sqlrelay-discussion — Discussion of topics related to SQL Relay

Re: [Sqlrelay-discussion] Running sqlr-scaler locally

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

It looks like the scaler really doesn't support -localstatedir.  It
should but it doesn't.  A quick fix would be to edit src/scaler/scaler.C
and replace instances of TMP_DIR with whatever directory you want to
use.  Then rebuild and reinstall.

I'll fix this in the next release.

David Muse
dav...@fi...

On Fri, 2008-05-16 at 23:35 -0700, Tim Piatenko wrote:
> Hi folks,
> 
> I have just successfully setup SQLRelay to connect to a MS SQL db, but
> all my efforts are being spoiled by the fact that I can't seem to be
> able to run sqlr-scaler ... The problem is that I don't have write
> permissions to /var/cache/sqlrelay/tmp/, which is where the scaler
> looks for the sqlr-listener-x file. I'm using the "sqlr-start -id ID
> -config CONFIGFILE [-localstatedir LOCALSTATEDIR]" form to start the
> listener and the connection daemons, and they work fine. I have them
> running and writing tmp files as well as debugging output in my home
> dir. So when I try to run "query -id ID -config MYCONFIG select ...",
> it just hangs forever. I looked in the debug files and saw the
> following:
> 
> tim@xyz debug: cat sqlr-listener.12775
> 
> 05/16/2008 22:57:44 PST listener [12775] : creating shared memory and semaphores
> 
>  ... blah blah blah ...
> 
> 05/16/2008 23:01:53 PST listener [12775] : forked a child: 12814
> 
> 05/16/2008 23:01:53 PST listener [12775] : waiting for client connection...
> 
> tim@xyz debug: cat sqlr-listener.12814
> 
> 05/16/2008 23:01:53 PST listener [12814] : getting authentication...
> 
> 05/16/2008 23:01:53 PST listener [12814] : 	listener-based
> authentication succeeded
> 
> 05/16/2008 23:01:53 PST listener [12814] : incrementing session count...
> 
> and then nothing ... it never gets incremented.
> 
> Is there really nothing I can do? I can't believe that there's a
> "-localstatedir" option for everything but the scaler. When I try to
> run the scaler, I get:
> 
> sqlr-scaler error:
> 	The file /var/cache/sqlrelay/tmp/sqlr-listener-sql1 was not found.
> 	This usually means that the sqlr-listener is not running.
> 	The sqlr-listener must be running for the sqlr-scaler to start.
> 
> Is there no way to tell it where my local tmp file is?
> 
> Thanks!
> 
> -Tim
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft 
> Defy all challenges. Microsoft(R) Visual Studio 2008. 
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Sqlrelay-discussion mailing list
> Sql...@li...
> https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion
> 
> 
> __________________________________________________
> D O T E A S Y - "Join the web hosting revolution!"
>              http://www.doteasy.com



__________________________________________________
D O T E A S Y - "Join the web hosting revolution!"
             http://www.doteasy.com

Re: [Sqlrelay-discussion] Starting listener: unsuccessful completion

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

The package installation should have created /var/sqlrelay/tmp and the
necessary subdirectories.  I'm surprised that it didn't.  They would
have been root-owned, but should have 777 permissions.

Make sure that the following directories exist and have 777 perms:

/var/sqlrelay/tmp
/var/sqlrelay/tmp/ipc
/var/sqlrelay/tmp/pids
/var/sqlrelay/tmp/sockets

/var/sqlrelay/cache

/var/sqlrelay/debug

If they already exist, remove any existing files in them.  If you ran
sqlrelay as root once, then try to run it as a different user afterward,
there could still be a file out there owned by root that can't be
overwritten or opened by the new user.

That should take care of it.  Hope this helps.

David
dav...@fi...


