivity. Cause Corruption with the JIT compiled version of Java code. Oraclei¿½ The database Java Developer's Guide 11g Release 1 (11.1) Part number B31225-05 "Overview of the JITStarting JVM Oracle with Oracle 11 g release 1 (11.1), There's a just-in-time (JIT) compiler for Oracle JVM environment. A JIT for Oracle JVM compiler allows much faster execution, because it manages the invalidation, recompilation and code without an external storage mechanism. Based on profiling data collected dynamically, this compiler seamlessly selects the Java methods to compile to native machine code and dynamically make them available for Java sessions running. In addition, the compiler can take advantage of the Oracle JAVA virtual machine class model to possibly persist methods Java compiled in the database of calls, sessions, or instances. This persistence avoids unnecessary recompilations overload through sessions or instances, when we know that semantically the Java code has not changed. " Solution Workaround solution: change the system java_jit_enabled = FALSE; This allows the JIT code to be bypassed, by simply interpreting the Java the code that is generated by bytecode, rather than using the JIT. There is no need to re - compile Java code. Just run the command above

  
  and repeat the Java stored procedure.

  To restore the functionality of JIT just do:

  change the system
  java_jit_enabled = TRUE;

  Solution:

  Delete all the lines in the java table $ mc$ and restart the database

  The java$ mc$ table stores the persistent version of the JIT code generated
  for performance reasons. remove this code apparently requires the JIT to
  regenerate the code.

• License/support options small developer company

  Hi all

  This probably isn't the right place to ask this question, but I couldn't find a better place...

  We are a small business (2 people) and we have developed a client/server application that uses either oracle (10 g 2 and more) or MS Sql server as a backend. We have close to 100 customers who use oracle version, and they are all on different versions of the DBMS. The problem which we face sometimes when solving problems for our customers, it's that we cannot reproduce the reported problem because we do not have the same version of the database that we don't have the downloadable versions for free form the oracle site. In addition, every now and then we run in oracle bugs for which patches appear to be available, but we are not can access tot these fixes. What we seek is an oracle license that gives us access to all versions of oracle and all patches that not too expensive us (we already have a hard time making a living out of our products ;-)). Anyone know if oracle offers such a license and give us a pointer to more information on that?

  Thank you

  Gertjan (aka no Nevaplus)

  Finally the discussion, I just got a message from Oracle that does the silver partnership offers early access to older versions (either electronically or through CD-ROMs), so we will apply for this partnership. Thank you all for the helpful tips.

  Gertjan

• Apply Patch 6400501

  Dear Experts,

  I put DB 11.1.0.7 on R12.1.1 to 11.2.0.4.

  As a conditional requirements I must apply Patch 6400501 .  I read by read me file, I read this note under conditions:

  -Must have located this patch via a bug database entry and have the exact symptoms described in the bug entry.

  My question: what is the bug database entry? How can I find the patch in? How can I determine if this patch is suitable or not?

  Kind regards

  Ahmad,

  Ahmad,

  There is nothing to do about this step, just make sure that you download the patch appropriate via the site Web of MOS and complete the rest of the steps in the README file.

  Entry database of bugs is the internal name of the site hosting for Oracle/bug fixes.

  Thank you

  Hussein

• Comparison of XPath to a node empty in xmltable oddity

  Hello

  I'm on 11.2.0.3 and trying to figure out if I can do a comparison logic in my XMLTABLE. I thought I'd seen pretty much sussed, except when it came to compare to an empty node. Then things began to get weird when I was trying to work where my test went wrong.

  So it works:

  Select x.*
  from xmltable (' / has "passer-by xmltype (" < a > < b > < c > < d > < / d > < e > vale < /e > < / c > < / b > < / has > ' "))
  number of columns c_val_check path "count(b/c[d=""])."
  d_val_check path number 'count(b/c[e="vale"])') x;

  C_VAL_CHECK D_VAL_CHECK
  ----------- -----------
  1           1

  The same way:

  with sd as (select 1 id, xmltype ("< a > < b > < c > < d > < / d > < e > vale < /e > < / c > < / b > < / has > '") double col1)
  Select x.*
  the SD,.
  XMLTable ('/ ' in passing sd.col1)
  number of columns c_val_check path "count(b/c[d=""])."
  d_val_check path number 'count(b/c[e="vale"])') x;

