End of file on communication channel when doing a (hash) join

I'm having a problem with a simple join between two tables (hash join, from the explain), one containing about 650k rows, the other 730k.
The query runs fine until I add to the select list a geometry field (oracle spatial). In that case I receive an end-of-file error, which seems to be caused by a memory saturation. I'm not sure about this, as I can't understand the trace very well...
I've increased the pga_aggregate_target and the pgamax_size but this have little effects on the problem...
Any hint? Could it be caused by the geometry field size?
thanks,
giovanni

Thanks for the quick reply. Here it is what you asked:
select from v$version;*
Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - Prod
PL/SQL Release 10.2.0.4.0 - Production
"CORE     10.2.0.4.0     Production"
TNS for 32-bit Windows: Version 10.2.0.4.0 - Production
NLSRTL Version 10.2.0.4.0 - Production
an excerpt from the trace file:
*** ACTION NAME:() 2010-04-16 12:48:17.796
*** MODULE NAME:(SQL*Plus) 2010-04-16 12:48:17.796
*** SERVICE NAME:(ABRUZZO) 2010-04-16 12:48:17.796
*** SESSION ID:(149.10) 2010-04-16 12:48:17.796
*** 2010-04-16 12:48:17.796
ksedmp: internal or fatal error
ORA-07445: trovata eccezione: dump della memoria [ACCESS_VIOLATION] [_kokekd2m+24] [PC:0x11CF9E0] [ADDR:0xBA8C193C] [UNABLE_TO_READ] []
Current SQL statement for this session:
select g.cr375_idobj,g.cr375_geometria,t.cr374_codicectr,t.cr374_scarpt_cont
from DBTI.Cr374_Scarpata t, DBTI.Cr375g_a_Scarp g
where t.cr374_idobj=g.cr375_idobj
check trace file c:\oracle\product\10.2.0\db_1\rdbms\trace\abruzzo_ora_0.trc for preloading .sym file messages
----- Call Stack Trace -----
calling call entry argument values in hex
location type point (? means dubious value)
_kokekd2m+24                  00000000
kokeq2iro+178       CALLrel kokekd2m+0 99384C8 9937624 BA8C18D8
9968588 3660D954 3A52AC00
__VInfreq__rworupo+ CALLrel _kokeq2iro+0         9968218 A05C810 EA7
321
_kxhrUnpack+71       CALLreg 00000000             99693C8 A05C810 0 9 0
qerhjWalkHashBucke CALLrel kxhrUnpack+0 9967DD0 99693C8 9 A05C810 0
t+210
__PGOSF352__qerhjIn CALLrel _qerhjWalkHashBucke A05CED4 99693C0 8
nerProbeHashTable+4 t+0
78
_kdstf0000101km+230 CALLreg 00000000
kdsttgr+1263        CALLrel kdstf0000101km+0

Similar Messages

"ORA-03113: end-of-file on communication channel" when a merge is called

We have two tables, one for the input of data and the other a production version that means the data has been approved. When the user approves the data, we use the merge command to update and insert the approved data into the production table. This worked great until now. We added a domain index to the production table. Since then, when I use the merge command, I get the ORA-03113 error. We are using Oracle 11g, but I thought that you could use merge with domain indexes in 11g. The merge does not affect the actual index column. It only affects certain columns so we do the update/insert just on those. Any idea? Do I need to do away with the merge for this table or is there another issue?
Thank you,
Susan

Hi,
Did you check if there is an error in the alert log? Can you post the error if exist?
What is the 11g version?
There are some bugs fixed in 11.1.0.7
Domain Indexes
5891213      ORA-6502 / LPX-7 from DBMS_METADATA.GET_DDL with domain index present
5908945      Dump [qxopqdca] with ancilliary operator
6051635      Wrong DDL from DBMS_METADATA for domain index or index on XMLType column
6497310      ORA-942 / dump for DML in XMLTYPE column with domain index
6685261      Dump [lnxmin] using ancilliary operators with inline views
So either see if you can patch or test to replace merge with insert/update or drop the index?
Regards,

Convrtd to Invterval Part- ORA-03113: end-of-file on communication channel

Hi all,
I had a table as Interval Partitioned. In order to create XML- Xpath indexes on it, I converted it to Range Partitioned table.
I am able to create the XPATH indexes but I get the error: ORA-03113: end-of-file on communication channel
- When I revert the code to Interval Partitioned without the XMLIndex, it works fine(although takes time as no XML Index)
- When I convert table to non partitioned table, create the XML Index, it works fine.
But I need the partitons, so when I create the partitioned table I get the error.
CREATE TABLE INT_PART_TABLE
DB_ID VARCHAR2(10 BYTE),
xML_mESSAGE SYS.XMLTYPE,
LOAD_TIMESTAMP TIMESTAMP(6)
XMLTYPE xML_mESSAGE STORE AS BINARY XML
PARTITION BY RANGE (LOAD_TIMESTAMP)
PARTITION MAX VALUES LESS THAN (TIMESTAMP' 2013-06-01 00:00:00')
TABLESPACE CSTR_STG_DATA
NOCOMPRESS
NOCACHE
ENABLE ROW MOVEMENT;
BEGIN
DBMS_XMLINDEX.dropparameter('Indx_Par');
END;
BEGIN
DBMS_XMLINDEX.REGISTERPARAMETER(
'Indx_Par',
'PATH TABLE Table1
PATHS (INCLUDE ( /abc:field1/xyz:field2
/abc:field1/def:field2
NAMESPACE MAPPING ( xmlns:abc="ABCD"
xmlns:def="DEFG"
xmlns:xyz="XYZA"
end;
create index INDX_XPATHS on "INT_PART_TABLE" (XML_MESSAGE) indextype is xdb.xmlindex
parameters ('PARAM Indx_Par') local;
Now if I execute the following statement in
SELECT T.xML_mESSAGE
FROM INT_PART_TABLE1 T
WHERE XMLEXISTS (
declare namespace abc="ABCD";
declare namespacedef="DEFG";
declare namespace xyz="XYZA";
let $tt as xs:boolean := fn:exists($p/main/id = ("144283","9085802")])
return if ($tt) then true()
else ()'
PASSING T.xML_mESSAGE AS "p");
- Is there any other way of writing this Select statement, which may work?
- Any other thing I need to take care of when defining the table and partitions script so that I don't get this error?

