Skip to Main Content
Forums

Berkeley DB Family

Announcement

For appeals, questions and feedback about Oracle Forums, please email oracle-forums-moderators_us@oracle.com. Technical questions should be asked in the appropriate category. Thank you!

OutOfMemoryError when writing?

nitroamosOct 5 2016 — edited Oct 5 2016

I've incorporated je berkeleydb into my application and it seems to be working very well. Today a user reported the exception I pasted below. As far as I can tell from the log file, heap memory was available. I've pasted my config below, too.

Any ideas?

java.lang.OutOfMemoryError: null

  at java.io.RandomAccessFile.writeBytes0(Native Method) ~[na:1.7.0_45]

  at java.io.RandomAccessFile.writeBytes(Unknown Source) ~[na:1.7.0_45]

  at java.io.RandomAccessFile.write(Unknown Source) ~[na:1.7.0_45]

  at com.sleepycat.je.log.FileManager.writeToFile(FileManager.java:1749) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.je.log.FileManager.writeLogBuffer(FileManager.java:1663) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.je.log.LogBufferPool.writeBufferToFile(LogBufferPool.java:449) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.je.log.LogBufferPool.writeDirty(LogBufferPool.java:408) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.je.log.LogManager.log(LogManager.java:365) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.je.tree.LN.logInternal(LN.java:728) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.je.tree.LN.optionalLog(LN.java:447) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.je.dbi.CursorImpl.insertRecordInternal(CursorImpl.java:1478) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.je.dbi.CursorImpl.insertOrUpdateRecord(CursorImpl.java:1280) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.je.Cursor.putNoNotify(Cursor.java:2504) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.je.Cursor.putNotify(Cursor.java:2365) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.je.Cursor.putNoDups(Cursor.java:2223) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.je.Cursor.putInternal(Cursor.java:2060) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.je.Cursor.put(Cursor.java:730) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.util.keyrange.RangeCursor.put(RangeCursor.java:1066) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.collections.DataCursor.put(DataCursor.java:788) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.collections.StoredContainer.putKeyValue(StoredContainer.java:359) ~[je-6.4.25.jar:6.4.25]

  at com.sleepycat.collections.StoredMap.put(StoredMap.java:315) ~[je-6.4.25.jar:6.4.25]

   private void initDb(File file) throws IOException {

      log.info("opening cache " + file);

      EnvironmentConfig envConfig = new EnvironmentConfig();

      envConfig.setTransactional(false);

      envConfig.setAllowCreate(true);

      DbInternal.disableParameterValidation(envConfig);

      envConfig.setConfigParam(EnvironmentConfig.CLEANER_MIN_UTILIZATION,"90");

      envConfig.setConfigParam(EnvironmentConfig.LOG_FILE_MAX,"10000000");

      envConfig.setDurability(Durability.COMMIT_NO_SYNC);

      envConfig.setCacheSize(128000000);

      env = new Environment(file,envConfig);

      DatabaseConfig dbConfig = new DatabaseConfig();

      dbConfig.setTransactional(false);

      dbConfig.setAllowCreate(true);

      Database catalogDb = env.openDatabase(null,CLASS_CATALOG,dbConfig);

      classCatalog = new StoredClassCatalog(catalogDb);

      db = env.openDatabase(null,"mydatastore",dbConfig);

      // setup binding

      EntryBinding<K> keyBinding = new SerialBinding<K>(classCatalog,keyType);

      EntryBinding<V> valueBinding = new SerialBinding<V>(classCatalog,

            valueType);

      keyValueMap = new StoredMap<K, V>(db,keyBinding,valueBinding,true);

   }

Comments

Keith Jamieson
excellent article. However there is one thing which is slightly erroneous. You don't need a type to be declared in the database to fetch the data, but you do need to declare a type;

here is one of my unit test scripts that does just that.

DECLARE
PN_CARDAPPL_ID NUMBER;
v_Return Cci_Standard.ref_cursor;
type getcardapplattrval_recordtype
Is record
(cardappl_id ci_cardapplattrvalue.cardappl_ID%TYPE,
tag ci_cardapplattrvalue.tag%TYPE,
value ci_cardapplattrvalue.value%TYPE
);
getcardapplattrvalue_record getcardapplattrval_recordtype;
BEGIN
PN_CARDAPPL_ID := 1; --value must be supplied
v_Return := CCI_GETCUSTCARD.GETCARDAPPLATTRVALUE(
PN_CARDAPPL_ID => PN_CARDAPPL_ID
);
loop
fetch v_return
into getcardapplattrvalue_record;

