ORA-01652: unable to extend temp segment by 128 in tablespace PSAPTEMP

Hi Everybody;
My coleague who is an bw consultant facing given below problem in BW production server.
Error reading the data of InfoProvider ZN_IVLTD
Error when generating the sql statement.
ORA-01652: unable to extend temp segment by 128 in tablespace PSAPTEMP
I checked table space PSAPTEMP as well but its showing o% used.
Kindly help how can i resolve this problem.
Regards
Santosh

Hi Santosh,
There is also this thread going on, check if anything will be useful for you.
Two SAP Instances (DEV & QAS) on singe windows host.
One problem that I have noticed on UNIX systems for PSAPTEMP, refer to the this note if your OS is UNIX.
548221 - Temporary Files are created as sparse files
Regards,
Nilesh

Similar Messages

  • ORA-1652: unable to extend temp segment by 128 in tablespace PSAPTEMP

    Hi Gurus,
    We are getting this alert continuosly , anybody please suggest me.
    OS: AIX
    DB:ORACLE 10g
    We disabled STAT file also,, but still we are getting this alert
    please check table spaces,,,
    PSAPTEMP                          10.485.760     10.481.664          0          0          0    Off                  0    ONLINE
    PSAPUNDO                           8.514.344      2.062.912         75         31        951    Off                 75    ONLINE     NOT ACTand file system also fine
    please find below error details..
    02/05/10 10:45:12 Europe/Middle (open.related - System Open Related):
    External Update
    05/02/2010 08:27:41 by CA
    Attached new CA message: 1535095
    .Last matching line that caused this alert: ORA-1652: unable to extend temp segment by 128 in tablespace PSAPTEMP
    BR0301E SQL error -1652 in thread 4 at location stats_tab_collect-20, SQL statement:
    'BEGIN DBMS_STATS.GATHER_TABLE_STATS (OWNNAME => '"SAPP01"', TABNAME => '"GLPCA"', ESTIMATE_PERCENT => NULL, METHOD_OPT => 'FOR ALL COLUMNS SI
    ORA-01652: unable to extend temp segment by 128 in tablespace PSAPTEMP
    ORA-06512: at "SYS.DBMS_STATS", line 13437
    ORA-06512: at "SYS.DBMS_STATS", line 13457
    ORA-06512: at line 1
    BR0886E Checking/collecting statistics failed for table SAPP01.GLPCA
    Regards,
    balaram

    HI,
    ORA-01652: This error is caused by a tablespace overflow. As a rule of thumb, PSAPTEMP should be at least twice as large as the largest index. If an overflow occurs despite this size, this is     frequently due to external reasons such as incorrect Oracle parameter settings, large-scale parallel processing or    unfavorable access paths. See also Note 3155.
    Check - 659946 for details.
    Best Regards
    Niraj

  • ORA-01652: unable to extend temp segment by 128 in tablespace TEMP2

    Dear all,
    received this error when generate a report.
    check the TEMP1 and TEMP2 they all have lot of spaces.
    check the env $APPLLCSP return empty.
    Please advise how to correct this error.
    Regards,
    Payables: Version : 12.0.0
    APXSSIMP module: Supplier Sites Open Interface Import
    APPLLCSP Environment Variable set to :
    Current NLS_LANG and NLS_NUMERIC_CHARACTERS Environment Variables are :
    American_America.UTF8
    Enter Password:
    REP-0069: Internal error
    REP-57054: In-process job terminated:Terminated with error:
    REP-300: MSG-00001: After SRWINIT
    MSG-00002: After Get_Company_Name
    MSG-00003: After Get_NLS_Strings
    MSG-00004: After Importing Suppliers
    MSG-00005: After Get_Header_Information
    ORA-01652: unable to extend temp segment by 128 in tablespace TEMP2
    ==> SELECT assi.org_id C_Rejected_Org_Id,
    MSG-00020: After SRWEXIT
    Report Builder: Release 10.1.2.0.2 - Production on Mon May 11 09:55:36 2009

    >
    check the TEMP1 and TEMP2 they all have lot of spaces.
    >
    How much is "lot" ? :-) How did you determine this "lot" ?
    Pl see MOS Doc 1025288.6 (How to Diagnose and Resolve UNABLE TO EXTEND Errors)
    HTH
    Srini