  C_VAL_CHECK D_VAL_CHECK
  ----------- -----------
  1           1

  But it does not work on the d = "" test, even if the two xmls are the same as in the previous two examples: "

  with sd as (select 1 id, xmltype ("< a > < b > < c > < d > < / d > < e > vale < /e > < / c > < / b > < / has > '") col1 in double union all)
  Select 2 id, xmltype ("< a > < b > < c > < d > < / d > < e > vale < /e > < / c > < / b > < / has > '") double col1)
  Select sd.id, x.*
  the SD,.
  XMLTable ('/ ' in passing sd.col1)
  number of columns c_val_check path "count(b/c[d=""])."
  d_val_check path number 'count(b/c[e="vale"])') x;

  ID C_VAL_CHECK D_VAL_CHECK
  ---------- ----------- -----------
  1           0           1
  2           0           1

  And if I change the < c > < e > value to something different, which does not help also:

  with sd as (select 1 id, xmltype ("< a > < b > < c > < d > < / d > < e > vale < /e > < / c > < / b > < / has > '") col1 in double union all)
  Select 2 id, xmltype ("< a > < b > < c > < d > < / d > < e > val0 < /e > < / c > < / b > < / has > '") double col1)
  Select sd.id, x.*
  the SD,.
  XMLTable ('/ ' in passing sd.col1)
  number of columns c_val_check path "count(b/c[d=""])."
  d_val_check path number 'count(b/c[e="vale"])') x;

  ID C_VAL_CHECK D_VAL_CHECK
  ---------- ----------- -----------
  1           0           1
  2           0           0

  What I am doing wrong? I guess I am doing something stupid here, and is not an Oracle bug!

  Hello

  I guess I am doing something stupid here, and is not an Oracle bug!

  Well, maybe it is.

  To use the indicator NO_XML_QUERY_REWRITE works around the problem, as does using a slightly different predicate:

  SQL > with sd as (select 1 id, xmltype ('vale') double col1)

  2 Union all the

  3 select 2 id, xmltype ('vale') double col1

  4              )

  5 Select sd.id, x.*

  6 of the sd,

  7 xmltable ('/ ' in passing sd.col1)

  path number 8 columns c_val_check ' count (b/c [not (d / text ())]). "

  9 number of d_val_check path 'count(b/c[e="vale"])') x;

  ID C_VAL_CHECK D_VAL_CHECK

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

  1           1           1

  2           1           1

• Inside the table in Widows7 Date.MinValue

  Column with date. MinValue added to a table allows you to enter the date
  in Windows XP by clicking on the calendar icon. In Windows 7, this has happened.
  Application shows no reaction on the mouse, click.

  < af:table value =...

  < af:column id = "c5" >

  < af:inputDate label = "Label 2" id = "id2" / >

  < / af:column >

  < / af:table >

  I have DDeveloper 11.1.1.5

  Thank you

  SNikiforov

  Hello

  Does anyone have answer to how to fix this bug in Oracle?

  
  

  We lack evidence that it is an Oracle bug and still think it might be a user error. Have you considered blockers popups as the origin of the problem? So please check and confirm that this problem happens on different browsers, or provide a test case that shows the default. If you want me to send the test case, please zip the test case and rename the file to "decompress." Then send it to frank nimphius oracle com I would be grateful if you have tested this on a second browser before you create and send the test case. My tests would be on 11.1.1.6 or 11.1.1.7 as I do not think that there is a problem it will be be set for 11.1.1.5 (although I am sure that there are no problems with the component we would have heard about this before)

  Frank

• transform xml into slow rational table

  my xml file looks like this

  has about 3000 'Sidonie' (tag < row >)

  and is really simple

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

  <? XML version = "1.0" encoding = "ISO-8859-1? >
  < adapt_xml >
  < row >
  < Field1 > 34093 < / field1 >
  < Field2 > < / Field2 >
  ...
  < field50 > 34 < / field50 >
  < liste_photos >
  adf_photo_2.jpg < picture > < / photo >
  adf_photo_7.jpg < picture > < / photo >
  adf_photo_4.jpg < picture > < / photo >
  ....
  < / liste_photos >
  ...

  
  < / row
  < row >
  < Field1 > 34093 < / field1 >
  < Field2 > < / Field2 >
  ...
  < field50 > 34 < / field50 >
  < liste_photos >
  adf_photo_2.jpg < picture > < / photo >
  adf_photo_7.jpg < picture > < / photo >
  adf_photo_4.jpg < picture > < / photo >
  ....
  < / liste_photos >
  ...

  
  < / row >
  < row >
  < Field1 > 34093 < / field1 >
  < Field2 > < / Field2 >
  ...
  < field50 > 34 < / field50 >
  < liste_photos >
  adf_photo_2.jpg < picture > < / photo >
  adf_photo_7.jpg < picture > < / photo >
  adf_photo_4.jpg < picture > < / photo >
  ....
  < / liste_photos >
  ...

