sqlrelay-discussion — Discussion of topics related to SQL Relay

Re: [Sqlrelay-discussion] getColumnType returns 'UNKNOWN' for PostgreSQL ARRAY column type.

[Alfred J F. <alf...@gm...>]

I noticed that I was running an old version of SQLRelay.  I have installed
version 0.39.2 and get the same results.  Just FYI :).  Since I'm already
bugging the list, I must say I'm surprised that there is no API call (at
least in Python) to find out the current version of the SQLRelay client.  I
guess nobody has had a need for it?

Cheers

On 9/5/07, Alfred J Fazio <alf...@gm...> wrote:
>
> Hi everyone,
>
>     I am using SQLRelay 0.38 via Python (2.4.4) to connect to a PostgreSQL
> (8.1.9) database.  I have noticed that if I create a table with an ARRAY
> column, that Sqlrelay returns the column type as UNKNOWN.  E.g., if I
> issue the following SQL:
>
> CREATE TABLE test (arr TEXT[][]);
>
> And fill it will some test data:
>
> INSERT INTO test (arr) VALUES (ARRAY[['fname', 'alfred'], ['lname',
> 'fazio'], ['age', '24'], ['employer', 'smoothstone']]);
>
> I will get an UNKNOWN value when calling getColumnType on the arr column
> above:
>
> >>> c.sendQuery("SELECT arr FROM test")
> >>> c.getField(0, 0)
> '{{fname,alfred},{lname,fazio},{age,24},{employer,smoothstone}}'
> >>> c.getColumnType(0)
> 'UNKNOWN'
>
> It would be very convenient if I could test whether the column I am
> fetching is an ARRAY.  If so, I can automatically parse the result of the
> query into a native Python list.  I briefly peeked in the source code and
> saw that there has been a datatype declared for arrays for PostgreSQL in the
> src/common/datatypes.h file (called 'ANYARRAY').  However, I didn't look
> around for long to see where I might implement the code needed to detect the
> column type.  Could somebody please point me to the right direction within
> the source code where I can perhaps add a clause for this situation?  I will
> then be happy to submit a patch to the list.
>
> Thanks for your time!
>
> --
> Alfred J Fazio,
> alf...@gm...




-- 
Alfred J Fazio,
alf...@gm...

[Sqlrelay-discussion] getColumnType returns 'UNKNOWN' for PostgreSQL ARRAY column type.

[Alfred J F. <alf...@gm...>]

Hi everyone,

    I am using SQLRelay 0.38 via Python (2.4.4) to connect to a PostgreSQL (
8.1.9) database.  I have noticed that if I create a table with an ARRAY
column, that Sqlrelay returns the column type as UNKNOWN.  E.g., if I issue
the following SQL:

CREATE TABLE test (arr TEXT[][]);

And fill it will some test data:

INSERT INTO test (arr) VALUES (ARRAY[['fname', 'alfred'], ['lname',
'fazio'], ['age', '24'], ['employer', 'smoothstone']]);

I will get an UNKNOWN value when calling getColumnType on the arr column
above:

>>> c.sendQuery("SELECT arr FROM test")
>>> c.getField(0, 0)
'{{fname,alfred},{lname,fazio},{age,24},{employer,smoothstone}}'
>>> c.getColumnType(0)
'UNKNOWN'

It would be very convenient if I could test whether the column I am fetching
is an ARRAY.  If so, I can automatically parse the result of the query into
a native Python list.  I briefly peeked in the source code and saw that
there has been a datatype declared for arrays for PostgreSQL in the
src/common/datatypes.h file (called 'ANYARRAY').  However, I didn't look
around for long to see where I might implement the code needed to detect the
column type.  Could somebody please point me to the right direction within
the source code where I can perhaps add a clause for this situation?  I will
then be happy to submit a patch to the list.

Thanks for your time!

-- 
Alfred J Fazio,
alf...@gm...

Re: [Sqlrelay-discussion] =?GB2312?Q?=BB=D8=B8=B4: [Spam] =BB=D8=B8=B4=A3=BA I got the following error, any idea??=

[David M. <dav...@fi...>]

Aaah,  I recently added a call to that stored procedure to get the
version of the db server at startup.  Looks like sp_version is only
valid in sybase db's, not in MS SQL Server.  Odd that not finding it
causes the connection to eat up a bunch of memory.  I'll fix that though
and put out an update.  This looks like a good workaround though.

Dave
dav...@fi...

On Fri, 2007-08-31 at 23:47 +0800, wmins wrote:
> I got same problem as u, my solution is write 'sp_version' myself. The create sql is:
> 
> CREATE proc [dbo].[sp_version]
> @script varchar(50)=null,
> @type varchar(50)=null
> as
> select 'TopView'
> 
> return
> 
> By the way, please tell me the Memory use status when ur 'sqlr-connection-freetds ' started successfully
> 
> 
> 
> 
> -----Original message-----
> From: "邱 承" cnm...@ya...
> Date: Sat,  1 Sep 2007 12:38:45 +0800
> To: sqlrelay sql...@li...
> Subject: [Spam] [Sqlrelay-discussion]回复： I got the following error, any idea?
> 
> > The problem has been worked around. I add a stored procedure named sp_versi> on in the DB connected. That's all. I don't know why the sqlr-connect keeps>  try to call this stored procedure.  I guess 'sp_version' is a stored proce> dure in Sybase, but don't exist in SQL2k.
> 
> any idea?
> 
> 
> ----- ?gt; 始邮?----
> 发件人： "邱 承" <cnma> rk...@ya...>
> 收件人： sql...@li...> eforge.net; cnm...@ya...; 邱 承 <qiu...@cs...>
> ?gt; 逊⑺停?2007/8/28(周二), 下午6:17:52
> 主> 题： I got the following error, any idea?
> 
> 
> I used sql-relay > 0.39.2 to connect to a sql server 2k server.  The following is my sqlrelay.> conf:
> <instances>
>         <!-- Regular SQL Relay Instance -->
>        >  <instance id="example" port="9000" socket="/tmp/example.socket" dbas> e="f
> reetds" connections="3" maxconnections="15" maxqueuelength="> 5" growby="1" ttl="6
> 0" endofsession="commit" sessiontimeout="600> " runasuser="nobody" runasgroup="nob
> ody" cursors="5" authtier="l> istener" handoff="pass" deniedips="" allowedips="" d
> ebug="connec> tion" maxquerysize="65536" maxstringbindvaluelength="4000" maxlobbin
> > dvaluelength="71680" idleclienttimeout="-1" maxlisteners="-1" listene> rtimeout="0
> " reloginatstart="false">
>                 <users>
>    >                      <user user="user1" password="password1"/>
>       >                   <user user="user2" password="password2"/>
>          >                <user user="user3" password="password3"/>
>             >     </users>
>                 <connections>
>                         <con> nection connectionid="db1" string="sybase=/etc/freet
> ds;user=sa;p> assword=***;server=MyServer2k;db=run;" metric="1" behindloadbalan> 
> cer="no"/>
>                 </connections>
>         </instance>
>  > 
> Then, after started by "sqlr-start -id example", I got the following err> ors:
> Server Name:
> JF-WEBJY-TEST2
> Procedure Name:
> Error:  Could not > find stored procedure 'sp_version'.
> Couldn't create cursors.
> *** glibc > detected *** free(): invalid pointer: 0xb7fd0000 ***
>   sqlr-connection-fr> eetds -id example -connectionid db1 -config /usr/local/fi
> works/etc/sqlre> lay.conf
> Server message:
>         severity(10)
>         number(252)
>  >        state(62)
>         line(1)
>  
> Could anybody help me?
> 
> 
> 
> > 雅虎免费邮箱，全球第一邮?gt; 淦放疲?
> 
>       _______________________________________> ____________________ 
> 雅虎免费邮箱，全?gt; 虻谝挥氏淦放疲?
> http://mail.yahoo.com.cn/
> 
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________ Sqlrelay-discussion mailing list Sql...@li... https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion

[Sqlrelay-discussion] the efficiency of sqlrelay

[<tss...@16...>]

hi,all:  why sqlrelay query a intricate sql will so slow ?the same sql query :   php-sqlrelay:0.16688680648804s   php-oci func:0.046548843383789s    my PC:intel c4 2.4 with 512rammy OS:rh linux 9oracle 9iapache 2php4.4  when i try to user php with sqlrelay link this: select i.P_PRODUCTNAME as P_PRODUCTNAME , i.p_companyid , i.p_id as p_id , i.p_type , to_char( I.P_CREATETIME, 'yy-mm-dd hh24:mi:ss') as createtime , i.p_user as user_name , to_char( I.P_CHECKTIME , 'yy-mm-dd hh24:mi:ss') as checktime , to_char( I.p_lastupdatetime, 'yy-mm-dd hh24:mi:ss') as updatetime , i.p_admin , i.P_STATUS as sta , l.g_languagename , l.g_id , i.p_productimage from c_product i , g_language l where l.g_id=i.p_languageid and i.p_languageid=1 and I.P_CREATETIME>=to_date('2006-08-13 00:00:00','YYYY-MM-DD HH24:MI:SS') and I.P_CREATETIME<=to_date('2007-08-13 23:59:59','YYYY-MM-DD HH24:MI:SS') and i.p_productimage>0 order by I.P_CREATETIME desc the timer tell me:conn time: 0.00015997886657715
prase time: 0.16688680648804
disconn time: 0.00068283081054688
all time: 0.16797995567322 but,when i try php self oci functions to query the same sql:conn time: 0.046813011169434
prase time: 0.046548843383789
disconn time: 0.00010800361633301
all time: 0.093719959259033 and i try php with sqlrelay agein:select a_admin from g_admin the timer tell me:conn time: 0.00015783309936523
prase time: 0.014791011810303
disconn time: 0.00018692016601562
all time: 0.015383958816528 and i try php self oci function:select a_admin from g_admin the timer tell me:conn time: 0.048692941665649
prase time: 0.0018470287322998
disconn time: 9.1075897216797E-05
all time: 0.050878047943115  my sqlrelay.conf:<?xml version="1.0"?>
<!DOCTYPE instances SYSTEM "sqlrelay.dtd">
<instances>        <instance id="oracletest" port="9000" socket="/tmp/oracletest.socket" dbase="oracle8" connections="10" maxconnections="100" maxqueuelength="5" growby="2" ttl="60" endofsession="commit" sessiontimeout="600" runasuser="nobody" runasgroup="nobody" cursors="5" authtier="listener" handoff="pass" debug="none">
                <users>
                        <user user="oracletest" password="oracletest"/>
                </users>
                <connections>
                        <connection connectionid="oracletest" string="user=bigfish_test;password=bigfish_test;oracle_sid=bigfish;oracle_home=/opt/oracle/product/9i2" metric="1"/>
                </connections>
        </instance></instances>
     

[Sqlrelay-discussion] =?GB2312?Q?=BB=D8=B8=B4: [Spam] =BB=D8=B8=B4=A3=BA I got the following error, any idea??=

[wmins <wm...@gw...>]

I got same problem as u, my solution is write 'sp_version' myself. The create sql is:

CREATE proc [dbo].[sp_version]
@script varchar(50)=null,
@type varchar(50)=null
as
select 'TopView'

return

By the way, please tell me the Memory use status when ur 'sqlr-connection-freetds ' started successfully




-----Original message-----
From: "邱 承" cnm...@ya...
Date: Sat,  1 Sep 2007 12:38:45 +0800
To: sqlrelay sql...@li...
Subject: [Spam] [Sqlrelay-discussion]回复： I got the following error, any idea?

> The problem has been worked around. I add a stored procedure named sp_versi> on in the DB connected. That's all. I don't know why the sqlr-connect keeps>  try to call this stored procedure.  I guess 'sp_version' is a stored proce> dure in Sybase, but don't exist in SQL2k.

any idea?


----- ?gt; 加始?----
发件人： "邱 承" <cnma> rk...@ya...>
收件人： sql...@li...> eforge.net; cnm...@ya...; 邱 承 <qiu...@cs...>
?gt; 逊⑺停?2007/8/28(周二), 下午6:17:52
主> 题： I got the following error, any idea?


I used sql-relay > 0.39.2 to connect to a sql server 2k server.  The following is my sqlrelay.> conf:
<instances>
        <!-- Regular SQL Relay Instance -->
       >  <instance id="example" port="9000" socket="/tmp/example.socket" dbas> e="f
reetds" connections="3" maxconnections="15" maxqueuelength="> 5" growby="1" ttl="6
0" endofsession="commit" sessiontimeout="600> " runasuser="nobody" runasgroup="nob
ody" cursors="5" authtier="l> istener" handoff="pass" deniedips="" allowedips="" d
ebug="connec> tion" maxquerysize="65536" maxstringbindvaluelength="4000" maxlobbin
> dvaluelength="71680" idleclienttimeout="-1" maxlisteners="-1" listene> rtimeout="0
" reloginatstart="false">
                <users>
   >                      <user user="user1" password="password1"/>
      >                   <user user="user2" password="password2"/>
         >                <user user="user3" password="password3"/>
            >     </users>
                <connections>
                        <con> nection connectionid="db1" string="sybase=/etc/freet
ds;user=sa;p> assword=***;server=MyServer2k;db=run;" metric="1" behindloadbalan> 
cer="no"/>
                </connections>
        </instance>
 > 
Then, after started by "sqlr-start -id example", I got the following err> ors:
Server Name:
JF-WEBJY-TEST2
Procedure Name:
Error:  Could not > find stored procedure 'sp_version'.
Couldn't create cursors.
*** glibc > detected *** free(): invalid pointer: 0xb7fd0000 ***
  sqlr-connection-fr> eetds -id example -connectionid db1 -config /usr/local/fi
works/etc/sqlre> lay.conf
Server message:
        severity(10)
        number(252)
 >        state(62)
        line(1)
 
Could anybody help me?



> 雅虎免费邮箱，全球第一邮?gt; 淦放疲?

      _______________________________________> ____________________ 
雅虎免费邮箱，全?gt; 虻谝挥氏淦放疲?
http://mail.yahoo.com.cn/

[Sqlrelay-discussion] =?gb2312?B?u9i4tKO6IEkgZ290IHRoZSBmb2xsb3dpbmcgZXJyb3IsIGFueSBpZGVhPw==?=

[&#37041; &#25215; <cnm...@ya...>]