  • ORA-01652: unable to extend temp segment by 128 in tablespace TEMP

    Hi,
    I am getting the following error while executing an insert query.
    ORA-01652: unable to extend temp segment by 128 in tablespace TEMP.
    I have tried increasing the TEMP tbsp size it got resolved temporarily but occurred again after few days.
    When the problem occured, following query shows all insert queries in TEMP tablespace and one of them consumes about 5GB and others are about 24MB.
    SELECT S.sid || ',' || S.serial# sid_serial, S.username,
    T.blocks * TBS.block_size / 1024 / 1024 mb_used, T.tablespace,
    T.sqladdr address, Q.hash_value, Q.sql_text
    FROM v$sort_usage T, v$session S, v$sqlarea Q, dba_tablespaces TBS
    WHERE T.session_addr = S.saddr
    AND T.sqladdr = Q.address (+)
    AND T.tablespace = TBS.tablespace_name
    ORDER BY S.sid;
    The insert query is :
    INSERT INTO SEOSDATA (ENTRYID,DOMAINNAME,USERNAME,EVENTTYPE,LOGNAME,TIMSTAMP,SOURCE,COMPUTERNAME,EVENTID,EVENTCATEGORY,SEARCHSTRINGS,MSGTEXT) VALUES (ENTRYID_SEQ.NEXTVAL,:1,:2,:3,:4,:5,:6,:7,:8,:9,:10,:11)
    It is not expected that an insert query consumes such a huge memory.
    Can someone help me in figuring out what the problem is.
    What might have gone wrong? Seems like sorting is used, why an insert query uses sorting?
    Regards
    Ishu

    Hi,
    Insert statement should not use TEMP tablespace. Please check with following query which user and session id is using TEMP tablespace.
    SELECT b.tablespace, b.segfile#, b.segblk#, b.blocks, a.sid, a.serial#,
    a.username, a.osuser, a.status
    FROM v$session a,v$sort_usage b
    WHERE a.saddr = b.session_addr
    ORDER BY b.tablespace, b.segfile#, b.segblk#, b.blocks;
    Dilipkumar Patel.

  • Unable to extend temp segment by 128 in tablespace PSAPTEMP

    Hi everybody!
    While processing a UNICODE coversion with SAPINST on my SAP NW2004s system (AIX 5,2  - Oracle 10.2 - BW 7.0) into my.SAPDFACT_1.log file the following errors occured:
    DbSl Trace: ORA-1403 when accessing table SAPUSER
    (DB) INFO: connected to DB
    (DB) INFO: DbSlControl(DBSL_CMD_NLS_CHARACTERSET_GET): UTF8
    (GSI) INFO: dbname   = "HBQ20080503070541                                                                                "
    (GSI) INFO: vname    = "ORACLE                          "
    (GSI) INFO: hostname = "si088ss1                                                        "
    (GSI) INFO: sysname  = "AIX"
    (GSI) INFO: nodename = "si088ss1"
    (GSI) INFO: release  = "2"
    (GSI) INFO: version  = "5"
    (GSI) INFO: machine  = "0031832B4C00"
    (DB) INFO: /BIC/FZHPY_LIM created #20080503200234
    (IMP) INFO: import of /BIC/FZHPY_LIM completed (133182692 rows) #20080503212420
    DbSl Trace: Error 1652 in exec_immediate() from oci_execute_stmt(), orpc=0
    DbSl Trace: ORA-1652 occurred when executing SQL stmt (parse error offset=42)
    (DB) ERROR: DDL statement failed
    (CREATE UNIQUE INDEX "/BIC/FZHPY_LIM~0" ON "/BIC/FZHPY_LIM" ( "KEY_ZHPY_LIMP", "KEY_ZHPY_LIMT", "KEY_ZHPY_LIMU", "KEY_ZHPY_LIM2", "KEY_ZHPY_LIM3", "KEY_ZHPY_LIM4", "KEY_ZH
    PY_LIM5", "KEY_ZHPY_LIM6", "KEY_ZHPY_LIM7" ) TABLESPACE PSAPSR3 STORAGE (INITIAL 65536 NEXT 0000002560K MINEXTENTS 0000000001 MAXEXTENTS 2147483645 PCTINCREASE 0 ) NOLOGGING COMPUTE STATISTICS )
    DbSlExecute: rc = 99
      (SQL error 1652)
      error message returned by DbSl:
    ORA-01652: unable to extend temp segment by 128 in tablespace PSAPTEMP
    (DB) INFO: disconnected from DB
    Anybody can help me?
    Thanks a lot!