On Tue, 2008-05-20 at 17:42 +0200, Anthony Gardner wrote:
> I sent this under yahoo but have just seen in the archives that it's
> totally HTML. So, sorry about that. Let's see how gmail copes!!
> 
> Using the following sqlrelay.conf file .....
> <?xml version="1.0"?>
> <!DOCTYPE instances SYSTEM "sqlrelay.dtd">
> <instances>
> 
>   <instance id="sqlitetest"
>             port="9000"
>             socket="/tmp/sqlitetest.socket"
>             dbase="sqlite"
>             connections="3"
>             maxconnections="5"
>             maxqueuelength="0"
>             growby="1"
>             ttl="60"
>             endofsession="commit"
>             sessiontimeout="600"
>             runasuser="anthony"
>             runasgroup="anthony"
>             cursors="5"
>             authtier="listener"
>             handoff="pass"
>             debug="listener_and_connection">
>     <users>
>       <!-- user user="sqlitetest" password="sqlitetest"/ -->
>       <user user="" password=""/>
>     </users>
>     <connections>
>       <connection connectionid="sqlitetest"
> string="db=/home/anthony/Development/sqlrelay/sqlrelay_test.db"
> metric="1"/>
>     </connections>
>   </instance>
> 
> </instances>
> 
> I get the following output from running ....  sudo sqlr-start -id
> sqlitetest -config sqlrelay.conf
> 
> 
> Starting listener:
>   sqlr-listener-debug -id sqlitetest -config sqlrelay.conf &
> Debugging to: /var/cache/sqlrelay/debug/sqlr-listener.10989
> Could not open: /var/cache/sqlrelay/tmp/ipc/sqlitetest
> Make sure that the file and directory are readable and writable.
> 
> unsuccessful completion
> 
> Starting 3 connections to sqlitetest :
>   sqlr-connection-sqlite-debug -id sqlitetest -connectionid sqlitetest
> -config sqlrelay.conf &
> Debugging to: /var/cache/sqlrelay/debug/sqlr-connection.10991
> 
> sqlr-connection error:
>         The pid
> file /var/cache/sqlrelay/tmp/pids/sqlr-listener-sqlitetest was not
> found.
>         This usually means that the sqlr-listener 
> is not running.
>         The sqlr-listener must be running for the sqlr-connection to
> start.
> 
>   sqlr-connection-sqlite-debug -id sqlitetest -connectionid sqlitetest
> -config sqlrelay.conf &
> Debugging to: /var/cache/sqlrelay/debug/sqlr-connection.10996
> 
> sqlr-connection error:
>         The pid
> file /var/cache/sqlrelay/tmp/pids/sqlr-listener-sqlitetest was not
> found.
>         This usually means that the sqlr-listener 
> is not running.
>         The sqlr-listener must be running for the sqlr-connection to
> start.
> 
>   sqlr-connection-sqlite-debug -id sqlitetest -connectionid sqlitetest
> -config sqlrelay.conf &
> Debugging to: /var/cache/sqlrelay/debug/sqlr-connection.10998
> 
> sqlr-connection error:
>         The pid
> file /var/cache/sqlrelay/tmp/pids/sqlr-listener-sqlitetest was not
> found.
>         This usually means that the sqlr-listener 
> is not running.
>         The sqlr-listener must be running for the sqlr-connection to
> start.
> 
> 
> Starting cache manager:
>   sqlr-cachemanager
> Warning: using default id.
> 
> 
> Thanks to MP3.com for sponsoring: 
>         Clustered/Replicated database support.
>         Perl API.
> Thanks to FeedLounge for sponsoring: 
>         Query routing and filtering.
> 
> /var/cache/sqlrelay/tmp/ipc/sqlitetest was automatically created by
> sqlrelay 
> 
> I manually created pids and sockets directory
> /var/cache/sqlrelay/tmp/pids
> /var/cache/sqlrelay/tmp/sockets
> 
> I have changed permissions to 777 from /var/cache/sqlrelay downwards
> 
> SQL Relay version: 0.38 installed from Ubuntu binaries on 8.04
> 
> Any pointers would be great.
> 
> -Ants __________________________________________________
> D O T E A S Y - "Join the web hosting revolution!"
> http://www.doteasy.com 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft 
> Defy all challenges. Microsoft(R) Visual Studio 2008. 
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________ Sqlrelay-discussion mailing list Sql...@li... https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion



__________________________________________________
D O T E A S Y - "Join the web hosting revolution!"
             http://www.doteasy.com

[Sqlrelay-discussion] Starting listener: unsuccessful completion

[Anthony G. <ant...@gm...>]

I sent this under yahoo but have just seen in the archives that it's totally
HTML. So, sorry about that. Let's see how gmail copes!!