Hi,
I think it's time you give a clear (and working) test case so that we can safely try to reproduce the issue.
What you've given so far has syntax error and name mismatch.
So please :
- database version (SELECT * FROM v$version)
- complete sequence of DLLs
- some sample XML documents (it doesn't have to be the real ones, but at least something realistic)
Thanks in advance.
declare namespace abc="ABCD";
declare namespacedef="DEFG";
declare namespace xyz="XYZA";
let $tt as xs:boolean := fn:exists($p/main/id = ("144283","9085802")])
return if ($tt) then true()
else ()'Why all that stuff? You don't have to return a boolean.
The following works for me on 11.2.0.3 :
SQL> CREATE TABLE int_part_table (
2    db_id          VARCHAR2(10)
3 , xml_message    XMLTYPE
4 , load_timestamp TIMESTAMP
5 )
6 XMLTYPE xml_message STORE AS BINARY XML
7 PARTITION BY RANGE (load_timestamp) (
8    PARTITION MAX VALUES LESS THAN (timestamp '2013-06-01 00:00:00')
9 )
10 NOCOMPRESS
11 NOCACHE
12 ENABLE ROW MOVEMENT;
Table created
SQL> insert into int_part_table values (1, xmltype('<main><id>144283</id></main>'), sysdate);
1 row inserted
SQL> insert into int_part_table values (1, xmltype('<main><id>9085802</id></main>'), sysdate);
1 row inserted
SQL> insert into int_part_table values (1, xmltype('<main><id>1</id></main>'), sysdate);
1 row inserted
SQL> commit;
Commit complete
SQL> create index int_part_table_uix on int_part_table (xml_message)
2 indextype is xdb.xmlindex
3 parameters (
4 'PATH TABLE INT_PART_TABLE_PT
5 PATHS ( INCLUDE ( /main/id ) )')
6 local;
Index created
SQL> SELECT xml_message
2 FROM int_part_table
3 WHERE XMLExists(
4         '/main[id=("144283","9085802")]'
5         PASSING xml_message
6       )
7 ;
XML_MESSAGE
<main>
<id>144283</id>
</main>
<main>
<id>9085802</id>
</main>
Execution Plan
Plan hash value: 3517234298
| Id | Operation                              | Name                         | Rows | Bytes | Cost (%CPU)| Time     | Pstart| Pstop |
|   0 | SELECT STATEMENT                       |                              |     1 |   155 |    34   (6)| 00:00:01 |       |       |
|   1 | NESTED LOOPS                          |                              |     1 |   155 |    34   (6)| 00:00:01 |       |       |
|   2 |   VIEW                                 | VW_SQ_1                      |     1 |    25 |    32   (4)| 00:00:01 |       |       |
|   3 |    HASH UNIQUE                         |                              |     1 |    47 |         |             |       |       |
|* 4 |     HASH JOIN SEMI                     |                              |     1 |    47 |    32   (4)| 00:00:01 |       |       |
|   5 |      PARTITION SYSTEM SINGLE           |                              |     2 |    90 |     2   (0)| 00:00:01 |     1 |     1 |
|* 6 |       TABLE ACCESS BY LOCAL INDEX ROWID| INT_PART_TABLE_PT            |     2 |    90 |     2   (0)| 00:00:01 |     1 |     1 |
|* 7 |        INDEX SKIP SCAN                 | SYS117585_INT_PART__PIKEY_IX |     3 |       |     1   (0)| 00:00:01 |     1 |     1 |
|   8 |      COLLECTION ITERATOR PICKLER FETCH | XQSEQUENCEFROMXMLTYPE        | 8168 | 16336 |    29   (0)| 00:00:01 |       |       |
|* 9 |   TABLE ACCESS BY USER ROWID           | INT_PART_TABLE               |     1 |   130 |     1   (0)| 00:00:01 | ROWID | ROWID |
Predicate Information (identified by operation id):
   4 - access("SYS_P3"."VALUE"=SYS_XQ_UPKXML2SQL(VALUE(KOKBF$),2,1,0) AND
              SUBSTRB("VALUE",1,1599)=SUBSTRB(SYS_XQ_UPKXML2SQL(VALUE(KOKBF$),2,1,0),1,1599))
   6 - filter(SYS_XMLI_LOC_ISNODE("SYS_P3"."LOCATOR")=1)
   7 - access("SYS_P3"."PATHID"=HEXTORAW('704E') )
       filter("SYS_P3"."PATHID"=HEXTORAW('704E') )
   9 - filter("ITEM_6"=TBL$OR$IDX$PART$NUM("DEV"."INT_PART_TABLE",0,7,65535,"INT_PART_TABLE".ROWID))
Note
   - Unoptimized XML construct detected (enable XMLOptimizationCheck for more information)
SQL> SELECT xml_message
2 FROM int_part_table
3 WHERE XMLExists(
4         '/main[id="144283" or id="9085802"]'
5         PASSING xml_message
6       )
7 ;
XML_MESSAGE
<main>
<id>144283</id>
</main>
<main>
<id>9085802</id>
</main>
Execution Plan
Plan hash value: 3748936130
| Id | Operation                                | Name                         | Rows | Bytes | Cost (%CPU)| Time     | Pstart| Pstop |
|   0 | SELECT STATEMENT                         |                              |     1 |   155 |    11 (10)| 00:00:01 |       |       |
|   1 | NESTED LOOPS                            |                              |     1 |   155 |    11 (10)| 00:00:01 |       |       |
|   2 |   VIEW                                   | VW_SQ_1                      |     2 |    50 |     8   (0)| 00:00:01 |       |       |
|   3 |    HASH UNIQUE                           |                              |     2 |   180 |         |             |       |       |
|   4 |     CONCATENATION                        |                              |       |       |         |             |       |       |
|   5 |      NESTED LOOPS                        |                              |       |       |         |             |       |       |
|   6 |       NESTED LOOPS                       |                              |     1 |    90 |     4   (0)| 00:00:01 |       |       |
|   7 |        PARTITION SYSTEM SINGLE           |                              |     1 |    45 |     2   (0)| 00:00:01 |     1 |     1 |