    Hi Stefan!
    desc SAPSR3."/BIC/FZHPY_LIM";
    Name                                      Null?    Type
    KEY_ZHPY_LIMP                             NOT NULL NUMBER(10)
    KEY_ZHPY_LIMT                             NOT NULL NUMBER(10)
    KEY_ZHPY_LIMU                             NOT NULL NUMBER(10)
    KEY_ZHPY_LIM2                             NOT NULL NUMBER(10)
    KEY_ZHPY_LIM3                             NOT NULL NUMBER(10)
    KEY_ZHPY_LIM4                             NOT NULL NUMBER(10)
    KEY_ZHPY_LIM5                             NOT NULL NUMBER(10)
    KEY_ZHPY_LIM6                             NOT NULL NUMBER(10)
    KEY_ZHPY_LIM7                             NOT NULL NUMBER(10)
    /BIC/ZHCONTING                            NOT NULL NUMBER(17,2)
    /BIC/ZHD_IMP                              NOT NULL NUMBER(17,2)
    /BIC/ZHPERC2                              NOT NULL NUMBER(10)
    SELECT AVG_ROW_LEN FROM ALL_TABLES WHERE TABLE_NAME = '/BIC/FZHPY_LIM';
    AVG_ROW_LEN
    SELECT BYTES, BLOCKS FROM DBA_SEGMENTS WHERE SEGMENT_NAME = '/BIC/FZHPY_LIM';
         BYTES     BLOCKS
    7180648448     876544
    SELECT FILE_NAME, TABLESPACE_NAME, BYTES, BLOCKS, AUTOEXTENSIBLE, MAXBYTES, MAXBLOCKS FROM DBA_TEMP_FILES;
    FILE_NAME
    TABLESPACE_NAME                     BYTES     BLOCKS AUT   MAXBYTES  MAXBLOCKS
    /oracle/HBQ/sapdata1/temp_1/temp.data1
    PSAPTEMP                       1048576000     128000 YES 1.0486E+10    1280000
    /oracle/HBQ/sapdata1/temp_1/temp2.data1
    PSAPTEMP                        298844160      36480 YES 3.4359E+10    4194176
    Thanks.

  • The Sky is Falling! ORA-01652: unable to extend temp segment by 128

    So we currently have a production problem and I'm not so in the know as a lowly java developer and not an Oracle expert.
    We keep getting this error(below) when a certain heavy query hits the DB.
    Our DBA claims that the tablespace for 'TABLE_SPACE_NAME_HERE' is 20GB of space and that the problem is the query.
    The query has been running fine for many many months but all of a sudden is presenting a problem and we have to do something quick.
    We tried bouncing the application server but the error came right back when the big select query gets hit.
    Any thoughts? Help! : )
    java.sql.SQLException: ORA-01652: unable to extend temp segment by 128 in tablespace TABLE_SPACE_NAME_HERE
         at oracle.jdbc.driver.DatabaseError.throwSqlException(DatabaseError.java:113)
         at oracle.jdbc.driver.T4CTTIoer.processError(T4CTTIoer.java:331)
         at oracle.jdbc.driver.T4CTTIoer.processError(T4CTTIoer.java:288)
         at oracle.jdbc.driver.T4C8Oall.receive(T4C8Oall.java:754)
         at oracle.jdbc.driver.T4CPreparedStatement.doOall8(T4CPreparedStatement.java:219)
         at oracle.jdbc.driver.T4CPreparedStatement.executeForRows(T4CPreparedStatement.java:972)
         at oracle.jdbc.driver.OracleStatement.executeMaybeDescribe(OracleStatement.java:1074)
         at oracle.jdbc.driver.T4CPreparedStatement.executeMaybeDescribe(T4CPreparedStatement.java:854)
         at oracle.jdbc.driver.OracleStatement.doExecuteWithTimeout(OracleStatement.java:1156)
         at oracle.jdbc.driver.OraclePreparedStatement.executeInternal(OraclePreparedStatement.java:3415)
         at oracle.jdbc.driver.OraclePreparedStatement.executeQuery(OraclePreparedStatement.java:3460)
         at org.jboss.resource.adapter.jdbc.WrappedPreparedStatement.executeQuery(WrappedPreparedStatement.java:296)