The problem has been worked around. I add a stored procedure named sp_versi=
on in the DB connected. That's all. I don't know why the sqlr-connect keeps=
 try to call this stored procedure.  I guess 'sp_version' is a stored proce=
dure in Sybase, but don't exist in SQL2k.=0A=0Aany idea?=0A=0A=0A----- =D4=
=AD=CA=BC=D3=CA=BC=FE ----=0A=B7=A2=BC=FE=C8=CB=A3=BA "=C7=F1 =B3=D0" <cnma=
rk...@ya...>=0A=CA=D5=BC=FE=C8=CB=A3=BA sql...@li...=
eforge.net; cnm...@ya...; =C7=F1 =B3=D0 <qiu...@cs...>=0A=D2=
=D1=B7=A2=CB=CD=A3=BA 2007/8/28(=D6=DC=B6=FE), =CF=C2=CE=E76:17:52=0A=D6=F7=
=CC=E2=A3=BA I got the following error, any idea?=0A=0A=0AI used sql-relay =
0.39.2 to connect to a sql server 2k server.  The following is my sqlrelay.=
conf:=0A<instances>=0A        <!-- Regular SQL Relay Instance -->=0A       =
 <instance id=3D"example" port=3D"9000" socket=3D"/tmp/example.socket" dbas=
e=3D"f=0Areetds" connections=3D"3" maxconnections=3D"15" maxqueuelength=3D"=
5" growby=3D"1" ttl=3D"6=0A0" endofsession=3D"commit" sessiontimeout=3D"600=
" runasuser=3D"nobody" runasgroup=3D"nob=0Aody" cursors=3D"5" authtier=3D"l=
istener" handoff=3D"pass" deniedips=3D"" allowedips=3D"" d=0Aebug=3D"connec=
tion" maxquerysize=3D"65536" maxstringbindvaluelength=3D"4000" maxlobbin=0A=
dvaluelength=3D"71680" idleclienttimeout=3D"-1" maxlisteners=3D"-1" listene=
rtimeout=3D"0=0A" reloginatstart=3D"false">=0A                <users>=0A   =
                     <user user=3D"user1" password=3D"password1"/>=0A      =
                  <user user=3D"user2" password=3D"password2"/>=0A         =
               <user user=3D"user3" password=3D"password3"/>=0A            =
    </users>=0A                <connections>=0A                        <con=
nection connectionid=3D"db1" string=3D"sybase=3D/etc/freet=0Ads;user=3Dsa;p=
assword=3D***;server=3DMyServer2k;db=3Drun;" metric=3D"1" behindloadbalan=
=0Acer=3D"no"/>=0A                </connections>=0A        </instance>=0A =
=0AThen, after started by "sqlr-start -id example", I got the following err=
ors:=0AServer Name:=0AJF-WEBJY-TEST2=0AProcedure Name:=0AError:  Could not =
find stored procedure 'sp_version'.=0ACouldn't create cursors.=0A*** glibc =
detected *** free(): invalid pointer: 0xb7fd0000 ***=0A  sqlr-connection-fr=
eetds -id example -connectionid db1 -config /usr/local/fi=0Aworks/etc/sqlre=
lay.conf=0AServer message:=0A        severity(10)=0A        number(252)=0A =
       state(62)=0A        line(1)=0A =0ACould anybody help me?=0A=0A=0A=0A=
=D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E4=A3=AC=C8=AB=C7=F2=B5=DA=D2=BB=D3=CA=CF=
=E4=C6=B7=C5=C6=A3=A1=0A=0A=0A      _______________________________________=
____________________ =0A=D1=C5=BB=A2=C3=E2=B7=D1=D3=CA=CF=E4=A3=AC=C8=AB=C7=
=F2=B5=DA=D2=BB=D3=CA=CF=E4=C6=B7=C5=C6=A3=A1 =0Ahttp://mail.yahoo.com.cn/

[Sqlrelay-discussion] Why the Opened Server Connections keep increasing?

[Zhuguo S. <blu...@gm...>]

When I use the SQL 'set names utf8' through sqlrealy, sometimes sqlrelay
doesn't response. I use 'sqlr-status' command to check what's going on, the
Opened Server Connections are keeping incrsasing. Some one can help me?

and I use 'sqlrcon_ping' to test if the connection is established, is that
right?

Re: [Sqlrelay-discussion] Bug report, empty cursor/blob, with code to reproduce problem

[David M. <dav...@fi...>]

Wow, that's a strange bug!

Using your code I am able to reproduce it here.  If I turn debug on, I
see some strange behavior when fetching from the bind cursor, so I
should be able to track down what's happening.

I'll let you know what I find.  That sure is odd.

Dave
dav...@fi...

On Thu, 2007-08-30 at 11:30 +0200, Ingmar Brouns wrote:
> BUG report, empty cursor/blob, with code to reproduce problem
> 
> sqlrelay 0.39.2
> Oracle 10g
> php 4.3.11
> 
> I've encountered a bug that causes blobs or cursors to be empty
> when they shouldn't be.  I've made an example which uses only oracle 
> functions so that it will be easy for you to verify/reproduce the
> bug. 
> I've encountered this bug several times, and in all these cases it
> happened
> two steps. First an oracle procedure with multiple out parameters was
> called. 
> And after that, a procedure was called which produces a reference
> cursor
> or a blob. When the reference cursor of the blob is read out, it
> produces
> the right results or nothing depending on the order in which the
> variables 
> are bound in the first procedure. By running the following code you
> will
> see what I mean. The code does two things. First a query is executed 
> that does a call to dbms_output.get_line. The corresponding cursor is 
> freed afterwards. Then another query is performed which opens a cursor
> for "SELECT * FROM DUAL". The rowcount for this cursor is echod,
> this should always display one. However, by changing the order of the
> binds 
> in the first query, the cursor is suddenly empty. The same thing
> happens when
> the second query reads out a blob. I suspect that something goes wrong
> with the freeing of resources, as in some more complex situations I
> believe 
> this problem has led to clients keeping connections while they should
> be
> freed.
> 
>     //creates a new connection which is stored in the variable $conn
>     //make_new_connection(.....);
>          
>     $sql = "BEGIN dbms_output.get_line(line => :line_out, status
> => :status_out);  END;"; 
>     $cur = sqlrcur_alloc($conn);
>     sqlrcur_prepareQuery($cur, $sql);                          
>       
>     /* In this situation, the cursor which references DUAL 
>      * in the next query will have a rowcount of 1, which is correct. 
>      * When the order of these two binds is switched, then the
>      * the cursor referencing DUAL will have a rowcount of 0 !!!
>      */ 
>     sqlrcur_defineOutputBindInteger($cur,"status_out");
>     sqlrcur_defineOutputBindString($cur,"line_out", 255); 
>         
>     sqlrcur_executeQuery($cur);        
>     $line = sqlrcur_getOutputBindString( $cur,
> "line_out" );                 
>     $status = sqlrcur_getOutputBindInteger( $cur, "status_out" ); 
>     sqlrcur_free($cur);
> 
>     /*
>      * This query declares a reference cursor which is opened
>      * for SELECT * FROM DUAL, so it's opened for DUAL. The
>      * reference cursor is bound to the out parameter curs. 
>      * Hence, echo sqlrcur_rowCount($cursor); should always display 1
>      * as SELECT * FROM DUAL will give 1 row
>      */
>     $sql = "DECLARE TYPE curtype IS REF CURSOR; ref_cur curtype; BEGIN
> OPEN ref_cur FOR SELECT * FROM DUAL; :curs := ref_cur ; END;"; 
>     $cur = sqlrcur_alloc($conn);
>     sqlrcur_prepareQuery($cur,
> $sql);                                          
>     sqlrcur_defineOutputBindCursor($cur,"curs");
>     sqlrcur_executeQuery($cur);             
>     $cursor = sqlrcur_getOutputBindCursor( $cur, "curs" );   
>     sqlrcur_fetchFromBindCursor($cursor);   
>    
>     echo sqlrcur_rowCount($cursor);     
>    
>     sqlrcur_free($cursor);       
>     sqlrcur_free($cur); 
>     sqlrcon_free($conn);   
> 
> I would very much like to hear whether the above code has helped
> to verify what I've discribed.
> 
> 
> Yours sincerely,
> 
> Ingmar Brouns
> 
>      
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________ Sqlrelay-discussion mailing list Sql...@li... https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion

Re: [Sqlrelay-discussion] dynamically adding connections

[Duncan M. <dwm...@gm...>]

Thank you - that is what I was looking for.