|* 8 |         TABLE ACCESS BY LOCAL INDEX ROWID| INT_PART_TABLE_PT            |     1 |    45 |     2   (0)| 00:00:01 |     1 |     1 |
|* 9 |          INDEX SKIP SCAN                 | SYS117585_INT_PART__PIKEY_IX |     3 |       |     1   (0)| 00:00:01 |     1 |     1 |
| 10 |        PARTITION SYSTEM SINGLE           |                              |     1 |       |     1   (0)| 00:00:01 |     1 |     1 |
|* 11 |         INDEX RANGE SCAN                 | SYS117585_INT_PART__PIKEY_IX |     1 |       |     1   (0)| 00:00:01 |     1 |     1 |
|* 12 |       TABLE ACCESS BY LOCAL INDEX ROWID | INT_PART_TABLE_PT            |     1 |    45 |     2   (0)| 00:00:01 |     1 |     1 |
| 13 |      NESTED LOOPS                        |                              |       |       |         |             |       |       |
| 14 |       NESTED LOOPS                       |                              |     1 |    90 |     4   (0)| 00:00:01 |       |       |
| 15 |        PARTITION SYSTEM SINGLE           |                              |     1 |    45 |     2   (0)| 00:00:01 |     1 |     1 |
|* 16 |         TABLE ACCESS BY LOCAL INDEX ROWID| INT_PART_TABLE_PT            |     1 |    45 |     2   (0)| 00:00:01 |     1 |     1 |
|* 17 |          INDEX SKIP SCAN                 | SYS117585_INT_PART__PIKEY_IX |     3 |       |     1   (0)| 00:00:01 |     1 |     1 |
| 18 |        PARTITION SYSTEM SINGLE           |                              |     1 |       |     1   (0)| 00:00:01 |     1 |     1 |
|* 19 |         INDEX RANGE SCAN                 | SYS117585_INT_PART__PIKEY_IX |     1 |       |     1   (0)| 00:00:01 |     1 |     1 |
|* 20 |       TABLE ACCESS BY LOCAL INDEX ROWID | INT_PART_TABLE_PT            |     1 |    45 |     2   (0)| 00:00:01 |     1 |     1 |
|* 21 |   TABLE ACCESS BY USER ROWID             | INT_PART_TABLE               |     1 |   130 |     1   (0)| 00:00:01 | ROWID | ROWID |
Predicate Information (identified by operation id):
   8 - filter("SYS_P5"."VALUE"='9085802' AND SYS_XMLI_LOC_ISNODE("SYS_P5"."LOCATOR")=1 AND SUBSTRB("VALUE",1,1599)='9085802')
   9 - access("SYS_P5"."PATHID"=HEXTORAW('704E') )
       filter("SYS_P5"."PATHID"=HEXTORAW('704E') )
11 - access("SYS_P5"."RID"="SYS_P3"."RID" AND "SYS_P3"."PATHID"=HEXTORAW('0BBD') AND
              "SYS_P3"."ORDER_KEY"<"SYS_P5"."ORDER_KEY")
       filter(SYS_ORDERKEY_DEPTH("SYS_P3"."ORDER_KEY")+1=SYS_ORDERKEY_DEPTH("SYS_P5"."ORDER_KEY") AND
              TBL$OR$IDX$PART$NUM("DEV"."INT_PART_TABLE",0,7,65535,"SYS_P3"."RID")=TBL$OR$IDX$PART$NUM("DEV"."INT_PART_TABLE_PT",0,7,65535,ROWI
              D) AND "SYS_P5"."ORDER_KEY"<SYS_ORDERKEY_MAXCHILD("SYS_P3"."ORDER_KEY"))
12 - filter(SYS_XMLI_LOC_ISNODE("SYS_P3"."LOCATOR")=1)
16 - filter("SYS_P5"."VALUE"='144283' AND SYS_XMLI_LOC_ISNODE("SYS_P5"."LOCATOR")=1 AND SUBSTRB("VALUE",1,1599)='144283' AND
              (LNNVL("SYS_P5"."VALUE"='9085802') OR LNNVL("SYS_P5"."PATHID"=HEXTORAW('704E') ) OR
              LNNVL(SYS_XMLI_LOC_ISNODE("SYS_P5"."LOCATOR")=1) OR LNNVL(SUBSTRB("VALUE",1,1599)='9085802')))
17 - access("SYS_P5"."PATHID"=HEXTORAW('704E') )
       filter("SYS_P5"."PATHID"=HEXTORAW('704E') )
19 - access("SYS_P5"."RID"="SYS_P3"."RID" AND "SYS_P3"."PATHID"=HEXTORAW('0BBD') AND
              "SYS_P3"."ORDER_KEY"<"SYS_P5"."ORDER_KEY")
       filter(SYS_ORDERKEY_DEPTH("SYS_P3"."ORDER_KEY")+1=SYS_ORDERKEY_DEPTH("SYS_P5"."ORDER_KEY") AND
              TBL$OR$IDX$PART$NUM("DEV"."INT_PART_TABLE",0,7,65535,"SYS_P3"."RID")=TBL$OR$IDX$PART$NUM("DEV"."INT_PART_TABLE_PT",0,7,65535,ROWI
              D) AND "SYS_P5"."ORDER_KEY"<SYS_ORDERKEY_MAXCHILD("SYS_P3"."ORDER_KEY"))
20 - filter(SYS_XMLI_LOC_ISNODE("SYS_P3"."LOCATOR")=1)
21 - filter("ITEM_2"=TBL$OR$IDX$PART$NUM("DEV"."INT_PART_TABLE",0,7,65535,"INT_PART_TABLE".ROWID))I asked in one of your other threads if /main/id was unique per XML document.
If so, you can use a simple function-based index instead of the XMLIndex :
SQL> drop index int_part_table_uix;
Index dropped.
SQL> create index int_part_table_ix1 on int_part_table (
2    xmlcast(
3      xmlquery('/main/id' passing XML_MESSAGE returning content)
4      as varchar2(10)
5    )
6 );
Index created.
SQL> SELECT xml_message
2 FROM int_part_table
3 WHERE XMLCast(
4          XMLQuery('/main/id' PASSING xml_message RETURNING CONTENT)
5          AS VARCHAR2(10)
6        )
7 IN ('144283', '9085802');
XML_MESSAGE
<main>
<id>144283</id>
</main>
<main>
<id>9085802</id>
</main>
Execution Plan
Plan hash value: 2864653096
| Id | Operation                           | Name               | Rows | Bytes | Cost (%CPU)| Time | Pstart| Pstop |
|   0 | SELECT STATEMENT                    |                    |     2 |   236 |     2   (0)| 00:00:01 |       |       |
|   1 | INLIST ITERATOR                    |                    |       |       |            |       | |       |
|   2 |   TABLE ACCESS BY GLOBAL INDEX ROWID| INT_PART_TABLE     |     2 |   236 |     2   (0)| 00:00:01 |     1 |     1 |
|* 3 |    INDEX RANGE SCAN                 | INT_PART_TABLE_IX1 |     2 |       |     1   (0)| 00:00:01 |       |       |
Predicate Information (identified by operation id):
   3 - access(CAST(EXTRACTVALUE(SYS_MAKEXML(0,"SYS_NC00003$"),'/main/id',null,0,0,524293,1073874944) AS
              varchar2(10)   )='144283' OR CAST(EXTRACTVALUE(SYS_MAKEXML(0,"SYS_NC00003$"),'/main/id',null,0,0,524293,1073874944
              ) AS varchar2(10)   )='9085802')

