FreeTDS Development Group

["Federico Alves" <sales AT minixel.com>]

Dear freddy
I downloaded the most current nightly version from
ftp://ftp.ibiblio.org/pub/Linux/ALPHA/freetds/current/freetds-current.tgz,
compiled it and it crashes Asterisk in 60 seconds, when I use my call
simulator. I have a test bed where you may login and reproduce the issue at
pleasure. I will appreciate if you contact me as a consultant or can
recommend somebody who can overtake this project, and fix it. 

-----Original Message-----
From: freetds-bounces AT lists.ibiblio.org
[mailto:freetds-bounces AT lists.ibiblio.org] On Behalf Of
freetds-request AT lists.ibiblio.org
Sent: Thursday, January 10, 2008 7:58 AM
To: freetds AT lists.ibiblio.org
Subject: FreeTDS Digest, Vol 60, Issue 11

Send FreeTDS mailing list submissions to
        freetds AT lists.ibiblio.org

To subscribe or unsubscribe via the World Wide Web, visit
        http://lists.ibiblio.org/mailman/listinfo/freetds
or, via email, send a message with subject or body 'help' to
        freetds-request AT lists.ibiblio.org

You can reach the person managing the list at
        freetds-owner AT lists.ibiblio.org

When replying, please edit your Subject line so it is more specific
than "Re: Contents of FreeTDS digest..."


Today's Topics:

   1. Re: Are LOBs supported? Two additional bytes are  fetched for
      text column. (Frediano Ziglio)
   2. Re: Are LOBs supported? (Frediano Ziglio)
   3. Re: kerberos authentication module for mssql [linux]
      (Frediano Ziglio)
   4. Re: Request for commercial support (Frediano Ziglio)
   5. Re: Are LOBs supported? - Binding before prepare
      (Sebastien FLAESCH)
   6. Bug tracking system for FreeTDS (Sebastien FLAESCH)
   7. Re: Bug tracking system for FreeTDS (ZIGLIO, Frediano, VF-IT)
   8. Re: SQLNumResultCols() is wrong (+1) (ZIGLIO, Frediano, VF-IT)
   9. Re: SQLNumResultCols() is wrong (+1) (Sebastien FLAESCH)


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

Message: 1
Date: Thu, 10 Jan 2008 07:46:47 +0100
From: Frediano Ziglio <freddy77 AT gmail.com>
Subject: Re: [freetds] Are LOBs supported? Two additional bytes are
        fetched for text column.
To: FreeTDS Development Group <freetds AT lists.ibiblio.org>
Message-ID: <1199947607.7542.7.camel@freddy>
Content-Type: text/plain


Il giorno mer, 09/01/2008 alle 11.11 +0100, Sebastien FLAESCH ha
scritto:
> Hello,
> 
> Here is a new version of the odbctest17.c sample...
> 
> Now the program uses SQL_LEN_DATA_AT_EXEC() macro for SQLBindParameter().
> 
> But still uses SQL_DATA_AT_EXEC for SQLBindCol() (I suppose this is
supported).
> 
> I think there is a problem when fetching text data:
> 
> By inserting 10000 bytes, then fetching, I get 2 additional bytes for the
text column.
> 
> I checked the values inserted in the database with:
> 
>    select datalength(t), datalength(b) from tt
> 
> returns:
> 
> 0       10000         10000
> 1       10000         10000
> 
> So to me the insert is ok.
> 
> Try the following:
> 
> ./odbctest17.bin dsn user pswd 2
> 
> where 2 is the number of rows to be inserted.
> 
> Output gives:
> 
>  >> insert... 0
>  >> SQLParamData: ptr = 0xbff4b990  rcode = 99
>  >> param 0xbff4b990: total bytes written = 10000
>  >> SQLParamData: ptr = 0xbff4b990  rcode = 99
>  >> param 0xbff4b990: total bytes written = 10000
>  >> SQLParamData: ptr = 0xbff4b990  rcode = 0
>  >> insert... 1
>  >> SQLParamData: ptr = 0xbff4b990  rcode = 99
>  >> param 0xbff4b990: total bytes written = 10000
>  >> SQLParamData: ptr = 0xbff4b990  rcode = 99
>  >> param 0xbff4b990: total bytes written = 10000
>  >> SQLParamData: ptr = 0xbff4b990  rcode = 0
>  >> fetch... 0  rcode = 0
>  >> readBlob field 1
>  >>     step 1: 4096 bytes read...
>  >>     step 2: 4096 bytes read...
>  >>     step 3: 1810 bytes read...
>  >>   total bytes read = 10002               <--- should be 10000
>  >> readBlob field 2