    LosLobo wrote:
    So, the next question... what is our lesson learned in this case?It depends on the root cause.
    Our DBA thinks 30GB is an unreasonable size for the tablespace and is fingering the select query that was causing the error to occur. Their solution is move to the query to a view and then reduce the tablespace back to 20GB.
    My thoughts are shouldn't the DB be able to handle a query that has been running fine for the last couple years? Also, if we do what is suggested what would prevent another query from coming along and causing the same issue all over again?Has the DBA identified the source of the issue? Did the query plan change? It's possible that something with statistics (or with some configuration change) causes Oracle to believe that two different query plans are roughly equally efficient. One plan might take substantially more TEMP space than another. It's possible that Oracle had been choosing the plan that involved less TEMP space being used and recently changed to preferring the plan that takes more TEMP space. If that's the issue, you may want to force Oracle to use the plan that involves less TEMP usage.
    Regardless of your TEMP tablespace size, another query may come along that causes TEMP to run out of space. Or data growth may cause TEMP to run out of space. Or an increase in the number of users may cause TEMP to run out of space. Ideally, the DBAs would be identifying how much TEMP space is used over the course of the day so that if things are growing steadily additional space can be added as necessary. If TEMP space increases dramatically because a query plan changes, however, even the best monitoring is unlikely to be able to predict that level of growth.
    Whether 30 GB is unreasonable (or whether 20 GB is unreasonable) will depend heavily on your application. We don't know enough to be able to comment. A TB-sized OLTP database serving millions of customers will have very different TEMP requirements than a multi-TB data warehouse which will have very different TEMP requirements than a small department-level application.
    My surmising is we must have just crossed a watermark threshold and the simplest most reasonable solution is to just leave the larger tablespace size.Why do you believe this is the case? It is entirely possible that you need more TEMP because your TEMP usage has been growing slowly over time. It is entirely possible that the query in question has always been using more TEMP space than it really should and that you finally have enough usage to cause the problem to bubble to the surface. It is entirely possible that the query used a reasonable amount of TEMP for the past couple years and suddenly started using far more because of a query plan change. Once you identify the source of the problem, we can figure out the appropriate solution. Without knowing the source of the problem, we're all just guessing.
    Justin

  • ORA-01652: unable to extend temp segment by 8192 in tablespace CDR_TAB

    Dear All,
    While create index command
    "create index idx_CALLED_NO on dbs(CALLED_NO) nologging"
    an error occure
    ORA-01652: unable to extend temp segment by 8192 in tablespace CDR_TAB
    how to retify it.what is the command to add datafile or tablespace......
    Using oracle 9i...
    Thanks in advance

    create tablespace or add index, that you can find out at http://tahiti.oracle.com
    You should check space on CDR_TAB.
    I think you can create new tablespace
    http://download.oracle.com/docs/cd/B10501_01/server.920/a96540/statements_74a.htm#SQLRF01403
    Create tablespace CDR_IND datafile 'xxx' ...After that create index
    create index idx_CALLED_NO on dbs(CALLED_NO) nologging tablespace CDR_IND ;Good Luck;)

  • ORA-01652: unable to extend temp segment by 256 in tablespace TEMP

    ORA-01652: unable to extend temp segment by 256 in tablespace HDM_DBUSER_TEMP
    HI GURUS PLEASE HELP ME,
    I Have 9i database on solaris version 9,
    i am repeatedly getting the following error in the alert.log file.
    Failed write to clobORA-01652: unable to extend temp segment by 256 in tablespace HDM_DBUSER_TEMP
    using the meta link Doc ID:     280578.1, i have increased the temp TS, but still getting the same alert message in alert.log file.
    In metalink it is given second solution,
    If there is severe temp segment space contention across instances, a
    slowdown can occur. The following are potential workarounds:
    - Increase size of the temp tablespace
    - Do not use the DEFAULT temp tablespace feature in RAC.
    - Increase sort_area_size and/or pga_aggregate_target "
    I have increased the size of the temp TS, but the error is poping up still in alert.log.
    I don't know how to assign DEFAULT temp tablespace feature in RAC????
    SQL> select inst_id, tablespace_name, total_blocks, used_blocks, free_blocks
    from gv$sort_segment; 2 ^R
    from gv$sort_segment;
    INST_ID TABLESPACE_NAME TOTAL_BLOCKS USED_BLOCKS FREE_BLOCKS
    1 TEMP 2304 0 2304
    1 HDM_DBUSER_TEMP 519424 519424 0
    PLEASE Help !!!
    Thanks in advance