Oracle error ORA-03113: end-of-file on communication channel

Dear Support
I am getting "ORA-03113: end-of-file on communication channel" when I am trying to access my application.
The current setup is following
- two node Sun Cluster with node name as sep1n1 and sep2n2
- running Oracle RAC (oracle database 10g) with instance name as ADVFRW1 and ADVFRW2
Please see attached oracle trace file which was generated at the time of problem.
Can you please help me to know what can be the reason of this oracle error/crash?
Thanks
Regards
VS

I don't see any Trace file.
Network Problems?
Are the databases up and running?

ORA-03113: end-of-file on communication channel ; On All Indexed fields

Hi Friends,
Please help
I am getting the following error,when querying a table in 10.2.0.4.0.
ERROR at line 1:
ORA-03113: end-of-file on communication channel
When I exclude one of the indexed fields in the where condition it works fine.I rebuilded/dropped index,But still oracle throws me out.
Please see some Infom from Alter.log and Trace files
Alter log
Errors in file d:\oracle\product\10.2.0\admin\cvs2\udump\cvs2_ora_3064.trc:
ORA-07445: exception encountered: core dump ACCESS_VIOLATION __VInfreq__qksqbFind1RowTabs+72 PC:0x313865C ADDR:0x131 UNABLE_TO_READ] [
Trace file
Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
Windows XP Version V5.1 Service Pack 3
CPU : 2 - type 586, 2 Physical Cores
Process Affinity : 0x00000000
Memory (Avail/Total): Ph:829M/2038M, Ph+PgF:2573M/3933M, VA:964M/2047M
Instance name: cvs2
Redo thread mounted by this instance: 1
Oracle process number: 28
Windows thread id: 3064, image: ORACLE.EXE (SHAD)
*** ACTION NAME:() 2009-06-09 16:50:51.357
*** MODULE NAME:(SQL*Plus) 2009-06-09 16:50:51.357
*** SERVICE NAME:(CVS2) 2009-06-09 16:50:51.357
*** SESSION ID:(148.7775) 2009-06-09 16:50:51.357
*** 2009-06-09 16:50:51.357
ksedmp: internal or fatal error
ORA-07445: exception encountered: core dump ACCESS_VIOLATION __VInfreq__qksqbFind1RowTabs+72 PC:0x313865C ADDR:0x131 UNABLE_TO_READ] [
Current SQL statement for this session:
Call Stack Trace calling call entry argument values in hex
location type point (? means dubious value) -------- ----------------------------
__VInfreq__qksqbFin 00000000
d1RowTabs+72
kkogtp+366 CALLrel qksqbFind1RowTabs+
0
vopastp+276 CALLrel kkogtp+0 9C17008 9C17008 9C16808 0
voppfdDescendents+ CALLrel vopastp+0 9C16808 1 0
374
voppfd+88 CALLrel voppfdDescendents+ 9C16808 1
0
opitca+854 CALLrel voppfd+0 9C16808 1
__PGOSF346__kksFull CALLrel _opitca+0 9ECB128 3D2396F0
TypeCheck+15
_rpiswu2+426 CALLreg 00000000 BF9C004
kksLoadChild+8074 CALLrel rpiswu2+0 457453CC 9D 3DE12CF8 5
3DE12A00 9D 3DE12D24 0 5CDB50
0 BF9C004 0
kxsGetRuntimeLock+ CALLrel kksLoadChild+0 C669558 3D738E24 BF9CA5C
and it goes and goes.....
Thanks In Advance
SSN

