PL/Dev over Instant Client quering "forever" 10.1.0.5 on large resultsets

[vhbtechnik]

So far we had Oracle DB 10.1.0.5 Patch 2 on W2k3 and XP-Clients using Instant Client 10.1.0.5 an PL/SQL Developer 5.16 or 6.05 to query your DB. This scenario worked well.

Now we upgraded to ORACLE 10G 10.1.0.5 PATCH 25 and now our PL/SQL Developer 5.16 or 6.05 (on IC 10.1.0.5) can logon the db and also query small tables. But as soon as the resultset reaches a certain size, the query on a table won't come to an end and is always showing "Executing...". And i can only press "BREAK" what the results in a "ORA-12152: TNS: unable to send break message" and "ORA-03114: not connected to ORACLE".
If i narrow the resultset down on the same table it works before.
If i watch the sessions on small resultset-queries, i see the corresponding session, on large resultset-queries the session seem to close immediately.

To solve this issue a already tried to install the newest PL/SQL developer 7.1.5(trail) or/and installing a newer instant client version (10.2.0.4), which both did not solve the problem.

Is there a new option in 10.1.0.5 Patch 25 (or before) which closes sessions if the resultsets getting to large over a slower internet connection?

btw. using sqlplus in the instantclient directory or even excel over odbc on the same client, returns the full resultset without problems.

Edit2:
Here is a snippet of the tracefile on the client right after Executing the select-statement. Some data seems to be retrieved and than it ends with these lines:

(1992) [20-AUG-2008 17:13:00:953] nsprecv: 2D 20 49 6E 74 72 61 6E |-.Intran|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: 65 74 2D 47 72 75 6E 64 |et-Grund|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: 6C 61 67 65 6E 02 C1 04 |lagen...|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: 02 C1 03 02 C1 0B 02 C1 |........|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: 51 00 02 C1 03 02 C1 2D |Q......-|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: 05 48 4B 4F 50 50 01 80 |.HKOPP..|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: 03 3E 64 66 01 80 07 78 |.>df...x|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: 65 0B 0F 01 01 01 07 76 |e......v|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: C7 01 01 09 01 01 07 76 |.......v|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: C7 01 01 18 01 01 07 78 |.......x|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: 65 0B 0F 01 01 01 07 76 |e......v|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: C7 01 01 09 01 01 07 76 |.......v|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: C7 01 01 18 01 01 02 C1 |........|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: 3B 02 C1 02 01 80 00 00 |;.......|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: 00 00 00 00 00 00 00 00 |........|
(1992) [20-AUG-2008 17:13:00:953] nsprecv: 00 00 01 80 15 0C 00 |....... |
(1992) [20-AUG-2008 17:13:00:953] nsprecv: normal exit
(1992) [20-AUG-2008 17:13:00:953] nsrdr: got NSPTDA packet
(1992) [20-AUG-2008 17:13:00:953] nsrdr: NSPTDA flags: 0x0
(1992) [20-AUG-2008 17:13:00:953] nsrdr: normal exit
(1992) [20-AUG-2008 17:13:00:953] snsbitts_ts: entry
(1992) [20-AUG-2008 17:13:00:953] snsbitts_ts: acquired the bit
(1992) [20-AUG-2008 17:13:00:953] snsbitts_ts: normal exit
(1992) [20-AUG-2008 17:13:00:953] snsbitcl_ts: entry
(1992) [20-AUG-2008 17:13:00:953] snsbitcl_ts: normal exit
(1992) [20-AUG-2008 17:13:00:953] nsdo: *what=1, *bl=2001
(1992) [20-AUG-2008 17:13:00:953] snsbitts_ts: entry
(1992) [20-AUG-2008 17:13:00:953] snsbitts_ts: acquired the bit
(1992) [20-AUG-2008 17:13:00:953] snsbitts_ts: normal exit
(1992) [20-AUG-2008 17:13:00:953] nsdo: nsctxrnk=0
(1992) [20-AUG-2008 17:13:00:953] snsbitcl_ts: entry
(1992) [20-AUG-2008 17:13:00:953] snsbitcl_ts: normal exit
(1992) [20-AUG-2008 17:13:00:953] nsdo: normal exit
(1992) [20-AUG-2008 17:13:00:953] nioqrc: exit
(1992) [20-AUG-2008 17:13:00:953] nioqrc: entry
(1992) [20-AUG-2008 17:13:00:953] nsdo: entry
(1992) [20-AUG-2008 17:13:00:953] nsdo: cid=0, opcode=85, *bl=0, *what=0, uflgs=0x0, cflgs=0x3
(1992) [20-AUG-2008 17:13:00:953] snsbitts_ts: entry
(1992) [20-AUG-2008 17:13:00:953] snsbitts_ts: acquired the bit
(1992) [20-AUG-2008 17:13:00:953] snsbitts_ts: normal exit
(1992) [20-AUG-2008 17:13:00:953] nsdo: rank=64, nsctxrnk=0
(1992) [20-AUG-2008 17:13:00:953] snsbitcl_ts: entry
(1992) [20-AUG-2008 17:13:00:953] snsbitcl_ts: normal exit
(1992) [20-AUG-2008 17:13:00:953] nsdo: nsctx: state=8, flg=0x100400d, mvd=0
(1992) [20-AUG-2008 17:13:00:953] nsdo: gtn=127, gtc=127, ptn=10, ptc=2011
(1992) [20-AUG-2008 17:13:00:953] snsbitts_ts: entry
(1992) [20-AUG-2008 17:13:00:953] snsbitts_ts: acquired the bit
(1992) [20-AUG-2008 17:13:00:953] snsbitts_ts: normal exit
(1992) [20-AUG-2008 17:13:00:953] snsbitcl_ts: entry
(1992) [20-AUG-2008 17:13:00:953] snsbitcl_ts: normal exit
(1992) [20-AUG-2008 17:13:00:953] nsdo: switching to application buffer
(1992) [20-AUG-2008 17:13:00:953] nsrdr: entry
(1992) [20-AUG-2008 17:13:00:953] nsrdr: recving a packet
(1992) [20-AUG-2008 17:13:00:953] nsprecv: entry
(1992) [20-AUG-2008 17:13:00:953] nsprecv: reading from transport...
(1992) [20-AUG-2008 17:13:00:968] nttrd: entry