dbms_output.put_line('Cardappl_id=>'||getcardapplattrvalue_record.cardappl_id);
dbms_output.put_line('Tag =>'||getcardapplattrvalue_record.tag);
dbms_output.put_line('Value =>'||getcardapplattrvalue_record.value);
exit when v_Return%NOTFOUND;
end loop;
END;
BluShadow
Keith Jamieson wrote:
excellent article. However there is one thing which is slightly erroneous. You don't need a type to be declared in the database to fetch the data, but you do need to declare a type;
Correct, if you are just fetching the data in PL/SQL. The article, however, is talking about fetching the data within SQL. You won't do that with a locally defined PL/SQL type. ;)
692864
thank you so much for a great article its very nice and helpful

thanks and regards
raj
247514
Nice work Blue and bump. The forum should have sticky or FAQ kinda list for each forum.
764763
Hi everyone !!

I have a question to ask you, and i hope you can help me.

Are Ref Cursors the best solution to call stored procedure from java code ?
Or there's another "best" way to do it ?

And thnx
BluShadow
user11951318 wrote:
Hi everyone !!

I have a question to ask you, and i hope you can help me.

Are Ref Cursors the best solution to call stored procedure from java code ?
Or there's another "best" way to do it ?

And thnx
If you mean, "Are Ref Cursors the best thing to pass back from a stored procedure in order to query data back from the database?" then I would guess the answer is Yes as that is what they are primarily intended for. You can't "call" a stored procedure using a ref cursor as your question implies. However, this is an Oracle forum, not a Java forum, so the best practice for Java would be best asked on a Java forum.
743277
I am executing this code toad but it does show the output of query , How can i get the out put of this code , I M USING TOAD

set serveroutput on ;
DECLARE
CURSOR c8
IS
SELECT (a.invoice_id) l_name,
(a.invoice_id) f_name
FROM ap_invoices_all a;
BEGIN
FOR r_c8 IN c8
LOOP
DBMS_OUTPUT.put_line (r_c8.l_name || ' ' || r_c8.l_name);
END LOOP;
END;
Boneist
Either add the following line before your procedure and run it as a script:
set serveroutput on
Or click the red button (it will turn green) on the DBMS Output (Disabled) tab, run your query and check in that tab for the dbms_output output.
BluShadow
user11995078 wrote:
I am executing this code toad but it does show the output of query , How can i get the out put of this code , I M USING TOAD
Boneist has already given you the answer, but in future, consider posting your question on a thread of your own rather than hijacking something that is completely unrelated to your question.
743277
thanks Boss
Eight Six
Hi blueshadow,

Nice explanation :)

Just quick question in the below code you are closing the ref cursor before fetching is that correct?

SQL> ed
Wrote file afiedt.buf

1 declare
2 v_rc sys_refcursor;
3 v_empno number;
4 v_ename varchar2(10);
5 v_mgr number;
6 v_sal number;
7 begin
8 v_rc := get_dept_emps(10); -- This returns an open cursor
9 loop
10 fetch v_rc into v_empno, v_ename, v_mgr, v_sal;
11 exit when v_rc%NOTFOUND; -- Exit the loop when we've run out of data
12 dbms_output.put_line('Row: '||v_rc%ROWCOUNT||' # '||v_empno||','||v_ename||','||v_mgr||','||v_sal);
13 end loop;
14 close v_rc;
15 fetch v_rc into v_empno, v_ename, v_mgr, v_sal;
16* end;
SQL> /
Row: 1 # 7782,CLARK,7839,2450
Row: 2 # 7839,KING,,5000
Row: 3 # 7934,MILLER,7782,1300
declare
*
ERROR at line 1:
ORA-01001: invalid cursor
ORA-06512: at line 15
BluShadow
user10857924 wrote:
Hi blueshadow,

Nice explanation :)

Just quick question in the below code you are closing the ref cursor before fetching is that correct?
It is if you were actually reading the article:
And what happens if we try and fetch more data after it's finished, just like we tried to do in SQL*Plus..
{snip code}
As expected we get an error.
manjukn
Thanks Blue..
you have explained the concept in a beautiful way
1 person found this helpful
user7540156
i am trying to execute the below code.
but it's returning date only.
i want date and time.
declare
type r_cursor is REF CURSOR;
c_emp r_cursor;
en date;
begin
open c_emp for select sysdate from dual;
loop
fetch c_emp into en;
exit when c_emp%notfound;
dbms_output.put_line(en);
end loop;
close c_emp;
end;