I'm not 100% sure, but I'd suggest you to open SR in metalink. And send these trace files them for analyze
Kamran Agayev A. (10g OCP)
http://kamranagayev.wordpress.com
[Step by Step install Oracle on Linux and Automate the installation using Shell Script |http://kamranagayev.wordpress.com/2009/05/01/step-by-step-installing-oracle-database-10g-release-2-on-linux-centos-and-automate-the-installation-using-linux-shell-script/]

End-of-file on communication channel error when starting database

Hi All,
I have small problem, when we specify at startup without any pfile, it will go to read spfile,in our case it gives error end-of-file on communication channel.
Can anyone please help me to resolve this issue?
Thank you
Chirag

Oracle unable to spawn job slave processes should not prevent the instance from starting - these are not critical system processes and the instance can function without them (as it can when the job queue is set to zero processes).
The o/s failure message also does not make much sense.. overlapped I/O is Microsoft Windows term for async I/O. This can be used at file I/O level, but is also supported for any other file handle like sockets - thus applicable at the network layer too.
I'm a bit confused to how a PFILE versus a SPFILE startup can cause this. If that is the case, then surely there are something different ito init instance settings that triggers this error?
Can you copy and paste the actual SQL*Plus session (as already requested). Show us startup command (and params) used in SQL*Plus and the resulting messages (and errors displayed) in response to the startup.

EAL extract fails with ORA-03113 end-of-file on communication channel error

Hi,
We are trying to run an EAL (v11.1.1.4) extract from an HFM (v9.3.1.4) environment onto Oracle database. We are able to run the extract successfully for few hundred records, but when the records exceeds 10 million we get the error below
[03 Dec 2011 06:02:41] [dbmgr] ERROR: OCI: HR#13579: Cannot execute statement for insert
[03 Dec 2011 06:02:41] [dbmgr] ERROR: OCI: HR#01428: OCI_ERROR: (ORA-03113: end-of-file on communication channel;Process ID: 16323;Session ID: 15 Serial number: 8288; (code=3113)) [rec#1]
[03 Dec 2011 06:02:41] [dbmgr] ERROR: OCI: HR#13580: Cannot rollback
[03 Dec 2011 06:02:41] [dbmgr] ERROR: OCI: HR#01428: OCI_ERROR: (ORA-03114: not connected to ORACLE; (code=3114)) [rec#1]
[03 Dec 2011 06:02:41] [dbmgr] WARN : HR#13674: Data move region job failed: HR#00000: some operations failed, see details in the log file; 81877900 cell(s) processed in 70348 second(s)
We checked on the DB and the Oracle session that was doing the insert was open even after the extract failed with the above error.
Any idea what could be causing this??
Thanks
Ramya Suresh

Hi,
I was able to fix this issue.
From the trace file identified the related objects and validated them.
More details can be found in below MOS note.
[ID 421711.1]--Understanding and Diagnosing ORA-00600 [17069] Errors
Thanks for all the inputs...

Getting end-of-file on communication channel error on this query

Hi guys,
I'm getting the end-of-file on communcation channel error when running the following query in SQL PLUS, If run it in a stored procedure called through code of JDBC, it gave "no more data to read from socket" error. Any idea what went wrong? kind of frustrated now. We are using the Oralce 9.0.1.4.0, which suppose to fix some sdo_uion and sdo_buffer bugs. Thanks a lot!!
select SDO_GEOM.SDO_BUFFER((SELECT SDO_AGGR_UNION(MDSYS.SDOAGGRTYPE(GEOLOC, 0.011119487)) FROM GEOTEL_SOURCE where MSA='5000' and COMPANY_NAME ='AMERICAN FIBER SYSTEMS'),1.0,5,'unit=MILE arc_tolerance=0.05') from GEOTEL_SOURCE a where a.MI_PRINX = 1;

Hi Daniel,
Just tried the 9.2.0.2 patch, seems to work better now. At least I don't get the end-of-file on communication channel error. But how about the speed thing? I isolated the problem, and find out that sdo_aggr_union is taking too long to finish( 1 Hour!!! for 2300 lines with the same city. Any clue to optimize it, I tried some hint like /*+ordered*/ No_Merge and something like that, does not improve though. Thanks!

8.1.7 Post-installation end-of-file on communication channel error

Hi everyone,
I got a problem with a new install of Oracle 8.1.7. First, I had 8.0.5 installed on Linux which contained a relative huge database. I wanted to upgrade to 8.1.7 and so, I installed it (with no default database). After I applied the patch (the setup_stubs.sh script) I tried to migrate the existing database with the odma (oracle data migration assistant). However, during the process odma gave me an error "An error occured while starting up the database..." I checked the log that came with the error message and it occured when it was trying to open the database i.e. the instance started, but when it came for mounting the database it displayed: ORA-03113: end-of-file on communication channel
I attempted to do the migration manually. So I tried to start the database using svrmgrl:
startup restrict pfile=$ORACLE_HOME/dbs/initSID.ora
and, of course, I get the instance started, but when it comes to mount the db, it gave me:
ORA-03113: end-of-file on communication channel
Anybody knows why I get this error on startup?
I checked my environment variables, my PATH and also made sure that svrmgrl was the version for 8.1.7. Everythings seem fine.
Any ideas?
Anything would be very much appreciated.
Thanks
joss

Oracle unable to spawn job slave processes should not prevent the instance from starting - these are not critical system processes and the instance can function without them (as it can when the job queue is set to zero processes).
The o/s failure message also does not make much sense.. overlapped I/O is Microsoft Windows term for async I/O. This can be used at file I/O level, but is also supported for any other file handle like sockets - thus applicable at the network layer too.
I'm a bit confused to how a PFILE versus a SPFILE startup can cause this. If that is the case, then surely there are something different ito init instance settings that triggers this error?
Can you copy and paste the actual SQL*Plus session (as already requested). Show us startup command (and params) used in SQL*Plus and the resulting messages (and errors displayed) in response to the startup.