    Hello,
    Add additional datafile to followign tablespace, also post out of the followign command
    SELECT dt.tablespace_name,
           file_name,
           bytes,
           extent_management,
           segment_space_management
    FROM dba_tablespaces dt, dba_data_files ddf
    WHERE dt.tablespace_name = ddf.tablespace_name;
    ALTER TABLESPACE hdm_dbuser_temp ADD DATAFILE '/location/...' SIZE 512M
          AUTOEXTEND ON NEXT 256M MAXSIZE 4096M;   -- or unlimited on 8192 MRegards

  • ORA-01652: unable to extend temp segment by 9600 in tablespace TEMP

    im just trying to execute a query with an analytic function used and this error occurs : ORA-01652: unable to extend temp segment by 9600 in tablespace TEMP
    select substr(b.eff_dt,1,10) "Effective",b.amt,sum(b.amt) over (order by substr(b.eff_dt,1,10)) "running_sum"
    from b_dl b, b_dl b2
    where b.st_cd = '1'
    and ( b.b_key is null or b.b_key not in(39,40,55,31,32,33,34,35,36,37,38,39,40,55,43,44))
    and b.amt <> 0
    please help

    If you are not a DBA, then get in touch with your Database administrator and show him this problem. Your database is running out of TEMP tablespace because of the sorting needed by this query.
    BTW, How many rows are being processed in this self join query?

  • ORA-01652: unable to extend temp segment by 64 in tablespace

    Hi,
    I am getting error ORA-01652: unable to extend temp segment by 64 in tablespace <Tablespace_Name>. I have checked the temp tablespace and the space of temp table space is sufficiently free.
    Can any one pls suggest on this.
    Thanks,

    Hi;
    Please see below note which could be helpful for your issue:
    TROUBLESHOOTING GUIDE (TSG) - UNABLE TO EXTEND Errors [ID 1025288.6]
    Regard
    Helios

  • ORA-01652: unable to extend temp segment by 1280 in tablespace TEMP

    We have a lot of version 8.5 crystal report templates, now, as required, we want to use JRC (pure java) to export those templates to PDF / RTF etc.
    We have done a lot, for the most of templates, works fine, but there are some reports, once the data is large, an error coming :ORA-01652: unable to extend temp segment by 1280 in tablespace TEMP
    for this problem:
    1.  I have read the migration guide first :http://support.businessobjects.com/communityCS/TechnicalPapers/cr_xi_designer_migration.pdf
    2. I have tested the templates with the same parameters in designer, they are just fine ( and obviously, the old templates can work with COM components in wintel platform , we used VB )
    3. Because 2 works fine, I tried to extend our oracle temporary tablespace, but everytime I extend, I can see it will always be fully occupied by the JRC export function.
    4. The problem is occured in the  repSrc.export(expOpt, reqCont) function
    I am now trying to find out the reasons:
    1. checking the parameters real value in the report
    2. enable log4j for JRC to see the real SQL statement
    This is really annoying problem, I would appreciate if anyone could give me help.

    I think this is one might have some relationship with my problem
    http://diamond.businessobjects.com/node/477
    I found in version 8.5, the created SQL Statement is simple, all conditions come under Where clause, but the new crystal report XI convert this to "join" types.
    But JRC engine will lost some very important join condition to make the result really large, so, my oracle error comes out.
    I am thinking about to write command object as a workaround for this issue, woundn't JRC engine support the simple SQL Statement?
    I found the release note of CR4E says some features are "Supported" instead of "Full Support", where can I find the detail list of these un-supported feature list?

  • ORA-1652: unable to extend temp segment by 128 in tablespace

    HI,
    i am getting an error in alert log file.....
    Thread 1 advanced to log sequence 1758
      Current log# 2 seq# 1758 mem# 0: /dev/vx/rdsk/racdg/orcl_raw_log12
    Mon Sep  8 12:34:16 2008
    ARC1: Evaluating archive   log 1 thread 1 sequence 1757
    ARC1: Beginning to archive log 1 thread 1 sequence 1757
    Creating archive destination LOG_ARCHIVE_DEST_1: '/arch/log/1_1757.dbf'
    ARC1: Completed archiving  log 1 thread 1 sequence 1757
    Mon Sep  8 13:04:26 2008
    Completed checkpoint up to RBA [0x6de.2.10], SCN: 0x0000.6c1f757f
    Mon Sep  8 13:49:16 2008
    ORA-1652: unable to extend temp segment by 128 in tablespace                 TEMPI don';t want to add datafile
    SQL> select bytes,maxbytes,increment_by from dba_temp_files;
         BYTES   MAXBYTES INCREMENT_BY
    6134169600          0            0
    SQL> select TABLESPACE_NAME, BYTES_USED, BYTES_FREE from V$TEMP_SPACE_HEADER;
    TABLESPACE_NAME                BYTES_USED BYTES_FREE
    TEMP                           6037700608   96468992i have also referred metalink note: 19047.1
    Is there any way to avoid this problem without adding any datafile........