I'll fix this problem is that it consider numm-termination cause source
is longvarchar while it should consider destination (witch is
SQL_C_BINARY)

freddy77




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

Message: 2
Date: Thu, 10 Jan 2008 07:48:02 +0100
From: Frediano Ziglio <freddy77 AT gmail.com>
Subject: Re: [freetds] Are LOBs supported?
To: FreeTDS Development Group <freetds AT lists.ibiblio.org>
Message-ID: <1199947682.7542.10.camel@freddy>
Content-Type: text/plain


Il giorno mer, 09/01/2008 alle 10.50 +0100, Sebastien FLAESCH ha
scritto:
> Little question:
> 
> I am using now SQL_LEN_DATA_AT_EXEC() for SQLBindParameter()...
> 
> Can I still use SQL_DATA_AT_EXEC for SQLBindCol()?
> = Is it supported to fetch data?
> 
> Thanks,
> Seb
> 

There are two cases where this parameter is used for deferred input:
- bulk adding (not supported)
- cursors updates, supported but still SQL_DATA_AT_EXEC not supported.

freddy77




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

Message: 3
Date: Thu, 10 Jan 2008 07:49:18 +0100
From: Frediano Ziglio <freddy77 AT gmail.com>
Subject: Re: [freetds] kerberos authentication module for mssql
        [linux]
To: FreeTDS Development Group <freetds AT lists.ibiblio.org>
Message-ID: <1199947758.7542.13.camel@freddy>
Content-Type: text/plain


Il giorno mer, 09/01/2008 alle 18.05 -0500, Christos Zoulas ha scritto:
> Hello,
> 
> A co-worker of mine wrote this module. It is available from:
> 
>     http://code.google.com/p/libsqljdbc-auth/
> 
> This works with the microsoft jdbc driver.
> 
> Enjoy,
> 
> christos

I'll have a look at this implementations. We implemented kerberos
support some months ago too. I think however that we don't add any
documentations :(

freddy77




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

Message: 4
Date: Thu, 10 Jan 2008 07:52:08 +0100
From: Frediano Ziglio <freddy77 AT gmail.com>
Subject: Re: [freetds] Request for commercial support
To: FreeTDS Development Group <freetds AT lists.ibiblio.org>
Message-ID: <1199947928.7542.17.camel@freddy>
Content-Type: text/plain


Il giorno gio, 10/01/2008 alle 01.23 -0500, Federico Alves ha scritto:
> Dear friends
> This is the case:
> I installed current version in a 64 Bit RHEL 5 machine, and it crashes
after
> 60 seconds when using Asterisk and ODBC, under heavy load. I have no
> expertise in house to do debugging and fixing. I want to hire somebody and
> pay by results, I mean, when the driver does not crash. I now it is
freetds
> the culprit, because I replace the driver in odbc.ini with a commercial
> driver and Asterisk works fine, "ceteris paribus". If somebody is
interested
> in the challenge, please contact me at: falves1 at hotmail. I believe that
a
> much more robust freetds will come out of this experience.
> 
> 

I remember we did some fixes for Asterix. I don't know if RHEL use our
post patches. I would suggest to try a recent snapshot. Currently we are
preparing RCs.