  
  < / row >
  < row >
  < Field1 > 34093 < / field1 >
  < Field2 > < / Field2 >
  ...
  < field50 > 34 < / field50 >
  < liste_photos >
  adf_photo_2.jpg < picture > < / photo >
  adf_photo_7.jpg < picture > < / photo >
  adf_photo_4.jpg < picture > < / photo >
  ....
  < / liste_photos >
  ...

  
  < / row >
  < row >
  < Field1 > 34093 < / field1 >
  < Field2 > < / Field2 >
  ...
  < field50 > 34 < / field50 >
  < liste_photos >
  adf_photo_2.jpg < picture > < / photo >
  adf_photo_7.jpg < picture > < / photo >
  adf_photo_4.jpg < picture > < / photo >
  ....
  < / liste_photos >
  ...

  
  < / row >
  ...
  < / adapt_xml >

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

  the XML is stored in an xml table in Oracle (Version 11.2...)

  I have to extract the values a store in a relational table

  to do this, I make the following declarations

  It's fast

  SELECT
  x1.field1,
  x 1. Field2,
  x1.field3,
  ....
  x 1. Field50
  OF tmp_xml.
  XMLTable)
  ' / / row '
  by the way OBJECT_VALUE
  columns
  path of Field1 "Field1."
  path of Field2 'Field2 ',.
  field3 path "field3"
  ...
  Field50 path 'field50 '.
  ) x 1

  
  It's really slow

  SELECT
  x1.field1,
  x 1. Field2,
  x1.field3,
  ....
  x 1. Field50,
  x 1 .photo_1,
  x 1 .photo_2,
  x 1 .photo_3,
  x 1 .photo_4,
  x 1 .photo_5,
  x 1 .photo_6,
  x 1 .photo_7,
  x 1 .photo_8,
  x 1 .photo_9,
  x 1 .photo_10,
  x 1 .photo_11,
  x 1 .photo_12,
  x 1 .photo_13
  OF tmp_xml.
  XMLTable)
  ' / / row '
  by the way OBJECT_VALUE
  columns
  path of Field1 "Field1."
  path of Field2 'Field2 ',.
  field3 path "field3"
  ...
  Field50 path "field50."
  picture_1 path "liste_photos/photo [1]."
  photo_2 path "liste_photos/photo [2]."
  picture_3 path "liste_photos/photo [3]."
  photo_4 path "liste_photos/photo [4]."
  photo_5 path "liste_photos/photo [5]."
  photo_6 path "liste_photos/photo [6]."
  photo_7 path "liste_photos/photo [7]."
  photo_8 path "liste_photos/photo [8]."
  photo_9 path "liste_photos/photo [9]."
  photo_10 path "liste_photos/photo [10]."
  photo_11 path "liste_photos/photo [11]."
  photo_12 path "liste_photos/photo [13]."
  photo_13 path "liste_photos/photo [13].
  ) x 1

  
  It's slow as well

  SELECT
  x1.field1,
  x 1. Field2,
  x1.field3,
  ....
  x 1. Field50,
  x 1 .liste_photos,
  OF tmp_xml.
  XMLTable)
  ' / / row '
  by the way OBJECT_VALUE
  columns
  path of Field1 "Field1."
  path of Field2 'Field2 ',.
  field3 path "field3"
  ...
  Field50 path "field50."
  path liste_photos xmltype "liste_photos."
  ) x 1

  
  I tried to cash some xmltype index
  but without success

  the selection of the Field1... field50 is fast

  But if I select the photo list the performance is down

  Thank you

  Martin

  Finally, it works

  It has been a long road to success

  in the end, it was an oracle bug

  I opened a SR with Oracle in September

  After some time without any progress in the SR (no reply) I let it worsen the SR.

  not much happened until 24 x 7 climbing

  Since I guess it was in December (3 months later) they are really working on my SR and they had the same problems that I have

  in the end, that they discovered that it was because of the UTF8 character set

  the bug is now fixed in paragraph 12.2 and I got a patch stop

  If someone is interested in the patch

  January 16, 2014

  QUERYING A TABLE WITH XMLTYPE COLUMN PERFORMANCE IS BAD ON THE UTF8 DB CHARACTER SET [Patch ID 17180523]

  

  January 19, 2014

  QUERYING A TABLE WITH XMLTYPE COLUMN PERFORMANCE IS BAD ON THE UTF8 DB CHARACTER SET [Bug ID 17928272]

• No more data to read from socket

  Hello
  I am trying to run this query which keep away causing the error no additional read decision-making data
  