ORA-03113: end-of-file on communication channel in Toad

Hi,
When i am running any commands from SQL Plus i am not getting the Error while logged in a s sys as sysdba
But the same command while running in Toad using sys as sysdba I am getting the Error as :ORA-03113: end-of-file on communication channel.
The Database does not occured any Error during Start up its mounted and opened
Kindly any help will be needful for me.
Thanks and Regards

Hi,
I checked the Alert log file in the bdump directory and its giving as:
Hi,
I checked the Alert log file in the bdump directory and its giving as:
Sun Mar 7 13:43:31 2010
Setting recovery target incarnation to 1
Sun Mar 7 13:43:31 2010
Successful mount of redo thread 1, with mount id 3004982143
Sun Mar 7 13:43:31 2010
Database mounted in Exclusive Mode
Completed: ALTER DATABASE MOUNT
Sun Mar 7 13:43:31 2010
ALTER DATABASE OPEN
Sun Mar 7 13:43:31 2010
Thread 1 opened at log sequence 199
Current log# 1 seq# 199 mem# 0: /oracle/oradata/testdb/redo01.log
Successful open of redo thread 1
Sun Mar 7 13:43:31 2010
MTTR advisory is disabled because FAST_START_MTTR_TARGET is not set
Sun Mar 7 13:43:31 2010
SMON: enabling cache recovery
Sun Mar 7 13:43:31 2010
Successfully onlined Undo Tablespace 1.
Sun Mar 7 13:43:31 2010
SMON: enabling tx recovery
Hex dump of (file 1, block 52484) in trace file /oracle/admin/testdb/bdump/testdb_smon_3460.trc
Corrupt block relative dba: 0x0040cd04 (file 1, block 52484)
Bad check value found during buffer read
Data in bad block:
type: 6 format: 2 rdba: 0x0040cd04
last change scn: 0x0000.0012d539 seq: 0x1 flg: 0x06
spare1: 0x0 spare2: 0x0 spare3: 0x0
consistency value in tail: 0xd5390601
check value in block header: 0x62e3
computed block checksum: 0x600
Reread of rdba: 0x0040cd04 (file 1, block 52484) found same corrupted data
Sun Mar 7 13:43:31 2010
Errors in file /oracle/admin/testdb/bdump/testdb_smon_3460.trc:
ORA-00604: error occurred at recursive SQL level 1
ORA-01578: ORACLE data block corrupted (file # 1, block # 52484)
ORA-01110: data file 1: '/oracle/oradata/testdb/system01.dbf'
Sun Mar 7 13:43:31 2010
Database Characterset is AL32UTF8
replication_dependency_tracking turned off (no async multimaster replication found)
Starting background process QMNC
QMNC started with pid=14, OS id=3474
Sun Mar 7 13:43:32 2010
db_recovery_file_dest_size of 2048 MB is 0.00% used. This is a
user-specified limit on the amount of space that will be used by this
database for recovery-related files, and does not reflect the amount of
space available in the underlying filesystem or ASM diskgroup.
Sun Mar 7 13:43:33 2010
Completed: ALTER DATABASE OPEN
Sun Mar 7 13:44:22 2010
Errors in file /oracle/admin/test/udump/testdb_ora_3956.trc:
ORA-07445: exception encountered: core dump [ttcclr()+1770] [SIGSEGV] [Address not mapped to object] [0x14] [] []
Sun Mar 7 13:44:31 2010
Errors in file /oracle/admin/test/udump/testdb_ora_3958.trc:
ORA-07445: exception encountered: core dump [ttcclr()+1770] [SIGSEGV] [Address not mapped to object] [0x12] [] []
Sun Mar 7 13:53:26 2010
Errors in file /oracle/admin/test/udump/testdb_ora_3960.trc:
ORA-07445: exception encountered: core dump [ttcclr()+1770] [SIGSEGV] [Address not mapped to object] [0x14] [] []
Sun Mar 7 14:12:09 2010
Errors in file /oracle/admin/test/udump/testdb_ora_3994.trc:
ORA-07445: exception encountered: core dump [ttcclr()+1770] [SIGSEGV] [Address not mapped to object] [0x1A] [] []

ExtractValue Problem----- ORA-03113: end-of-file on communication channel

Dear Friend
My system is now on this version
Oracle9i Enterprise Edition Release 9.2.0.4.0 - Production
PL/SQL Release 9.2.0.4.0 - Production
CORE 9.2.0.3.0 Production
TNS for Linux: Version 9.2.0.4.0 - Production
NLSRTL Version 9.2.0.4.0 - Production
earlier it was
Oracle9i Enterprise Edition Release 9.2.0.1.0 - Production
PL/SQL Release 9.2.0.1.0 - Production
CORE 9.2.0.1.0 Production
TNS for Linux: Version 9.2.0.1.0 - Production
NLSRTL Version 9.2.0.1.0 - Production
I took the cold backup and restore in this version. after that my total system is working very fine, is very fast using than the previous. But only one SQL command is not working which is very urgent to solve
when i am using this
SELECT extractValue(value(t),'/produit/cod_pro') cod_pro
FROM XMLTABLE X,
TABLE ( xmlsequence (extract(XML_data,'/com/lProducts_list/produit'))) t;
the table structure is
SQL> desc XMLTABLE
Name Null? Type
DOC_ID NUMBER
FILE_NAME VARCHAR2(100)
FILE_DATE DATE
XML_DATA XMLTYPE
I got this error
ORA-03113: end-of-file on communication channel
Please guide me what to do to resolve this problem
Farhad