freddy77




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

Message: 5
Date: Thu, 10 Jan 2008 09:16:56 +0100
From: Sebastien FLAESCH <sf AT 4js.com>
Subject: Re: [freetds] Are LOBs supported? - Binding before prepare
To: FreeTDS Development Group <freetds AT lists.ibiblio.org>
Message-ID: <4785D478.8060206 AT 4js.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Frediano Ziglio wrote:
>>> examples you bind parameters after SQLPrepare. I would suggest to bind
>>> before. In many cases (using many drivers) it's faster.
>> Thanks for the tip, but actually we have to do SQLPrepare() before
because
>> the basic SQL instructions in Informix 4gl are PREPARE, then EXECUTE with
>> parameter definition and values:
>>
>> PREPARE s1 FROM "INSERT INTO tab VALUES( ?, ?, ? )"  -- SQLPrepare()
>> EXECUTE s1 USING var1, var2, var3  -- SQLExecute()
>>
>> ...
>>
> 
> I would try to prepare before even for Informix. Driver should just not
> use binding informations if protocol does not require them. Binding data
> is a way to say "take missing data from here" it do not force anything.
> 
> If it works (as it should) you will gain performance with drivers that
> work better with type informations on prepare.

Maybe I was not clear:

I have *no* SQL paramater type information at prepare time...
So I can't SQLBindParameter() before SQLPrepare()...

I could postpone SQLPrepare() just before SQLExecute(), but then I
would not get SQL Error at prepare time (for example to get "table not
found" errors).
= we do not use deferred prepare.

Anyway thanks for the tip, I keep it in mind.
Seb


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

Message: 6
Date: Thu, 10 Jan 2008 09:39:46 +0100
From: Sebastien FLAESCH <sf AT 4js.com>
Subject: [freetds] Bug tracking system for FreeTDS
To: FreeTDS Development Group <freetds AT lists.ibiblio.org>
Message-ID: <4785D9D2.4050606 AT 4js.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Dear all,

I could not find any bug tracking system on the web site...
I can't imagine a serious project without a bug tracking system.
This will get critical if we start to support FreeTDS with our product.

We are using bugzilla for years now, and we are very satisfied.
Any chance to have someone installing and configuring bugzilla?
(we can help if needed)

Maybe the 0.82 release could be the right time to set this up.

Thanks!
Seb


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

Message: 7
Date: Thu, 10 Jan 2008 09:43:32 +0100
From: "ZIGLIO, Frediano, VF-IT" <Frediano.Ziglio AT vodafone.com>
Subject: Re: [freetds] Bug tracking system for FreeTDS
To: "FreeTDS Development Group" <freetds AT lists.ibiblio.org>
Message-ID:
        <72EBB5FADDA71343B78E1F09FF471F6E452612 AT OBOMEXO02.omnitel.it>
Content-Type: text/plain;       charset=US-ASCII

> 
> Dear all,
> 
> I could not find any bug tracking system on the web site...
> I can't imagine a serious project without a bug tracking system.
> This will get critical if we start to support FreeTDS with 
> our product.
> 
> We are using bugzilla for years now, and we are very satisfied.
> Any chance to have someone installing and configuring bugzilla?
> (we can help if needed)
> 
> Maybe the 0.82 release could be the right time to set this up.
> 
> Thanks!
> Seb

We use bug tracking at sourceforge.net site.

freddy77


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

Message: 8
Date: Thu, 10 Jan 2008 09:52:40 +0100
From: "ZIGLIO, Frediano, VF-IT" <Frediano.Ziglio AT vodafone.com>
Subject: Re: [freetds] SQLNumResultCols() is wrong (+1)
To: "FreeTDS Development Group" <freetds AT lists.ibiblio.org>
Message-ID:
        <72EBB5FADDA71343B78E1F09FF471F6E452613 AT OBOMEXO02.omnitel.it>
Content-Type: text/plain;       charset=US-ASCII