  SELECT distinct SOR_COMPLIANCE_ADMIN.get_Completed_Letter_Id(o.offender_id, 'OFFENDER'), 'Mailed', sysdate, null, o.offender_id, 
              l.location_id, o.first_name, o.middle_name, o.last_name, o.sir_name, l.address1, l.address2, l.city, REPLACE(sor_data_admin.getCode(l.state),'0','O'), l.zip, l.county, 
              sor_data_admin.getOffenderType(o.offender_id), 
              null, l.JURISDICTION, decode(sor_data_admin.getCode(o.sex),'M','Mr.', 'F','Ms.','') 
              from sor_compliance_com_v t, registration_offender_xref x, sor_location l, sor_offender o
              where x.status='Active'
              and x.offender_id=o.offender_id
              and o.offender_id=l.offender_id
              and t.offender_id=o.offender_id
              and t.location_id=l.location_id ;

  definition of get_Completed_Letter_Id

   function get_Completed_Letter_Id(p_offender_id number,  p_destination varchar2) return number
          IS
     
     
                  /*cursor C is used to get letter template ID for offenders whose tier are 1, 2*/
                  CURSOR c IS 
                  SELECT  sle.letter_id
                  FROM sor_letter sle
                  WHERE sle.destination=p_destination
                  and upper(sle.letter_type)='COM';
      
                  tLetterId number;
                  error_message varchar2(500);
                  
          BEGIN
          
          
  
                      OPEN c;
                      FETCH c INTO tLetterId;
                      CLOSE c;
                    
                      RETURN tLetterId;
  
    
          EXCEPTION
              when others then
                  error_message:=substr(SQLERRM, 1, 400);
                  raise_application_error(SOR_ERROR_CONSTANTS_PKG.errnum_GENERAL, 'SOR_COMPLIANCE_ADMIN.getLetterId '||SOR_ERROR_CONSTANTS_PKG.errmsg_GENERAL||error_message);
                  
          END get_Completed_Letter_Id;

  definition of sor_data_admin.getOffenderType
  

    function getOffenderType(p_offender_id number) return varchar2 AS
        
              error_message varchar2(500);
        
              CURSOR c IS 
              SELECT decode(MAX(habitual)||MAX(aggravated),'NN','STANDARD','LIFETIME')
              FROM registration_offender_xref
              WHERE offender_id = p_offender_id
              AND status = 'Active';
              -- doc/hkt 111407 New Law: calculate offender type by tier (level)     
              
              cursor tierNum is
              select max(c.tier) from sor_offense o, sor_offense_code c
              where o.offender_id = p_offender_id
              and o.offense_code = c.CODE_ID
              and upper(o.status) = 'ACTIVE';
            
              tier number;
              vOffenderType varchar2(30);
              
        BEGIN
        
              OPEN c;
              FETCH c INTO vOffenderType;
              
              IF c%NOTFOUND THEN
                vOffenderType := 'INACTIVE';
              END IF;
              CLOSE c;
          
              open tierNum;
              Fetch tierNum into tier;
              if tierNum%NotFound then
                tier := Null;
              end if;
              Close tierNum;
        
              if vOffenderType <> 'INACTIVE' then
                    if tier = 3 then 
                      vOffenderType := 'LIFETIME';
                    end if;
              end if;
          
              RETURN vOffenderType;
              
        EXCEPTION
        
            WHEN OTHERS THEN
                  error_message:=substr(SQLERRM, 1, 400);
                  raise_application_error(SOR_ERROR_CONSTANTS_PKG.errnum_GENERAL, 'SOR_COMPLIANCE_ADMIN.getOffenderType '||SOR_ERROR_CONSTANTS_PKG.errmsg_GENERAL||error_message);
        
        END getOffenderType;

  Definition of getcode

   FUNCTION getCode(codeID IN NUMBER) RETURN VARCHAR2 AS
            code sor_code.code%TYPE;
            
        BEGIN
                SELECT code INTO code FROM sor_code WHERE code_id = codeID;
                return(code);
        
        END getCode;

  Thanks for your help

  ORA-00600/ORA-07445/ORA-03113 = Oracle bug-online research on Metalink or contact Oracle support

• A lot of questions of a model in Web Forms

  Hello, I recently installed in my Pc (Windows 7 Home Premium), home
  Database 11 g and forms 11.1.2.00 (64-bit) just for training,
  because I have one experience with 6i (Client-Server) and I wanted to see the new environment.
  
  I have my previous employment (unemployed now) a small (30 fmb) application to the 6i and want to run in the Web.
  I looked a little pdf "Upgrading Oracle Forms 6i with Oracle Forms 11g (E24478-01)".
  