The error, ORA-03113: end-of-file on communication channel, is a client driver error.
The client driver simply says "hey, the database server process servicing me has torn down our communication channel!". The communication channel is usually a TCP socket connection.
The client driver has no idea why. Thus you have to find the answer on the server.
The usual cause of this communication channel being closed, is the Oracle server process servicing the client session, crashing.
When an Oracle server process crashes, it does not tell the client. It usually writes an entry to the Oracle alert log, dump a trace file, and terminates. When terminating, it tears down any communication channels (TCP sockets) it may have opened.
Therefore you must look in
- the alert log file for an error message listing the crash (which will include the dump file name)
- look at this trace dump file (which is difficult if you're not familiar with Oracle internals)
- investigate the stack trace and error codes on Metalink (which often turns up the bug notes for this, plus any workarounds or available patches)
Usually however, one should log these types of Oracle server errors as a TAR with Oracle Support.

ERROR at line 1: ORA-03113: end-of-file on communication channel

I have a DRDA gateway set up between my Oracle 10.2.0.4 instance (linux 64 bit), and a DB2 UDB instance (v 8.2 32 Bit). Within a SQLPlus session, I'm able to successfully interact with tables (select, update, delete) that do not have CLOB columns defined. However, whenever I try to access a table with a CLOB column (even doing a select on a single row), I get an end-of-file communication error and it kicks me out of my session. I can interact with the table as long as I don't include the CLOB/BLOB columns. Is there some additional configuration or memory settings I'm missing?
Connected to:
Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - 64bit Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
SQL> select * from mytablewithaclob@db2 where l_column= 1;
select * from dbo.mytablewithaclob@db2 where l_column= 1
ERROR at line 1:
ORA-03113: end-of-file on communication channelI add the following codepage.map (as per another post) - packages are bound properly, and tables without clobs I can interact with just fine.:
S 367 > US7ASCII
D 13488 > AL16UTF16
D 1200 = AL16UTF16
M 1208 = UTF8
MBC 1208 = 1208 1200
column in question is defined as follows in DB2:
mycol CLOB(1G) DEFAULT '' NOT NULL NOT LOGGEDAny thoughts? Much appreciated.
Edited by: kpw on Sep 3, 2009 4:21 PM

Unfortunately TG4DRDA 10.2.0 does not support DB2 (C)LOB datatype. A crash is a rather unpleasant way to report this inability... but that could depend on the version of TG4DRDA you use (10.2.0.1.0). I guess when you upgrade TG4DRDA to 10.2.0.4.0 you will get a proper error message (ORA-28527 - datatype conversion error).
As workaround you could:
- create a view on the DB2 table and exclude the (C)LOB columns
- move/copy the (C)LOB data into a TG4DRDA supported datatype ((long) varchar for bit data)
Regards,
Ed

FULL JOIN Error - (ORA-03113: end-of-file on communication channel)

Hello,
well my following query is running fine, no errors but not showing join records from table B and E.
Query is as following:
SELECT D.EMPLOYEE_ID, F.EMP_NAME,
F.COMPANY_ID, F.COMP_NAME, F.BRANCH_ID, F.BR_NAME,
TO_CHAR(F.BIRTH_DATE,'DD/MM/YYYY') DOB,
((NVL(A.PF_OWN,0) + NVL(A.PF_COMP,0) + NVL(A.PROF_OWN,0) + NVL(A.PROF_COMP,0) + NVL(B.PROF_OWN,0) +
NVL(B.PROF_COMP,0) + NVL(B.TOT_PF_OWN,0) + NVL(B.TOT_PF_COMP,0) +
NVL(D.SAL_PF_OWN,0) + NVL(D.SAL_PF_COMP,0) -
(NVL(E.REV_PF_OWN,0) + NVL(E.REV_PF_COMP,0) + NVL(C.WD_PF_OWN,0) + NVL(C.WD_PF_COMP,0) +
NVL(C.WD_PROF_OWN,0) + NVL(C.WD_PROF_COMP,0)))) PF_BALANCE
FROM
(SELECT EMPLOYEE_ID, SUM(PF_OWN) SAL_PF_OWN, SUM(PF_COMP) SAL_PF_COMP
FROM EMPLOYEE.EMP_SAL_DETAILS
WHERE SAL_DATE >= (SELECT MAX(AS_ON_DATE) FROM EMPLOYEE.EMP_PF_OPBALS WHERE AS_ON_DATE <= '01-DEC-06')
AND SAL_DATE <= '01-DEC-06'
GROUP BY EMPLOYEE_ID) D
LEFT JOIN
(SELECT EMPLOYEE_ID, PF_OWN, PF_COMP, PROF_OWN, PROF_COMP
FROM EMPLOYEE.EMP_PF_OPBALS
WHERE AS_ON_DATE IN (SELECT MAX(AS_ON_DATE) FROM EMPLOYEE.EMP_PF_OPBALS WHERE AS_ON_DATE <= '01-DEC-06')) A
ON (D.EMPLOYEE_ID = A.EMPLOYEE_ID)
LEFT JOIN
(SELECT EMPLOYEE_ID, SUM(TOT_PF_OWN) TOT_PF_OWN, SUM(TOT_PF_COMP) TOT_PF_COMP, SUM(PROF_OWN) PROF_OWN, SUM(PROF_COMP) PROF_COMP
FROM EMPLOYEE.EMP_PF_PROF_DETAILS WHERE END_DATE >= (SELECT MAX(AS_ON_DATE) FROM EMPLOYEE.EMP_PF_OPBALS WHERE AS_ON_DATE <= '01-DEC-06')
GROUP BY EMPLOYEE_ID) B
ON (D.EMPLOYEE_ID = B.EMPLOYEE_ID)
LEFT JOIN
(SELECT EMPLOYEE_ID, SUM(PF_OWN) WD_PF_OWN, SUM(PF_COMP) WD_PF_COMP, SUM(PROF_OWN) WD_PROF_OWN, SUM(PROF_COMP) WD_PROF_COMP
FROM EMPLOYEE.EMP_PF_WITHDRAWALS WHERE PF_WDRAW_DATE >= (SELECT MAX(AS_ON_DATE) FROM EMPLOYEE.EMP_PF_OPBALS WHERE AS_ON_DATE <= '01-DEC-06')
GROUP BY EMPLOYEE_ID) C
ON (D.EMPLOYEE_ID = C.EMPLOYEE_ID)
LEFT JOIN
(SELECT EMPLOYEE_ID, SUM(PF_OWN) REV_PF_OWN, SUM(PF_COMP) REV_PF_COMP
FROM EMPLOYEE.EMP_SAL_REVERSALS
WHERE SAL_DATE >= (SELECT MAX(AS_ON_DATE) FROM EMPLOYEE.EMP_PF_OPBALS WHERE AS_ON_DATE >= '01-DEC-06')
AND SAL_DATE <= '01-DEC-06'
GROUP BY EMPLOYEE_ID) E
ON (D.EMPLOYEE_ID = E.EMPLOYEE_ID)
LEFT JOIN
(SELECT EMPLOYEE_ID, COMPANY_ID, COMP_NAME, BRANCH_ID, BR_NAME, EMP_NAME, BIRTH_DATE, CONF_DATE FROM V_SEL_SYS_EMP) F
ON (D.EMPLOYEE_ID = F.EMPLOYEE_ID)
ORDER BY D.EMPLOYEE_ID
And when i try to full join my tables and replace LEFT JOIN with FULL OUTER JOIN following errors accurs:
(ORA-03113: end-of-file on communication channel) and oracle gets disconnect.
Query will only show records its tables are FULL JOINED.
Please help what is the solution. Its very urgent also.
I am thankful to you.
Regards,
Imran