On 8/30/07, David Muse <dav...@fi...> wrote:
> There is a dynamic scaling option.  In the sqlrelay.conf file, the
> "connections" parameter defines the number of db connections that start
> up when sqlrelay starts.  There's another parameter called
> "maxconnections".  If it's greater than "connections", then new
> connections will start up as-needed.  The "ttl" parameter determines how
> long they will hang around if they go unused.  The "maxqueuelength"
> parameter determines how many waiting clients must be queued up before
> new connections will be started.  The "growby" parameter determines how
> many new connections will be started when the maxqueuelength is
> exceeded.
>
> I believe that you can even set connections="0" and no connections will
> be started when sqlrelay is started up.  In effect, it will be entirely
> dynamic.
>
> You should also be able to manually start sqlr-connection deamons if you
> like, just by running them from the command line.  They will register
> themselves with the sqlr-listener process and service requests just as
> if they had been started by sqlr-start.
>
> Hope this helps.
>
> David Muse
> dav...@fi...
>
> On Mon, 2007-08-27 at 22:55 -0500, Duncan McQueen wrote:
> > I looked thru the documentation and it appears that you define all the
> > database connections at startup.  Is there a way to dynamically add
> > new connections after sqlrelay has started?
> >
> > -------------------------------------------------------------------------
> > This SF.net email is sponsored by: Splunk Inc.
> > Still grepping through log files to find problems?  Stop.
> > Now Search log events and configuration files using AJAX and a browser.
> > Download your FREE copy of Splunk now >>  http://get.splunk.com/
> > _______________________________________________
> > Sqlrelay-discussion mailing list
> > Sql...@li...
> > https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion
> >
> >
> > _________________________________________________________________
> > Need personalized email and website? Look no further. It's easy
> > with Doteasy $0 Web Hosting! Learn more at www.doteasy.com
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________
> Sqlrelay-discussion mailing list
> Sql...@li...
> https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion
>

Re: [Sqlrelay-discussion] dynamically adding connections

[David M. <dav...@fi...>]

There is a dynamic scaling option.  In the sqlrelay.conf file, the
"connections" parameter defines the number of db connections that start
up when sqlrelay starts.  There's another parameter called
"maxconnections".  If it's greater than "connections", then new
connections will start up as-needed.  The "ttl" parameter determines how
long they will hang around if they go unused.  The "maxqueuelength"
parameter determines how many waiting clients must be queued up before
new connections will be started.  The "growby" parameter determines how
many new connections will be started when the maxqueuelength is
exceeded.

I believe that you can even set connections="0" and no connections will
be started when sqlrelay is started up.  In effect, it will be entirely
dynamic.

You should also be able to manually start sqlr-connection deamons if you
like, just by running them from the command line.  They will register
themselves with the sqlr-listener process and service requests just as
if they had been started by sqlr-start.

Hope this helps.

David Muse
dav...@fi...

On Mon, 2007-08-27 at 22:55 -0500, Duncan McQueen wrote:
> I looked thru the documentation and it appears that you define all the
> database connections at startup.  Is there a way to dynamically add
> new connections after sqlrelay has started?
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________
> Sqlrelay-discussion mailing list
> Sql...@li...
> https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion
> 
> 
> _________________________________________________________________
> Need personalized email and website? Look no further. It's easy
> with Doteasy $0 Web Hosting! Learn more at www.doteasy.com

[Sqlrelay-discussion] Bug report, empty cursor/blob, with code to reproduce problem

[Ingmar B. <sw...@gm...>]

BUG report, empty cursor/blob, with code to reproduce problem

sqlrelay 0.39.2
Oracle 10g
php 4.3.11

I've encountered a bug that causes blobs or cursors to be empty
when they shouldn't be.  I've made an example which uses only oracle
functions so that it will be easy for you to verify/reproduce the bug.
I've encountered this bug several times, and in all these cases it happened
two steps. First an oracle procedure with multiple out parameters was
called.
And after that, a procedure was called which produces a reference cursor
or a blob. When the reference cursor of the blob is read out, it produces
the right results or nothing depending on the order in which the variables
are bound in the first procedure. By running the following code you will
see what I mean. The code does two things. First a query is executed
that does a call to dbms_output.get_line. The corresponding cursor is
freed afterwards. Then another query is performed which opens a cursor
for "SELECT * FROM DUAL". The rowcount for this cursor is echod,
this should always display one. However, by changing the order of the binds
in the first query, the cursor is suddenly empty. The same thing happens
when
the second query reads out a blob. I suspect that something goes wrong
with the freeing of resources, as in some more complex situations I believe
this problem has led to clients keeping connections while they should be
freed.

    //creates a new connection which is stored in the variable $conn
    //make_new_connection(.....);

    $sql = "BEGIN dbms_output.get_line(line => :line_out, status =>
:status_out);  END;";
    $cur = sqlrcur_alloc($conn);
    sqlrcur_prepareQuery($cur, $sql);

    /* In this situation, the cursor which references DUAL
     * in the next query will have a rowcount of 1, which is correct.
     * When the order of these two binds is switched, then the
     * the cursor referencing DUAL will have a rowcount of 0 !!!
     */
    sqlrcur_defineOutputBindInteger($cur,"status_out");
    sqlrcur_defineOutputBindString($cur,"line_out", 255);

    sqlrcur_executeQuery($cur);
    $line = sqlrcur_getOutputBindString( $cur, "line_out" );

    $status = sqlrcur_getOutputBindInteger( $cur, "status_out" );
    sqlrcur_free($cur);

    /*
     * This query declares a reference cursor which is opened
     * for SELECT * FROM DUAL, so it's opened for DUAL. The
     * reference cursor is bound to the out parameter curs.
     * Hence, echo sqlrcur_rowCount($cursor); should always display 1
     * as SELECT * FROM DUAL will give 1 row
     */
    $sql = "DECLARE TYPE curtype IS REF CURSOR; ref_cur curtype; BEGIN OPEN
ref_cur FOR SELECT * FROM DUAL; :curs := ref_cur ; END;";
    $cur = sqlrcur_alloc($conn);
    sqlrcur_prepareQuery($cur,
$sql);
    sqlrcur_defineOutputBindCursor($cur,"curs");
    sqlrcur_executeQuery($cur);
    $cursor = sqlrcur_getOutputBindCursor( $cur, "curs" );
    sqlrcur_fetchFromBindCursor($cursor);

    echo sqlrcur_rowCount($cursor);

    sqlrcur_free($cursor);
    sqlrcur_free($cur);
    sqlrcon_free($conn);

I would very much like to hear whether the above code has helped
to verify what I've discribed.


Yours sincerely,

Ingmar Brouns

[Sqlrelay-discussion] dynamically adding connections

[Duncan M. <dwm...@gm...>]

I looked thru the documentation and it appears that you define all the
database connections at startup.  Is there a way to dynamically add
new connections after sqlrelay has started?

[Sqlrelay-discussion] Session Count Problem (Router)

[Chris C. <cc...@gm...>]
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Re: [Sqlrelay-discussion] ./sqlr-start -id oracletest with Error

[<tss...@16...>]

well!thx!