  My fmxs and PLL is in the C:\Retail_app\bin folder.
  Also in this folder I copied EnableDisableItem.pll and rp2rro.pll
  Sources (fmb, rdf) located in the folder C:\Retail_app\Scr.
  
  When I open the web, it seems to me FRM-40039 cannot attach retail.pll library while opening the form shop.fmx
  
  I have search the forum for this FRM and watched error FRM-40039 post 32-bit to 64-bit migration. Please help.
  
  Here, a professional oracle says that we have to generate all the PLL in plx.
  
  But when I try to do, he told me FRM-91507 internal error: unable to generate the library.
  
  The expert also said that it is important that you make sure that the file locations are included in FORMS_PATH (default.env).
  
  My FORMS_PATH has 'C:\Retail_App\Bin; C:\Oracle\Middleware\FRMHome\forms; C:\Oracle\Middleware\FrmInst\FormsComponent\forms ".
  
  I do not understand what default.env means. I'm new to Web Forms environment.
  
  
  Also another question, because I'm Greek, the form builder environment is half English and half Greek.
  He does not say, 'Libraries attached' or 'Units of program' or in the Menu it says "Parathyro' instead of 'Window' I want everyone in English.
  Of course my application language is Greek, but I want to Form Builder to be English.
  My NLS_LANG is
  GREEK_GREECE. EL8MSWIN1253
  
  Is this wrong?
  
  Also I want to say
  When I opened shop.fmb in FormBuilder and go to the retail.pll of the library, then go to the properties of the library, to the location of functional PL/SQL library, he say
  Retail.PLL.
  It does not say C:\Retail_App\bin.
  
  I tried this.
  
  remove the shop.fmb library, then fix it again, when asked "Remove path? I said no, but when I tried to launch the application it says
  
  FRM-10221, cannot read file C:\Oracle\Middleware\FrmIns «»
  Error FRM-41810 menu creation.
  
  Preform Trigger raised an unhandled exception 'ORA-06508 ".
  
  
  I put messages in the trigger of PREFORM of shop.fmb,.
  relaxation begins...
  
  : system.message_level: = '20';
  message ('1');
  Create_Block_Rg; -There is a prgram of the Retail.pll library unit
  message ('2');
  : system.message_level: = '0';
  
  
  When I started the application again, it occurred to me message 1 and failure, which means that he could not even read the pll.
  
  
  Thank you all and wish you a good year (with joy and work).
  
  Published by: apollon27 on December 31, 2012 15:47

  In short, the passage from one version to the other is usually easy. In many cases, all that is required is that your generate new 'X' files. In other words, you must use the compiler of forms 11 or the constructor to generate new FMX, MMX and PLX files. Also, you can't wait to use a pll during execution. Pll files should be generated in plx files. New files that results should be located in a path that is included in FORMS_PATH. For the time of execution (browser), FORMS_PATH is located in default.env. For the moment of conception (while working in Report Builder), FORMS_PATH is set in the registry in the same area that contains other entries for forms. Keep in mind that when I say 'DTC' this does not refer to the launch of the constructor of a form in a browser. Any time a form is running in a browser, its config from default.env.

  In addition, in most cases, I would not recommend that when you attach a library you choose not to delete the path. In other words, do not include the path when prompted. This is because if the path is chosen, the form will not work correctly if the application is moved. Instead, choose to remove the path when attaching and then make sure that the FORMS_PATH include the location of your libraries.

  So, to your accurate comments:

  1 FRM-91507 Internal error: unable to generate the library. It is usually the result of FORMS_PATH not defined correctly or not, including the path where the files of references. It was also reported this error resulting from setting NLS_LANG in some languages. This problem has been reported in Oracle bug 4131359. This bug has not yet been fixed. To determine if this is the problem, close the generator and try to set NLS_LANG to something like the following:
  

  AMERICAN_AMERICA. UTF8

  
  This can either be defined in the registry or as a Windows environment variable. After you set this value, restart the contractor or the compiler and retest.

  2. you said, ' + I don't understand what default.env + '. You should familiarize yourself with this file like you he will come back often. It is the configuration file that contains the environment settings used by the runtime of forms (frmweb.exe). In version 11, this file can be found here:
  

  C:\Oracle\Middleware\user_projects\domains\ClassicDomain\config\fmwconfig\servers\WLS_FORMS\applications\formsapp_11.1.2\config

  
  In most cases, Oracle recommends that all configuration changes are made to in Enterprise Manager or Console WLS. The only exceptions would be if either of these tools fail to a configuration file individual or if you are using the type of installation of 'development '. Installation types are explained here:

  http://docs.Oracle.com/CD/E24269_01/doc.11120/e23960/install_screens.htm#BABCHIAF