    how to find out which session is using temp tablespace
    SQL> select * from v$sort_usage;
    USERNAME                       USER                           SESSION_ADDR
    SESSION_NUM SQLADDR             SQLHASH TABLESPACE
    CONTENTS  SEGTYPE     SEGFILE#    SEGBLK#    EXTENTS     BLOCKS   SEGRFNO#
    DATA3                        DATA3                        000000044D84F558
          25860 00                        0 TEMP
    TEMPORARY DATA             201     183561          1        128          1pls correct me if the above query is worng to find out the no. of session s using temp tablespace
    i think we have already allocated good amt of space to pga_aggregate_target
    SQL> show parameter pga
    NAME                                 TYPE        VALUE
    pga_aggregate_target                 big integer 524288000
    SQL>Edited by: user00726 on Sep 8, 2008 2:29 AM

  • ORA-1652: unable to extend temp segment by 128 in tablespace   TEMP

    Hi All,
    Today in we have faced the below issue in our production database
    ORA-1652: unable to extend temp segment by 128 in tablespace                 TEMP
    ORA-1652: unable to extend temp segment by 128 in tablespace                 TEMP
    when we checked we are having 99% of free space in temp tablespace,we are not able to understand why we got this error.
    earlier also we got the same issue adding datafile to the tablespace solved the issue. wanted to know if free space exsists why we getting this error.
    Thanks....

    Hi,
    Please check Oracle support id : TROUBLESHOOTING GUIDE (TSG) : ORA-1652: unable to extend temp segment (Doc ID 1267351.1)
    and OTN Form :
    https://forums.oracle.com/thread/457771?start=0&tstart=0
    https://forums.oracle.com/thread/703095?start=0&tstart=0
    Thank you

  • ORA-1652: unable to extend temp segment by 699912 in tablespace ##PSAPTEMP

    Dear Expert
    I found this error ORA-1652: unable to extend temp segment by 699912 in tablespace ##PSAPTEMP. How can I fix this problem. Anyone help me...pls
    best regards

    Hi,
    We had the same error in our Production system.
    That basically occured when there were a lot of Background jobs that were running.
    As more than 5 jobs were running which were utisiling PSAPTEMP to the fullest invariably the jobs got cancelled with a dump and the error u have mentioned
    Though we tried adding 3 datafiles the error kept repeating..
    We finally added almost 20 GB to PSAPTEMP for  those jobs to get completed.

  • ORA-1652: unable to extend temp segment by 128 in tablespace TEMP1/TEMP2

    db version: 11.1.0.7
    I detect the error log on dbTier's alertlog as below, BUT I think the temp tbs is enough because of the SQL output(you can see as below too)
    so I want to know why and how can I fix it. thanks
    ORA-1652: unable to extend temp segment by 128 in tablespace TEMP1
    ORA-1652: unable to extend temp segment by 128 in tablespace TEMP2
    SQL> select TABLESPACE_NAME,TOTAL_BLOCKS,USED_BLOCKS,FREE_BLOCKS from v$sort_segment;
    TABLESPACE_NAME-------TOTAL_BLOCKS-----------USED_BLOCKS--------------FREE_BLOCKS
    TEMP1 -------------1048448------------9472--------1038976
    TEMP2---------------1048448 -----------5248---------1043200
    TEMP--------0--------0-------------0

    Unless you are running the query against v$sort_segment at the point in time that you are getting the errors, I would strongly suspect that something tried to use more than 16 GB of TEMP space, the query died, and when you subsequently looked, nothing was attempting to use much TEMP space. That doesn't mean that some process wasn't eating up all your TEMP space earlier.
    My guess (no evidence, just a hunch) is that you have a query that is doing a Cartesian join, that query rapidly consumed all the space in the TEMP1 and TEMP2 tablespaces, encountered this error before exiting and freeing all the TEMP1 and TEMP2 space it had acquired before you ran your query.
    Justin

Maybe you are looking for