> And when i try to full join my tables and replace LEFT JOIN with FULL OUTER
JOIN following errors accurs:
(ORA-03113: end-of-file on communication channel) and oracle gets disconnect.
This is not an error, but a symptom of an error. An ORA-03113 results when the Oracle Server Process that services your client, terminates abnormally. When that server process terminates, the connection (TCP socket) to your client is torn down - without your client knowing about it.
The ORA-03113 is generated by your client's Oracle driver when your client suddenly discovers that the connection to the Oracle Server Process is no longer there.
So why does the Server Process terminate abnormally? Usually due to an internal error (an ORA-600 and/or an ORA-7445).
These errors results in:
- error messages to be generated in the Oracle instance's alert log
- a trace file generated by the server process that includes stack and memory dumps
You need to determine what these errors are, and the use the ORA-600/ORA-7445 Troubleshooter (Metalink Note 153788.1) on [url http://metalink.oracle.com]Metalink to troubleshoot the error, determine whether it is a bug, if there is a patch or a workaround available, etc.
> Please help what is the solution. Its very urgent also.
I do not mind helping out where I can. But I do have a problem with people claiming there problem is urgent, and deserves quicker/better attention that other peoples' problems around here,
If your problem is urgent then you are in the wrong place. I do not get paid to solve urgent problems quickly. I and others, spend our free time providing assistance. You cannot demand any kind of urgent attention from any of us.
If you like urgent and special attention, use Oracle Support.

ORA-03113: end-of-file on communication channel while running utlrp.sql

Hi All,
I am getting ORA-03113: end-of-file on communication channel while running utlrp.sql.
Actually I ran utlrp.sql before in my test database and terminated the session in middle after that when ever I am running utlrp.sql i am getting the above error.
how to resolve the issue.
Kiran

Hi,
Check as follows;
select owner,object_name,object_type from dba_objects where object_name like
'UTL%'The objects should only belong to SYS, if they do belong do some other user (example HR, etc). Drop objects belong to other users (not sys).
Then connect as sys as sysdba and rerun the script utlrp.sql.
You can also review Oracle Metalink Document: ORA-03113 Error When Executing Utlrp.sql: Doc ID: Note:413922.1
https://metalink.oracle.com/metalink/plsql/f?p=130:14:5496440446026841247::::p14_database_id,p14_docid,p14_show_header,p14_show_help,p14_black_frame,p14_font:NOT,413922.1,1,1,1,helvetica
Adith

Context Index World Lexer ORA-03113: end-of-file on communication channel

I have release 10g Release 1 (10.1.0.2.0) for Windows and trying to take advantage of the World Lexer.
My table is:
create table worldtest(
filename char(32),
content blob
I've created a preference for the WORLD_LEXER:
     begin
     ctx_ddl.create_preference('wlex', 'WORLD_LEXER');
     end;
Right now I'm working with 10,000 records of PDF, MS-Word, Text, and HTM documents. When I try and create a context index using this lexer:
create index i_ctx_wc on worldtest(content)
indextype is ctxsys.context
parameters ('lexer wlex');
The following error is returned, and I have to use drop index force to remove the index. Without 'Force' a message is return saying the index is an a loading state, but nothing is occuring.
create index i_ctx_wc on worldtest(content)
indextype is ctxsys.context
ERROR at line 1:
ORA-03113: end-of-file on communication channel
The indexing works fine when I leave the world lexer preference out.
Any suggestions would be wonderful. I've been banging my head on this one for a while.
Thanks.

Hi,
I couldn't reproduce (different version which is perhaps the problem). Check out bug 4056162. ORA-3113 related to the use of the world lexer. The resolution is not published externally, so no help there, but the bug is closed so perhaps support can provide some insight.
Thanks.

End of file on communication channel when doing a (hash) join

Similar Messages

Maybe you are looking for