Using the following sqlrelay.conf file .....
<?xml version="1.0"?>
<!DOCTYPE instances SYSTEM "sqlrelay.dtd">
<instances>

  <instance id="sqlitetest"
            port="9000"
            socket="/tmp/sqlitetest.socket"
            dbase="sqlite"
            connections="3"
            maxconnections="5"
            maxqueuelength="0"
            growby="1"
            ttl="60"
            endofsession="commit"
            sessiontimeout="600"
            runasuser="anthony"
            runasgroup="anthony"
            cursors="5"
            authtier="listener"
            handoff="pass"
            debug="listener_and_connection">
    <users>
      <!-- user user="sqlitetest" password="sqlitetest"/ -->
      <user user="" password=""/>
    </users>
    <connections>
      <connection connectionid="sqlitetest"
string="db=/home/anthony/Development/sqlrelay/sqlrelay_test.db" metric="1"/>
    </connections>
  </instance>

</instances>

I get the following output from running ....  sudo sqlr-start -id sqlitetest
-config sqlrelay.conf


Starting listener:
  sqlr-listener-debug -id sqlitetest -config sqlrelay.conf &
Debugging to: /var/cache/sqlrelay/debug/sqlr-listener.10989
Could not open: /var/cache/sqlrelay/tmp/ipc/sqlitetest
Make sure that the file and directory are readable and writable.

unsuccessful completion

Starting 3 connections to sqlitetest :
  sqlr-connection-sqlite-debug -id sqlitetest -connectionid sqlitetest
-config sqlrelay.conf &
Debugging to: /var/cache/sqlrelay/debug/sqlr-connection.10991

sqlr-connection error:
        The pid file /var/cache/sqlrelay/tmp/pids/sqlr-listener-sqlitetest
was not found.
        This usually means that the sqlr-listener
is not running.
        The sqlr-listener must be running for the sqlr-connection to start.

  sqlr-connection-sqlite-debug -id sqlitetest -connectionid sqlitetest
-config sqlrelay.conf &
Debugging to: /var/cache/sqlrelay/debug/sqlr-connection.10996

sqlr-connection error:
        The pid file /var/cache/sqlrelay/tmp/pids/sqlr-listener-sqlitetest
was not found.
        This usually means that the sqlr-listener
is not running.
        The sqlr-listener must be running for the sqlr-connection to start.

  sqlr-connection-sqlite-debug -id sqlitetest -connectionid sqlitetest
-config sqlrelay.conf &
Debugging to: /var/cache/sqlrelay/debug/sqlr-connection.10998

sqlr-connection error:
        The pid file /var/cache/sqlrelay/tmp/pids/sqlr-listener-sqlitetest
was not found.
        This usually means that the sqlr-listener
is not running.
        The sqlr-listener must be running for the sqlr-connection to start.


Starting cache manager:
  sqlr-cachemanager
Warning: using default id.


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

/var/cache/sqlrelay/tmp/ipc/sqlitetest was automatically created by sqlrelay


I manually created pids and sockets directory
/var/cache/sqlrelay/tmp/pids
/var/cache/sqlrelay/tmp/sockets

I have changed permissions to 777 from /var/cache/sqlrelay downwards

SQL Relay version: 0.38 installed from Ubuntu binaries on 8.04

Any pointers would be great.

-Ants

[Sqlrelay-discussion] Starting listener: unsuccessful completion

[Anthony G. <cyc...@ya...>]