=E5=9C=A82007-08-13=EF=BC=8C"David Muse" <dav...@fi...> =E5=86=
=99=E9=81=93=EF=BC=9A
>"oracle8" just refers to the version of OCI that the code uses.
>
>oracle8 =3D OCI8
>
>OCI8 is supported by oracle 8,8i,9i and 10g.
>
>I used to have an oracle7 connection which used OCI7.   I added
>"oracle8" later when oracle stopped supporting OCI7.
>
>
>On Fri, 2007-08-10 at 16:45 +0800, =E5=AD=99=E5=BA=9A wrote:
>&gt; i modified the sqlrelay.conf:
>&gt;=20
>&gt; dbase=3D"oracle8" the part and it is running.
>&gt;=20
>&gt; but ,setup like this ,can it working well with oracle9i2?
>&gt;=20
>&gt;=20
>&gt;=20
>&gt; _____________________________________________________________________=
_
>&gt; =E5=8A=B2 =E7=88=86 150 =E4=B8=87 =E5=90=8C =E6=97=B6 =E5=9C=A8 =E7=
=BA=BF=EF=BC=8C=E4=BC=97 =E4=BA=BA =E8=BF=BD =E6=8D=A7 =E6=A2=A6 =E5=B9=BB =
=E8=A5=BF =E6=B8=B8=20
>&gt; ---------------------------------------------------------------------=
----
>&gt; This SF.net email is sponsored by: Splunk Inc.
>&gt; Still grepping through log files to find problems?  Stop.
>&gt; Now Search log events and configuration files using AJAX and a browse=
r.
>&gt; Download your FREE copy of Splunk now &gt;&gt;  http://get.splunk.com=
/
>&gt; _______________________________________________ Sqlrelay-discussion m=
ailing list Sql...@li... https://lists.sourcef=
orge.net/lists/listinfo/sqlrelay-discussion
>
>
>-------------------------------------------------------------------------
>This SF.net email is sponsored by: Splunk Inc.
>Still grepping through log files to find problems?  Stop.
>Now Search log events and configuration files using AJAX and a browser.
>Download your FREE copy of Splunk now &gt;&gt;  http://get.splunk.com/
>_______________________________________________
>Sqlrelay-discussion mailing list
>Sql...@li...
>https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion
>

Re: [Sqlrelay-discussion] Build error of 0.39.2

[Stephen B. <sql...@st...>]

Henning Markussen wrote:
> Hi
>
> I'm trying to build 0.39.2 on a RHEL3ES
>
> But it fail's and I need a tip to what it is.
>
>   
My first thing to check is that you have the latest version of
Rudiments installed..


Stephen Barclay
www.Call-Em-All.com

[Sqlrelay-discussion] Build error of 0.39.2

[Henning M. <hm...@mi...>]

Hi

I'm trying to build 0.39.2 on a RHEL3ES

But it fail's and I need a tip to what it is.


***** Summary ***********************************************
  Version      : 0.39.2

  APIs         : C/C++       yes           Perl       yes
                 Python      yes           Ruby       yes
                 PHP         yes           Java       no
                 TCL         yes           Zope       no

  Connections  : Oracle8     no            MySQL      no
                 PostgreSQL  dynamic       SQLite     dynamic
                 FreeTDS     no            Sybase     no
                 ODBC        no            DB2        dynamic
                 Firebird    no            MDB Tools  no
*************************************************************

If you expected a Connection or API that doesn't show up in the Summary
then the configure script probably couldn't find a package it needed to
build it.  You can manually specify package locations using command line
options.

Type:  configure --help   for a list of options.


Thanks to MP3.com for sponsoring:
         Clustered/Replicated database support.
         Perl API.
Thanks to FeedLounge for sponsoring:
         Query routing and filtering.

[root@xxx sqlrelay-0.39.2]# make
make -C src all
make[1]: Entering directory `/usr/src/redhat/BUILD/sqlrelay-0.39.2/src'
make -C util all
make[2]: Entering directory `/usr/src/redhat/BUILD/sqlrelay-0.39.2/src/util'
/bin/sh ../../libtool --mode=compile g++ -Wall -pipe  -pthread
-DTMP_DIR=\"/usr/local/firstworks/var/sqlrelay/tmp\"
-DDEBUG_DIR=\"/usr/local/firstworks/var/sqlrelay/debug\"
-DDEFAULT_CONFIG_FILE=\"/usr/local/firstworks/etc/sqlrelay.conf\" -I./
-I../../ -I../../src/common  -c cmdline.C -o cmdline.lo
  g++ -Wall -pipe -pthread
-DTMP_DIR=\"/usr/local/firstworks/var/sqlrelay/tmp\"
-DDEBUG_DIR=\"/usr/local/firstworks/var/sqlrelay/debug\"
-DDEFAULT_CONFIG_FILE=\"/usr/local/firstworks/etc/sqlrelay.conf\" -I./
-I../../ -I../../src/common -c cmdline.C  -fPIC -DPIC -o .libs/cmdline.o
cmdline.C: In member function `void cmdline::setId()':
cmdline.C:20: `getValue' undeclared (first use this function)
cmdline.C:20: (Each undeclared identifier is reported only once for each
    function it appears in.)
make[2]: *** [cmdline.lo] Error 1
make[2]: Leaving directory `/usr/src/redhat/BUILD/sqlrelay-0.39.2/src/util'
make[1]: *** [all] Error 2
make[1]: Leaving directory `/usr/src/redhat/BUILD/sqlrelay-0.39.2/src'
make: *** [all] Error 2

- Henning

[Sqlrelay-discussion] Reading out blobs

[Ingmar B. <sw...@gm...>]

Dear fellow mailinglist members,

I am having trouble reading out a blob with sql-relay. I use sql-relay
version 0.39.2
and oracle 10g .

The code I used comes directly from the sqlrelay manual:

    $query = "-- debug\n BEGIN select large_value into :large_value
                from wf_attribute_instances i
               where i.atri_id = 6011
                 and i.prin_id = 2090477 ; END;";


    $cur = sqlrcur_alloc($conn);

     sqlrcur_prepareQuery($cur,$query);
     sqlrcur_defineOutputBindBlob($cur,"large_value");
     sqlrcur_executeQuery($cur);
     $large_value=sqlrcur_getOutputBindBlob($cur,"large_value");
     $length=sqlrcur_getOutputBindLength($cur,"large_value");
     sqlrcon_endSession($conn);
     echo "<br>waarde blob=".$large_value;
     echo "<br>lengte blob=".$length;
     sqlrcur_free($cur);

The result ($large_value) was empty. So we turned on debugging:
sqlr-connection-oracle8-debug, to locate
the problem. And now, with debugging, suddenly the result is not empty
anymore. Below you will find
the query debug (-- debug\n) information with and without debugging. Has
anyone encountered this problem ?
I would greatly appreciate any suggestions/ advice :-)

output without debugging:


Sending Query:
-- debug
 BEGIN select large_value into :large_value
                from wf_attribute_instances i
               where i.atri_id = 6011
                 and i.prin_id = 2090477 ; END;
Length: %ld
Requesting a new cursor.
Sending %ld Input Bind Variables:
Sending Output Bind Variables:
large_value(%ld)
Send Column Info: yes
Skipping and Fetching
Skipping %ld rows
Fetching %ld rows
Checking For An Error... none.
Getting Cursor ID...
Cursor ID: %ld
Previous result set was not suspended.
Parsing Column Info
Actual row count: unknown
Affected row count: %ld
Column count: %ld
Receiving Output Bind Values:
    getting type...
    done getting type: %ld
    NULL output bind
        done fetching.
large_value=


    getting type...
    done getting type: %ld
Parsing Data
Got end of result set.
Ending Session
Deallocated cursor
Result Set Buffer Size: %ld
Connecting to listener...
Unix socket: /tmp/MYDATABASE.sqlr_socket
Authenticating with listener : MYDATABASE:MYDATABASE
Waiting for auth success/failure...
No authentication error.
Must Reconnect.
Reconnecting to
    unix port: /usr/local/firstworks/var/sqlrelay/tmp/sockets/4662
Connected.
Authenticating with connection : MYDATABASE:MYDATABASE
Waiting for auth success/failure...
No authentication error.
Sending Query:
BEGIN dbms_output.get_line( line => :line, status => :status ); END;
Length: %ld
Requesting a new cursor.
Sending %ld Input Bind Variables:
Sending Output Bind Variables:
line(%ld)
status(%ld)
Send Column Info: yes
Skipping and Fetching
    row to get: %ld
Skipping %ld rows
Fetching %ld rows
Checking For An Error... none.
Getting Cursor ID...
Cursor ID: %ld
Previous result set was not suspended.
Parsing Column Info
Actual row count: unknown
Affected row count: %ld
Column count: %ld
Receiving Output Bind Values:
    getting type...
    done getting type: %ld
    NULL output bind
        done fetching.
line=
    getting type...
    done getting type: %ld
    STRING output bind
        length=%ld
        done fetching
status=1
    getting type...
    done getting type: %ld
Parsing Data
Got end of result set.
Creating Column Arrays...
Aborting Result Set For Cursor: %ld
Deallocated cursor
Ending Session
Deallocated connection

