Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Debugging an Oracle connection

Re: Debugging an Oracle connection

From: Marian Aldenhövel <marian_at_mba-software.de>
Date: Mon, 31 Oct 2005 18:04:40 +0100
Message-ID: <dk5irj$pjp$2@newsreader3.netcologne.de>


Hi,

Well, I added a few lines to my sqlnet.ora to enable tracing:

> trace_level_client=admin
> trace_unique_client=yes
> TNSPING.TRACE_LEVEL = admin
> TNSPING.TRACE_DIRECTORY = d:\projekte\oralog
> trace_file_client = cli.trc
> trace_directory_client = d:\projekte\oralog
> log_file_client = sqlnet.log
> log_directory_client = d:\projekte\oralog

I clicked my connection online at designtime - which works - and got two trc files. I then ran my program up to the point of failure and got two other trace files. To compare them I concatenated those with the same client numbers in order of suffix.

The first difference is that in the non-working-case there is a number in braces before the timestamp in the trace file. That is it goes like this:

(1236) [date time] Text

whereas the trace of the working connection does not have that first part and starts like:

[date time] Text

For comparison purposes I removed the timestamp. I then put the files through a diff. They start out the same except for a few lines with - I believe - inconsequential differences:

> New trace stream is d:\projekte\oralog\cli_636.trc
> nncpcin_maybe_init: first request sent to name server will have ID 8874

vs.

> New trace stream is d:\projekte\oralog\cli_1292.trc
> nncpcin_maybe_init: first request sent to name server will have ID 0

I don't think there is something wrong with that, is it? Then about 150 lines into the traces the working sample goes:

> nrigbni: Unable to get data from navigation file tnsnav.ora
> nrigbni: exit
> nrigbi: exit
> nigini: exit
> niqname: Hst is already an NVstring.
> niqname: Inserting CID.
> niotns: entry
> niotns: niotns: setting up interrupt handler...
> niotns: Not trying to enable dead connection detection.
> niotns: Calling address: (
>

DESCRIPTION=(ADDRESS_LIST=(ADDRESS=(PROTOCOL=TCP)(HOST=dodo-mm)(PORT=1521)))(CONNECT_DATA=(SERVER=DEDICATED)(SERVICE_NAME=CMS)(CID=(PROGRAM=D:\prg\Delphi5\Delphi5\Bin\delphi32.exe)(HOST=DODO-MM)(USER=marian))))
> nsmal: 164 bytes at 0xbae35f0
> nscall: connecting...
> nladini: entry
> nladini: exit
> nladget: entry
> nladget: exit
> nttbnd2addr: entry
> nttgetport: entry
> nttgetport: port resolved to 1521
> nttgetport: exit
> nttbnd2addr: looking up IP addr for host: dodo-mm
> nttbnd2addr: exit
> nsmal: 492 bytes at 0xbae3ff8
> nsmal: 2348 bytes at 0xbb19040
> nsopen: opening transport...

And then goes on for about a thousand lines full of interesting information. The only part making sense to me is the one about DESCRIPTION. This is my login information.

The non-working sample stops after "Inserting CID".

Does that yield any useful information as to what might cause the failure?

Yes, I did google some of the keywords here, but to no avail.

Ciao, MM Received on Mon Oct 31 2005 - 11:04:40 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US