 

[Marco Kalter]

To obtain some more diagnostic information, can you modify the 7.1.5 shortcut and add the debugsql parameter? For example:

"C:\Program Files\PLSQL Developer\plsqldev.exe" debugsql

Reproduce the problem and send me the debug.txt file that is generated in the PL/SQL Developer directory.

 

[vhbtechnik]

I sent you the debugfiles via email.

 

[vhbtechnik]

This is also generated in a the \udump trace file in the moment is execute the query that would have a large resultset:

------------------------------------
Dump file \udump\_ora_4148.trc
Fri Aug 22 09:12:18 2008
ORACLE V10.1.0.5.0 - Production vsnsta=0
vsnsql=13 vsnxtr=3
Oracle Database 10g Release 10.1.0.5.0 - Production
With the OLAP and Data Mining options
Windows Server 2003 Version V5.2 Service Pack 2
CPU : 2 - type 586, 1 Physical Cores
Process Affinity : 0x00000000
Memory (Avail/Total): Ph:898M/3071M, Ph+PgF:2675M/4967M, VA:812M/2047M
Instance name:

Redo thread mounted by this instance: 1

Oracle process number: 33

Windows thread id: 4148, image: ORACLE.EXE (SHAD)

*** 2008-08-22 09:12:18.731
*** ACTION NAMESQL Window - select * from stude) 2008-08-22 09:12:18.731
*** MODULE NAMEPL/SQL Developer) 2008-08-22 09:12:18.731
*** SERVICE NAME) 2008-08-22 09:12:18.731
*** SESSION ID145.23131) 2008-08-22 09:12:18.731
opitsk: network error occurred while two-task session server trying to send break; error code = 12152

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

If i narrow down the resultset no trace is written, well and the query then works of course.

 

[vhbtechnik]

So i finally found out what the problem was:

Our firewall guys updated our CISCO PIX 525 to Firmware 8.0(4) and ASDN 6.1(3) and enabled the SQLnet "Fixup" in the software. That caused the "long" sql-queries to fail. After disabling the SQLnet Fixup in the firewall all is running smooth again.

Thanks for all your help