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

With debugging




Sending Query:
-- debug
 BEGIN select large_value into :large_value
                from wf_attribute_instances i
               where i.atri_id = 6011
                 and i.prin_id = 2090477 ; END;
Length: %ld
Requesting a new cursor.
Sending %ld Input Bind Variables:
Sending Output Bind Variables:
large_value(%ld)
Send Column Info: yes
Skipping and Fetching
Skipping %ld rows
Fetching %ld rows
Checking For An Error... none.
Getting Cursor ID...
Cursor ID: %ld
Previous result set was not suspended.
Parsing Column Info
Actual row count: unknown
Affected row count: %ld
Column count: %ld
Receiving Output Bind Values:
    getting type...
    done getting type: %ld
    LOB/CLOB output bind
        length=%ld
        fetching...
        fetching...
        fetching...
        done fetching.
large_value=
a:16:{s:5:"dummy";s:0:"";s:3:"sel";s:416:"238c3e4a29aaa449c9e2fd8e27c6e84a3ab18f
7198951473df85ae9448e109c0e1aba818742d9cb084a2d2ca273178b5e9272f381cfb8a2d4e6e37
a29a96fcb5a1e3601dedf4b46d44f003b2a384d5795e4e3ea500bad7c767e1f3d54c4c69b1dfce00
dcc6d189fd7edcd8cf63f50a0fb85a0ea98b39360fd2c4da1ee505f59a1611d65e7c9212a5fb3f5e
e295765cf7b7414a32ba5268227bee09a6211b534a99221db576ffba91e21ce248c428ba226f9a8e
e9a412c77eb122804511b189ba7b4259132ea2b8e0d09a0577c5435297";s:6:"act_cd";s:9:"17
0777000";s:12:"act_class_cd";s:9:"ORGANIZER";s:13:"vraagstelling";s:12:".eoeuoeu
oaeu";s:5:"datum";s:10:"06-07-2007";s:11:"streefdagen";s:2:"30";s:7:"ind_via";s:
1:"I";s:4:"orga";s:28:"HUISARTSENPRAKTIJK Het
Viooltje";s:7:"orga_id";s:4:"6251";s:
10:"ind_inform";s:1:"J";s:13:"ind_inform_me";s:1:"J";s:14:"inform_asen_id";s:0:"
";s:14:"inform_pati_id";s:0:"";s:23:"ind_vervolg_actie_enter";s:1:"0";s:6:"Submi
t";s:5:"Enter";}

    getting type...
    done getting type: %ld
Parsing Data
Got end of result set.
Ending Session
Deallocated cursor
Result Set Buffer Size: %ld
Connecting to listener...
Unix socket: /tmp/MYDATABASE.sqlr_socket
Authenticating with listener : MYDATABASE:MYDATABASE
Waiting for auth success/failure...
No authentication error.
Must Reconnect.
Reconnecting to
    unix port: /usr/local/firstworks/var/sqlrelay/tmp/sockets/2153
Connected.
Authenticating with connection : MYDATABASE:MYDATABASE
Waiting for auth success/failure...
No authentication error.
Sending Query:
BEGIN dbms_output.get_line( line => :line, status => :status ); END;
Length: %ld
Requesting a new cursor.
Sending %ld Input Bind Variables:
Sending Output Bind Variables:
line(%ld)
status(%ld)
Send Column Info: yes
Skipping and Fetching
    row to get: %ld
Skipping %ld rows
Fetching %ld rows
Checking For An Error... none.
Getting Cursor ID...
Cursor ID: %ld
Previous result set was not suspended.
Parsing Column Info
Actual row count: unknown
Affected row count: %ld
Column count: %ld
Receiving Output Bind Values:
    getting type...
    done getting type: %ld
    NULL output bind
        done fetching.
line=
    getting type...
    done getting type: %ld
    STRING output bind
        length=%ld
        done fetching
status=1
    getting type...
    done getting type: %ld
Parsing Data
Got end of result set.
Creating Column Arrays...
Aborting Result Set For Cursor: %ld
Deallocated cursor
Ending Session
Deallocated connection


Thnx in advance,

Ingmar

Re: [Sqlrelay-discussion] ./sqlr-start -id oracletest with Error

[David M. <dav...@fi...>]

"oracle8" just refers to the version of OCI that the code uses.

oracle8 = OCI8

OCI8 is supported by oracle 8,8i,9i and 10g.

I used to have an oracle7 connection which used OCI7.   I added
"oracle8" later when oracle stopped supporting OCI7.


On Fri, 2007-08-10 at 16:45 +0800, 孙庚 wrote:
> i modified the sqlrelay.conf:
> 
> dbase="oracle8" the part and it is running.
> 
> but ,setup like this ,can it working well with oracle9i2?
> 
> 
> 
> ______________________________________________________________________
> 劲 爆 150 万 同 时 在 线，众 人 追 捧 梦 幻 西 游 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________ Sqlrelay-discussion mailing list Sql...@li... https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion

Re: [Sqlrelay-discussion] ./sqlr-start -id oracletest with Error

[<tss...@16...>]

i modified the sqlrelay.conf:

dbase="oracle8" the part and it is running.

but ,setup like this ,can it working well with oracle9i2?

Re: [Sqlrelay-discussion] ./sqlr-start -id oracletest with Error

[<boc...@du...>]

It seems you don=A1=AFt have SQLRelay in the default location.

Then just add the bin directory of your sqlrelay to PATH variable.

=20

As sqlr-connection-oracle-debug is not in your binary directory. You may =
try
set debug=3D=A1=B1none=A1=B1 in your sqlrelay.conf and it will use
sqlr-connection-oracle (no debug suffix) instead. I suppose there must =
be
this file.

Any way, If this program  is also missed, You should confirm your
installation.

=20

=20

From: sql...@li...
[mailto:sql...@li...] On Behalf Of =
??
Sent: Friday, August 10, 2007 4:13 PM
To: Discussion of topics related to SQL Relay
Subject: Re: [Sqlrelay-discussion] ./sqlr-start -id oracletest with =
Error

=20

hi, i come from china hehe .
i have add /usr/local/firstworks/bin/ to your PATH environment variable.
but ,look here:
sh: line 1: sqlr-connection-oracle-debug: command not found
there was no file named sqlr-connection-oracle-debug in
/usr/local/sqlrelay/bin