O/P:25-JAN-12

i want output as date with time.
BluShadow
user7540156 wrote:
i am trying to execute the below code.
but it's returning date only.
i want date and time.
declare
type r_cursor is REF CURSOR;
c_emp r_cursor;
en date;
begin
open c_emp for select sysdate from dual;
loop
fetch c_emp into en;
exit when c_emp%notfound;
dbms_output.put_line(en);
end loop;
close c_emp;
end;

O/P:25-JAN-12

i want output as date with time.
That has absolutely nothing to do with ref cursors.

a) you should declare ref cursors using the inbuilt type SYS_REFCURSOR. There's absolutely no need to decalre your own type for it.
b) the display format of your date is exactly that, a display format issue.
SQL> declare
  2    c_emp SYS_REFCURSOR;
  3    en    date;
  4  begin
  5    open c_emp for select sysdate from dual;
  6    loop
  7      fetch c_emp into en;
  8      exit when c_emp%notfound;
  9      dbms_output.put_line(en);
 10    end loop;
 11    close c_emp;
 12  end;
 13  /
25-JAN-12

PL/SQL procedure successfully completed.
The date variable is being implicitly converted to a varchar2 datetype by the put_line statement which is expecting a varchar2 input parameter. This implicit conversion is being done based on your sessions NLS_DATE_FORMAT setting.

You can either change your sessions NLS_DATE_FORMAT...
SQL> alter session set nls_date_format = 'DD-MON-YYYY HH24:MI:SS';

Session altered.

SQL> declare
  2    c_emp SYS_REFCURSOR;
  3    en    date;
  4  begin
  5    open c_emp for select sysdate from dual;
  6    loop
  7      fetch c_emp into en;
  8      exit when c_emp%notfound;
  9      dbms_output.put_line(en);
 10    end loop;
 11    close c_emp;
 12  end;
 13  /
25-JAN-2012 09:29:13

PL/SQL procedure successfully completed.
or you can explicitly convert the date to a varchar2 in your code using the to_char function, specifying the display format you want...
SQL> ed
Wrote file afiedt.buf

  1  declare
  2    c_emp SYS_REFCURSOR;
  3    en    date;
  4  begin
  5    open c_emp for select sysdate from dual;
  6    loop
  7      fetch c_emp into en;
  8      exit when c_emp%notfound;
  9      dbms_output.put_line(to_char(en,'DD-MM-YYYY HH24:MI'));
 10    end loop;
 11    close c_emp;
 12* end;
SQL> /
25-01-2012 09:29

PL/SQL procedure successfully completed.
As I said, absolutely nothing to do with ref cursors and all to do with a complete lack of understanding of basic datatypes and implicit conversions.
1 person found this helpful
SamFisher
Good one, Blue.

Thx.
gaverill
Nicely done. I would suggest adding a section with a "pipelined" version of your "populate_emps" table function, as often there's no need to fully materialize the table...
SQL> ed
Wrote file afiedt.buf
 
  1  create or replace function /*populate*/pipe_emps(deptno in number := null)
  2  return t_emptype pipelined is
  3    -- v_emptype t_emptype := t_emptype();  -- Declare a local table structure and initialize it
  4    -- v_cnt     number := 0;
  5    v_rc      sys_refcursor;
  6    v_empno   number;
  7    v_ename   varchar2(10);
  8    v_mgr     number;
  9    v_sal     number;
 10  begin
 11    v_rc := get_dept_emps(deptno);
 12    loop
 13      fetch v_rc into v_empno, v_ename, v_mgr, v_sal;
 14      exit when v_rc%NOTFOUND;
 15      --v_emptype.extend;
 16      --v_cnt := v_cnt + 1;
 17      /*v_emptype(v_cnt) :=*/ pipe row ( emptype(v_empno, v_ename, v_mgr, v_sal) );
 18    end loop;
 19    close v_rc;
 20    return /*v_emptype*/;
 21* end;
SQL> / 
 
Function created.
Gerard
BluShadow
gaverill wrote:
Nicely done. I would suggest adding a section with a "pipelined" version of your "populate_emps" table function, as often there's no need to fully materialize the table...
Thanks for the suggestion, however pipelining the data is beyond the 101 basics of ref cursors which is what the article is designed to demonstrate. ;)
1 - 18
Locked Post
New comments cannot be posted to this locked post.

Post Details

Locked on Nov 2 2016
Added on Oct 5 2016
#berkeley-db-java-edition
3 comments
472 views