Database Software

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!

Import datapump Client disconnect

578869Aug 18 2010

Hello everyone,

I have a small problem with my import via datapump (Oracle Version : 10.2.0.4).

I make my import for a full schema, it was normal and after about 1 hour/ 1 hour 15 the client is disconnected.

I noticed that it happened when he begins to apply the constraints of tables but it can also be a chance.

So my client receives an error message like :
UDI-00008: operation generated ORACLE error 3113
UDI-00008: operation generated ORACLE error 3114
ORA-03114: not connected to ORACLE

I tested by disabling in parallell from 4 to 1 but without result.

Following this, I enabled the trace datapump (parameter : TRACE=1FF0300)

In the master file record (file in bdump : oracleSID_dm00_XXX.trc) , I noticed this:

*** 2010-08-16 18:41:37.772
KUPC: 18:41:37.772: Before Listen: consumer = MCP from queue = SYS.KUPC$C_1_20100816173452
KUPM: 18:41:42.778: Client count is: 1
KUPM: 18:41:42.783: In check_workers...
kupptwa: [08-16-2010 18:41:42.783]
Entered - count total workers alive.
kuppwami: [08-16-2010 18:41:42.784]
Entered - what am I, return process type.
kupptwa: [08-16-2010 18:41:42.784]
Count of workers is 1
KUPM: 18:41:42.784: Live worker count is: 1
KUPM: 18:41:42.784: In set_longops
KUPM: 18:41:42.784: Work so far is: 28516,47222900390625
KUPM: 18:41:42.784: Checking for resumable waits
KUPC: 18:41:42.785: Before Listen: consumer = MCP from queue = SYS.KUPC$C_1_20100816173452
*** 2010-08-16 18:42:27.829
KUPC: 18:42:27.829: Before Listen: consumer = MCP from queue = SYS.KUPC$C_1_20100816173452
KUPC: 18:42:32.834: Before Listen: consumer = MCP from queue = SYS.KUPC$C_1_20100816173452
KUPC: 18:42:35.003: Before Dequeue: msg type is: 1
KUPC: 18:42:35.004: Dequeued msg type: 3031 ID: DG,KUPC$C_1_20100816173452,KUPC$A_1_20100816173455,MCP,930,Y
KUPM: 18:42:35.004: *******IN DISPATCH at 67355, request type=3031
KUPM: 18:42:35.004: Current user is: SYS
KUPM: 18:42:35.004: Log message received from worker DG,KUPC$C_1_20100816173452,KUPC$A_1_20100816173455,MCP,930,Y
KUPM: 18:42:35.004: Processing object type SCHEMA_EXPORT/TABLE/CONSTRAINT/CONSTRAINT
KUPC: 18:42:35.006: Before ENQ: Sending Type: 2023 ID: DG,KUPC$S_1_20100816173452,MCP, ,752,Y
KUPM: 18:42:35.007: *******OUT DISPATCH , request type=3031 response type =2041
KUPC: 18:42:35.008: Before Listen: consumer = MCP from queue = SYS.KUPC$C_1_20100816173452
KUPM: 18:42:36.010: Client count is: 0
KUPM: 18:42:36.010: In check_workers...
kupptwa: [08-16-2010 18:42:36.010]
Entered - count total workers alive.
kuppwami: [08-16-2010 18:42:36.010]
Entered - what am I, return process type.
kupptwa: [08-16-2010 18:42:36.010]
Count of workers is 1
KUPM: 18:42:36.010: Live worker count is: 1
KUPM: 18:42:36.010: In set_longops
KUPM: 18:42:36.011: Work so far is: 28516,47222900390625
KUPM: 18:42:36.011: Checking for resumable waits
KUPC: 18:42:36.012: Before Listen: consumer = MCP from queue = SYS.KUPC$C_1_20100816173452
*** 2010-08-16 18:42:41.016

I think my break is rather like an external network or the other, as if the communication channel,
no message transit and blow a timeout. But I do not know how to verify this is just an idea like this.

If you have ideas, advice, I tried but I'm running out of ideas.

Thank you in advance.