=D4=...@du... =D0=B4=B5=C0=A3=BA
>
>
>
>
>You should add /usr/local/firstworks/bin/ to your PATH environment
>variable.
>
>=20
>
>=20
>
>
>
>From:
>sql...@li...
>[mailto:sql...@li...] On Behalf Of =
??
>Sent: Friday, August 10, 2007 3:15 PM
>To: Sql...@li...
>Subject: [Sqlrelay-discussion] ./sqlr-start -id oracletest with Error
>
>
>
>=20
>
>hi,all:
>my sqlrelay.conf:
><?xml version=3D"1.0"?>
><!DOCTYPE instances SYSTEM "sqlrelay.dtd">
><instances>
>
> <instance
>id=3D"oracletest" port=3D"9000"
>socket=3D"/tmp/oracletest.socket" dbase=3D"oracle"
>connections=3D"3" maxconnections=3D"5"
>maxqueuelength=3D"0" growby=3D"1" ttl=3D"60" endofsession=3D"commit"
>sessiontimeout=3D"600" runasuser=3D"nobody"
>runasgroup=3D"nobody" cursors=3D"5" authtier=3D"connection"
>handoff=3D"pass" debug=3D"listener_and_connection">
>=20
><users>
>=20
><user user=3D"oracletest" password=3D"oracletest"/>
>=20
></users>
>=20
><connections>
>=20
><connection connectionid=3D"oracletest"
>string=3D"user=3DUSER;password=3DPASSWORD;oracle_sid=3DORACLE_SID;oracle=
_home=3DORACL
E_HOME;nls_lang=3DNLS_LANG;autocommit=3Dyes/no;fetchatonce=3DFETCHATONCE;=
maxselect
listsize=3DMAXSELECTLISTSIZE;maxitembuffersize=3DMAXITEMBUFFERSIZE"
>metric=3D"1"/>
>=20
></connections>
> </instance>
>
></instances>
>
>
>and i run it:
>[root@bigfish bin]# ./sqlr-start -id oracletest
>
>Starting listener:
> sqlr-listener-debug -id oracletest -config
>/usr/local/sqlrelay/etc/sqlrelay.conf &
>sh: line 1: sqlr-listener-debug: command not found
>
>Starting 3 connections to oracletest :
> sqlr-connection-oracle-debug -id oracletest -connectionid oracletest
>-config /usr/local/sqlrelay/etc/sqlrelay.conf &
>sh: line 1: sqlr-connection-oracle-debug: command not found
> sqlr-connection-oracle-debug -id oracletest -connectionid oracletest
>-config /usr/local/sqlrelay/etc/sqlrelay.conf &
>sh: line 1: sqlr-connection-oracle-debug: command not found
> sqlr-connection-oracle-debug -id oracletest -connectionid oracletest
>-config /usr/local/sqlrelay/etc/sqlrelay.conf &
>sh: line 1: sqlr-connection-oracle-debug: command not found
>
>Starting scaler:
> sqlr-scaler -id oracletest -debug -config
>/usr/local/sqlrelay/etc/sqlrelay.conf
>sh: line 1: sqlr-scaler: command not found
>
>sqlr-scaler failed to start.
>
>
>Thanks to MP3.com for sponsoring:
> Clustered/Replicated database
>support.
> Perl API.
>Thanks to FeedLounge for sponsoring:
> Query routing and filtering.
>[root@bigfish bin]#
>
>
>why?
>
>
>
>=20
>
>
>
>
>
>=20
>
>
>
>
>
>=20
>
>
>
>
>
>=20
>
>
>
>=20
>
>
>
>
>
>
>
>=C4=E3 =CD=E6 =B9=FD =B4=F3 =B8=BB =CE=CC =C2=F0=A3=BF=D3=EB =C5=F3 =
=D3=D1 =CD=F8
>=C2=E7 =B6=D4 =D5=BD=A3=AC=BB=B9 =C4=DC =D2=BB =C6=F0 =CC=F4 =D5=BD GM =
>>=20
>
>
>
>
>



  _____ =20

 <http://event.mail.163.com/chanel/xyq.htm?from=3D163_NO7> =BE=A2 =B1=AC =
150 =CD=F2 =CD=AC =CA=B1
=D4=DA =CF=DF=A3=AC=D6=DA =C8=CB =D7=B7 =C5=F5 =C3=CE =BB=C3 =CE=F7 =
=D3=CE=20

Re: [Sqlrelay-discussion] ./sqlr-start -id oracletest with Error

[<tss...@16...>]

hi, i come from china hehe .
i have add /usr/local/firstworks/bin/ to your PATH environment variable.
but ,look here:
sh: line 1: sqlr-connection-oracle-debug: command not found
there was no file named sqlr-connection-oracle-debug in /usr/local/sqlrelay=
/bin



=D4=...@du... =D0=B4=B5=C0=A3=BA
>
>
>
>
>You should add /usr/local/firstworks/bin/ to your PATH environment
>variable.
>
>=20
>
>=20
>
>
>
>From:
>sql...@li...
>[mailto:sql...@li...] On Behalf Of ??
>Sent: Friday, August 10, 2007 3:15 PM
>To: Sql...@li...
>Subject: [Sqlrelay-discussion] ./sqlr-start -id oracletest with Error
>
>
>
>=20
>
>hi,all:
>my sqlrelay.conf:
><?xml version=3D"1.0"?>
><!DOCTYPE instances SYSTEM "sqlrelay.dtd">
><instances>
>
>        <instance
>id=3D"oracletest" port=3D"9000"
>socket=3D"/tmp/oracletest.socket" dbase=3D"oracle"
>connections=3D"3" maxconnections=3D"5"
>maxqueuelength=3D"0" growby=3D"1" ttl=3D"60" endofsession=3D"commit"
>sessiontimeout=3D"600" runasuser=3D"nobody"
>runasgroup=3D"nobody" cursors=3D"5" authtier=3D"connection"
>handoff=3D"pass" debug=3D"listener_and_connection">
>              =20
><users>
>                      =20
><user user=3D"oracletest" password=3D"oracletest"/>
>              =20
></users>
>              =20
><connections>
>                      =20
><connection connectionid=3D"oracletest"
>string=3D"user=3DUSER;password=3DPASSWORD;oracle_sid=3DORACLE_SID;oracle_h=
ome=3DORACLE_HOME;nls_lang=3DNLS_LANG;autocommit=3Dyes/no;fetchatonce=3DFET=
CHATONCE;maxselectlistsize=3DMAXSELECTLISTSIZE;maxitembuffersize=3DMAXITEMB=
UFFERSIZE"
>metric=3D"1"/>
>              =20
></connections>
>        </instance>
>
></instances>
>
>
>and i run it:
>[root@bigfish bin]# ./sqlr-start -id oracletest
>
>Starting listener:
>  sqlr-listener-debug -id oracletest -config
>/usr/local/sqlrelay/etc/sqlrelay.conf &
>sh: line 1: sqlr-listener-debug: command not found
>
>Starting 3 connections to oracletest :
>  sqlr-connection-oracle-debug -id oracletest -connectionid oracletest
>-config /usr/local/sqlrelay/etc/sqlrelay.conf &
>sh: line 1: sqlr-connection-oracle-debug: command not found
>  sqlr-connection-oracle-debug -id oracletest -connectionid oracletest
>-config /usr/local/sqlrelay/etc/sqlrelay.conf &
>sh: line 1: sqlr-connection-oracle-debug: command not found
>  sqlr-connection-oracle-debug -id oracletest -connectionid oracletest
>-config /usr/local/sqlrelay/etc/sqlrelay.conf &
>sh: line 1: sqlr-connection-oracle-debug: command not found
>
>Starting scaler:
>  sqlr-scaler  -id oracletest -debug  -config
>/usr/local/sqlrelay/etc/sqlrelay.conf
>sh: line 1: sqlr-scaler: command not found
>
>sqlr-scaler failed to start.
>
>
>Thanks to MP3.com for sponsoring:
>        Clustered/Replicated database
>support.
>        Perl API.
>Thanks to FeedLounge for sponsoring:
>        Query routing and filtering.
>[root@bigfish bin]#
>
>
>why?
>
>
>
>=20
>
>
>
>
>
>=20
>
>
>
>
>
>=20
>
>
>
>
>
>=20
>
>
>
>=20
>
>
>
>
>
>
>
>=C4=E3 =CD=E6 =B9=FD =B4=F3 =B8=BB =CE=CC =C2=F0=A3=BF=D3=EB =C5=F3 =D3=D1=
 =CD=F8
>=C2=E7 =B6=D4 =D5=BD=A3=AC=BB=B9 =C4=DC =D2=BB =C6=F0 =CC=F4 =D5=BD GM &gt=
;>=20
>
>
>
>
>

Re: [Sqlrelay-discussion] ./sqlr-start -id oracletest with Error

[<boc...@du...>]

You should add /usr/local/firstworks/bin/ to your PATH environment =
variable.