> > > 
> > > Try attached sample, SQLNumResultCols() gives numcols + 1
> > > 
> > >  >> Number of columns in result set: 4
> > > 
> > > should be:
> > > 
> > >  >> Number of columns in result set: 3
> > > 
> > > 
> > > Seb
> > > 
> > 
> > >From dump
> > 
> > 15:19:50.359222 (net.c:740):Sending packet
> > 0000 03 01 00 5e 00 00 01 00-ff ff 02 00 00 00 00 01 
> |...^.... ........|
> > 0010 26 04 00 00 00 63 20 00-00 00 09 04 d0 00 00 20 
> |&....c . ....... |
> > 0020 00 00 00 53 00 45 00 4c-00 45 00 43 00 54 00 20 
> |...S.E.L .E.C.T. |
> > 0030 00 2a 00 20 00 46 00 52-00 4f 00 4d 00 20 00 74 |.*. 
> .F.R .O.M. .t|
> > 0040 00 31 00 00 01 26 04 04-04 00 00 00 00 01 26 04 
> |.1...&.. ......&.|
> > 0050 04 04 20 00 00 00 01 26-04 04 00 00 00 00       |.. 
> ....& ......|
> > 
> > 15:19:50.359863 (util.c:162):Changed query state from QUERYING to
> > PENDING
> > 15:19:50.360069 (token.c:495):tds_process_tokens(0x8060c50, 
> 0xbfe02374,
> > 0xbfe02370, 0x100)
> > 15:19:50.360230 (util.c:162):Changed query state from 
> PENDING to READING
> > 15:19:50.360813 (net.c:547):Received header
> > 0000 04 01 00 b8 00 38 01 00-                        |.....8..|
> > 
> > 15:19:50.361031 (net.c:627):Received packet
> > 0000 04 01 00 b8 00 38 01 00-81 04 00 00 00 09 00 26 
> |.....8.. .......&|
> > 0010 04 01 6b 00 00 00 09 00-af 0a 00 09 04 d0 00 00 
> |..k..... ........|
> > 0020 01 63 00 00 00 09 00 a7-0a 00 09 04 d0 00 00 02 
> |.c...... ........|
> > 0030 76 00 63 00 00 00 00 00-38 07 72 00 6f 00 77 00 
> |v.c..... 8.r.o.w.|
> > 0040 73 00 74 00 61 00 74 00-a4 07 00 01 02 00 74 00 
> |s.t.a.t. ......t.|
> > 
> > rowstat columns
> > 
> > 0050 31 00 a5 0c 00 01 01 00-02 01 00 03 01 00 04 00 
> |1....... ........|
> > 0060 14 ff 01 00 c1 00 00 00-00 00 79 00 00 00 00 ac 
> |........ ..y.....|
> > 
> > column 4 is hidden :(
> > 
> > >From TODO:
> > 
> > . test and fix: hidden fields (FOR BROWSE select, see flag 
> test on tds)
> > 
> > yes, correctly there is something to do with hidden fields:
> > - what happen if we bind to an hidden field??
> > - if we use SQLGetData??
> > - if we request informations with
> > SQLDescribeCol/SQLColAttribute(s)/SQLGetDescField??
> > - as you noted returning # columns hidden fields are not 
> counted (there
> > is however a setting which is a mssql extension which threat hidden
> > columns as normal)
> > 
> > 0070 0d 00 00 01 38 00 00 00-26 04 04 f0 de bc 0a ac 
> |....8... &.......|
> > 0080 0d 00 00 01 38 00 00 00-26 04 04 04 00 00 00 ac 
> |....8... &.......|
> > 0090 0d 00 00 01 38 00 00 00-26 04 04 04 00 00 00 ac 
> |....8... &.......|
> > 00a0 0d 00 00 01 38 00 00 00-26 04 04 ff ff ff ff fe 
> |....8... &.......|
> > 00b0 00 00 e0 00 00 00 00 00-                        |........|
> > 
> 
> I'll write tests to TODO and apply a patch that filter entirely hidden
> fields
> 

Done

freddy77


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

Message: 9
Date: Thu, 10 Jan 2008 13:56:47 +0100
From: Sebastien FLAESCH <sf AT 4js.com>
Subject: Re: [freetds] SQLNumResultCols() is wrong (+1)
To: FreeTDS Development Group <freetds AT lists.ibiblio.org>
Message-ID: <4786160F.1050003 AT 4js.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed

Frediano,

Just extracted from CVS, works now as expected.