=20

=20

From: sql...@li...
[mailto:sql...@li...] On Behalf Of =
??
Sent: Friday, August 10, 2007 3:15 PM
To: Sql...@li...
Subject: [Sqlrelay-discussion] ./sqlr-start -id oracletest with Error

=20

hi,all:
my sqlrelay.conf:
<?xml version=3D"1.0"?>
<!DOCTYPE instances SYSTEM "sqlrelay.dtd">
<instances>

        <instance id=3D"oracletest" port=3D"9000"
socket=3D"/tmp/oracletest.socket" dbase=3D"oracle" connections=3D"3"
maxconnections=3D"5" maxqueuelength=3D"0" growby=3D"1" ttl=3D"60"
endofsession=3D"commit" sessiontimeout=3D"600" runasuser=3D"nobody"
runasgroup=3D"nobody" cursors=3D"5" authtier=3D"connection" =
handoff=3D"pass"
debug=3D"listener_and_connection">
                <users>
                        <user user=3D"oracletest" =
password=3D"oracletest"/>
                </users>
                <connections>
                        <connection connectionid=3D"oracletest"
string=3D"user=3DUSER;password=3DPASSWORD;oracle_sid=3DORACLE_SID;oracle_=
home=3DORACLE
_HOME;nls_lang=3DNLS_LANG;autocommit=3Dyes/no;fetchatonce=3DFETCHATONCE;m=
axselectl
istsize=3DMAXSELECTLISTSIZE;maxitembuffersize=3DMAXITEMBUFFERSIZE" =
metric=3D"1"/>
                </connections>
        </instance>

</instances>


and i run it:
[root@bigfish bin]# ./sqlr-start -id oracletest

Starting listener:
  sqlr-listener-debug -id oracletest -config
/usr/local/sqlrelay/etc/sqlrelay.conf &
sh: line 1: sqlr-listener-debug: command not found

Starting 3 connections to oracletest :
  sqlr-connection-oracle-debug -id oracletest -connectionid oracletest
-config /usr/local/sqlrelay/etc/sqlrelay.conf &
sh: line 1: sqlr-connection-oracle-debug: command not found
  sqlr-connection-oracle-debug -id oracletest -connectionid oracletest
-config /usr/local/sqlrelay/etc/sqlrelay.conf &
sh: line 1: sqlr-connection-oracle-debug: command not found
  sqlr-connection-oracle-debug -id oracletest -connectionid oracletest
-config /usr/local/sqlrelay/etc/sqlrelay.conf &
sh: line 1: sqlr-connection-oracle-debug: command not found

Starting scaler:
  sqlr-scaler  -id oracletest -debug  -config
/usr/local/sqlrelay/etc/sqlrelay.conf
sh: line 1: sqlr-scaler: command not found

sqlr-scaler failed to start.


Thanks to MP3.com for sponsoring:
        Clustered/Replicated database support.
        Perl API.
Thanks to FeedLounge for sponsoring:
        Query routing and filtering.
[root@bigfish bin]#


why?

=20

=20

=20

=20

=20

  _____ =20

 <http://event.mail.163.com/chanel/click.htm?from=3DNO_13&domain=3D163> =
=C4=E3 =CD=E6 =B9=FD
=B4=F3 =B8=BB =CE=CC =C2=F0=A3=BF=D3=EB =C5=F3 =D3=D1 =CD=F8 =C2=E7 =
=B6=D4 =D5=BD=A3=AC=BB=B9 =C4=DC =D2=BB =C6=F0 =CC=F4 =D5=BD GM >>=20

[Sqlrelay-discussion] ./sqlr-start -id oracletest with Error

[<tss...@16...>]

hi,all:
my sqlrelay.conf:
<?xml version="1.0"?>
<!DOCTYPE instances SYSTEM "sqlrelay.dtd">
<instances>
        <instance id="oracletest" port="9000" socket="/tmp/oracletest.socket" dbase="oracle" connections="3" maxconnections="5" maxqueuelength="0" growby="1" ttl="60" endofsession="commit" sessiontimeout="600" runasuser="nobody" runasgroup="nobody" cursors="5" authtier="connection" handoff="pass" debug="listener_and_connection">
                <users>
                        <user user="oracletest" password="oracletest"/>
                </users>
                <connections>
                        <connection connectionid="oracletest" string="user=USER;password=PASSWORD;oracle_sid=ORACLE_SID;oracle_home=ORACLE_HOME;nls_lang=NLS_LANG;autocommit=yes/no;fetchatonce=FETCHATONCE;maxselectlistsize=MAXSELECTLISTSIZE;maxitembuffersize=MAXITEMBUFFERSIZE" metric="1"/>
                </connections>
        </instance>
</instances>
and i run it:
[root@bigfish bin]# ./sqlr-start -id oracletest
Starting listener:
  sqlr-listener-debug -id oracletest -config /usr/local/sqlrelay/etc/sqlrelay.conf &
sh: line 1: sqlr-listener-debug: command not found
Starting 3 connections to oracletest :
  sqlr-connection-oracle-debug -id oracletest -connectionid oracletest -config /usr/local/sqlrelay/etc/sqlrelay.conf &
sh: line 1: sqlr-connection-oracle-debug: command not found
  sqlr-connection-oracle-debug -id oracletest -connectionid oracletest -config /usr/local/sqlrelay/etc/sqlrelay.conf &
sh: line 1: sqlr-connection-oracle-debug: command not found
  sqlr-connection-oracle-debug -id oracletest -connectionid oracletest -config /usr/local/sqlrelay/etc/sqlrelay.conf &
sh: line 1: sqlr-connection-oracle-debug: command not found
Starting scaler:
  sqlr-scaler  -id oracletest -debug  -config /usr/local/sqlrelay/etc/sqlrelay.conf
sh: line 1: sqlr-scaler: command not found
sqlr-scaler failed to start.
Thanks to MP3.com for sponsoring:
        Clustered/Replicated database support.
        Perl API.
Thanks to FeedLounge for sponsoring:
        Query routing and filtering.
[root@bigfish bin]#
why?
    

[Sqlrelay-discussion] SQL Relay - 0.39.2

[David M. <dav...@fi...>]

Sometimes, it seems no matter how much testing I do, I always find a bug
right after making a release :)

In this case though, I found an important bug.  There were cases where
server-side cursors weren't getting marked available when badly-behaved
clients disconnected, leading to error messages like "No server-side
cursors were available to process the query." and requiring a restart of
the sqlrelay server.

This is now fixed in SQL Relay 0.39.2.

Yay!

Also, some folks have had trouble extracting the tarball after
downloading it from sourceforge.  This seems to happen every time I
upload to them.  I'll upload a file, and it works fine when I first test
it, but then later, as it gets propagated to the mirror sites, it gets
messed up.  Maybe they use text rather than binary ftp to copy it or
something.  At any rate, I just uploaded this new release and downloaded
it and tested it and it worked, so hopefully you'll have the same
experience.  If not, let me know and I'll try again.  Usually when it
happens, re-uploading the file fixes the issue.  I'm not sure why, but
it seems to.

Dave

[Sqlrelay-discussion] SQL Relay 0.39.1

[David M. <dav...@fi...>]

Aaaah, wouldn't you know it.  I spent a ton of time testing SQL Relay
0.39, but still missed a couple of things.

So, SQL Relay 0.39.1 is now out with 2 fixes:

* readline detection in the configure script works now
* fixed an uninitialized variable in sqlr-import

Dave
dav...@fi...