Thanks.
Seb

ZIGLIO, Frediano, VF-IT wrote:
>>>> Try attached sample, SQLNumResultCols() gives numcols + 1
>>>>
>>>>  >> Number of columns in result set: 4
>>>>
>>>> should be:
>>>>
>>>>  >> Number of columns in result set: 3
>>>>
>>>>
>>>> Seb
>>>>
>>> >From dump
>>>
>>> 15:19:50.359222 (net.c:740):Sending packet
>>> 0000 03 01 00 5e 00 00 01 00-ff ff 02 00 00 00 00 01 
>> |...^.... ........|
>>> 0010 26 04 00 00 00 63 20 00-00 00 09 04 d0 00 00 20 
>> |&....c . ....... |
>>> 0020 00 00 00 53 00 45 00 4c-00 45 00 43 00 54 00 20 
>> |...S.E.L .E.C.T. |
>>> 0030 00 2a 00 20 00 46 00 52-00 4f 00 4d 00 20 00 74 |.*. 
>> .F.R .O.M. .t|
>>> 0040 00 31 00 00 01 26 04 04-04 00 00 00 00 01 26 04 
>> |.1...&.. ......&.|
>>> 0050 04 04 20 00 00 00 01 26-04 04 00 00 00 00       |.. 
>> ....& ......|
>>> 15:19:50.359863 (util.c:162):Changed query state from QUERYING to
>>> PENDING
>>> 15:19:50.360069 (token.c:495):tds_process_tokens(0x8060c50, 
>> 0xbfe02374,
>>> 0xbfe02370, 0x100)
>>> 15:19:50.360230 (util.c:162):Changed query state from 
>> PENDING to READING
>>> 15:19:50.360813 (net.c:547):Received header
>>> 0000 04 01 00 b8 00 38 01 00-                        |.....8..|
>>>
>>> 15:19:50.361031 (net.c:627):Received packet
>>> 0000 04 01 00 b8 00 38 01 00-81 04 00 00 00 09 00 26 
>> |.....8.. .......&|
>>> 0010 04 01 6b 00 00 00 09 00-af 0a 00 09 04 d0 00 00 
>> |..k..... ........|
>>> 0020 01 63 00 00 00 09 00 a7-0a 00 09 04 d0 00 00 02 
>> |.c...... ........|
>>> 0030 76 00 63 00 00 00 00 00-38 07 72 00 6f 00 77 00 
>> |v.c..... 8.r.o.w.|
>>> 0040 73 00 74 00 61 00 74 00-a4 07 00 01 02 00 74 00 
>> |s.t.a.t. ......t.|
>>> rowstat columns
>>>
>>> 0050 31 00 a5 0c 00 01 01 00-02 01 00 03 01 00 04 00 
>> |1....... ........|
>>> 0060 14 ff 01 00 c1 00 00 00-00 00 79 00 00 00 00 ac 
>> |........ ..y.....|
>>> column 4 is hidden :(
>>>
>>> >From TODO:
>>>
>>> . test and fix: hidden fields (FOR BROWSE select, see flag 
>> test on tds)
>>> yes, correctly there is something to do with hidden fields:
>>> - what happen if we bind to an hidden field??
>>> - if we use SQLGetData??
>>> - if we request informations with
>>> SQLDescribeCol/SQLColAttribute(s)/SQLGetDescField??
>>> - as you noted returning # columns hidden fields are not 
>> counted (there
>>> is however a setting which is a mssql extension which threat hidden
>>> columns as normal)
>>>
>>> 0070 0d 00 00 01 38 00 00 00-26 04 04 f0 de bc 0a ac 
>> |....8... &.......|
>>> 0080 0d 00 00 01 38 00 00 00-26 04 04 04 00 00 00 ac 
>> |....8... &.......|
>>> 0090 0d 00 00 01 38 00 00 00-26 04 04 04 00 00 00 ac 
>> |....8... &.......|
>>> 00a0 0d 00 00 01 38 00 00 00-26 04 04 ff ff ff ff fe 
>> |....8... &.......|
>>> 00b0 00 00 e0 00 00 00 00 00-                        |........|
>>>
>> I'll write tests to TODO and apply a patch that filter entirely hidden
>> fields
>>
> 
> Done
> 
> freddy77
> _______________________________________________
> FreeTDS mailing list
> FreeTDS AT lists.ibiblio.org
> http://lists.ibiblio.org/mailman/listinfo/freetds
> 



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

_______________________________________________
FreeTDS mailing list
FreeTDS AT lists.ibiblio.org
http://lists.ibiblio.org/mailman/listinfo/freetds


End of FreeTDS Digest, Vol 60, Issue 11
***************************************