<table cellspacing='0' cellpadding='0' border='0' ><tr><td style='font: inherit;'>Using the following sqlrelay.conf file .....<br>&lt;?xml version="1.0"?&gt;<br>&lt;!DOCTYPE instances SYSTEM "sqlrelay.dtd"&gt;<br>&lt;instances&gt;<br><br>&nbsp; &lt;instance id="sqlitetest"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; port="9000"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; socket="/tmp/sqlitetest.socket"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; dbase="sqlite"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; connections="3"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; maxconnections="5"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; maxqueuelength="0"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; growby="1"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
 ttl="60"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; endofsession="commit"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; sessiontimeout="600"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; runasuser="anthony"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; runasgroup="anthony"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; cursors="5"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; authtier="listener"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; handoff="pass"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; debug="listener_and_connection"&gt;<br>&nbsp;&nbsp;&nbsp; &lt;users&gt;<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; &lt;!-- user user="sqlitetest" password="sqlitetest"/ --&gt;<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; &lt;user user="" password=""/&gt;<br>&nbsp;&nbsp;&nbsp;
 &lt;/users&gt;<br>&nbsp;&nbsp;&nbsp; &lt;connections&gt;<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; &lt;connection connectionid="sqlitetest" string="db=/home/anthony/Development/sqlrelay/sqlrelay_test.db" metric="1"/&gt;<br>&nbsp;&nbsp;&nbsp; &lt;/connections&gt;<br>&nbsp; &lt;/instance&gt;<br><br>&lt;/instances&gt;<br><br>I get the following output from running ....&nbsp; sudo sqlr-start -id sqlitetest -config sqlrelay.conf<br><br><br>Starting listener:<br>&nbsp; sqlr-listener-debug -id sqlitetest -config sqlrelay.conf &amp;<br>Debugging to: /var/cache/sqlrelay/debug/sqlr-listener.10989<br>Could not open: /var/cache/sqlrelay/tmp/ipc/sqlitetest<br>Make sure that the file and directory are readable and writable.<br><br>unsuccessful completion<br><br>Starting 3 connections to sqlitetest :<br>&nbsp; sqlr-connection-sqlite-debug -id sqlitetest -connectionid sqlitetest -config sqlrelay.conf &amp;<br>Debugging to:
 /var/cache/sqlrelay/debug/sqlr-connection.10991<br><br>sqlr-connection error:<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; The pid file /var/cache/sqlrelay/tmp/pids/sqlr-listener-sqlitetest was not found.<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; This usually means that the sqlr-listener <br>is not running.<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; The sqlr-listener must be running for the sqlr-connection to start.<br><br>&nbsp; sqlr-connection-sqlite-debug -id sqlitetest -connectionid sqlitetest -config sqlrelay.conf &amp;<br>Debugging to: /var/cache/sqlrelay/debug/sqlr-connection.10996<br><br>sqlr-connection error:<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; The pid file /var/cache/sqlrelay/tmp/pids/sqlr-listener-sqlitetest was not found.<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; This usually means that the sqlr-listener <br>is not running.<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; The sqlr-listener must be running for the
 sqlr-connection to start.<br><br>&nbsp; sqlr-connection-sqlite-debug -id sqlitetest -connectionid sqlitetest -config sqlrelay.conf &amp;<br>Debugging to: /var/cache/sqlrelay/debug/sqlr-connection.10998<br><br>sqlr-connection error:<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; The pid file /var/cache/sqlrelay/tmp/pids/sqlr-listener-sqlitetest was not found.<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; This usually means that the sqlr-listener <br>is not running.<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; The sqlr-listener must be running for the sqlr-connection to start.<br><br><br>Starting cache manager:<br>&nbsp; sqlr-cachemanager<br>Warning: using default id.<br><br><br>Thanks to MP3.com for sponsoring: <br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Clustered/Replicated database support.<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Perl API.<br>Thanks to FeedLounge for sponsoring: <br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Query routing and
 filtering.<br><br>/var/cache/sqlrelay/tmp/ipc/sqlitetest was automatically created by sqlrelay <br><br>I manually created pids and sockets directory<br>/var/cache/sqlrelay/tmp/pids<br>/var/cache/sqlrelay/tmp/sockets<br><br>I have changed permissions to 777 from /var/cache/sqlrelay downwards<br><br>SQL Relay version: 0.38 installed from Ubuntu binaries on 8.04<br><br>Any pointers would be great.<br><br>-Ants<br><br>Disclaimer: Technically speaking, I am always wrong!</td></tr></table><br>



      <hr size=1>
Sent from <a  
href="http://us.rd.yahoo.com/mailuk/taglines/isp/control/*http://us.rd.yahoo.com/evt=52418/*http://uk.docs.yahoo.com/nowyoucan.html" target=_blank>Yahoo! Mail</a>.
<br>
A Smarter Email.

Re: [Sqlrelay-discussion] Building Sqlrelay 0.39.4 on Solaris 10

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

It appears that the build is finding an older version of rudiments.  You
may need to install the most current version of the rudiments library.
Or, if it's already installed, then you'll need to remove any older
versions.

David Muse
dav...@fi...

On Fri, 2008-05-09 at 08:36 +0200, Sebastien BROCHET wrote:
> Hello,
> 
> I can't build sqlrelay 0.39.4 on my solaris 10 box.
> 
> # gmake
> gmake -C src all
> gmake[1]: Entering directory `/export/home/seb/sqlrelay-0.39.4/src'
> gmake -C util all
> gmake[2]: Entering directory `/export/home/seb/sqlrelay-0.39.4/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 -I/usr/pkg/include -pthread
> -I/usr/sfw/include -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 -I/usr/pkg/include -pthread
> -I/usr/sfw/include -c cmdline.C  -fPIC -DPIC -o .libs/cmdline.o
> g++: unrecognized option `-pthread'
> g++: unrecognized option `-pthread'
> cmdline.C: In member function `void cmdline::setId()':
> cmdline.C:20: error: `getValue' undeclared (first use this function)
> cmdline.C:20: error: (Each undeclared identifier is reported only once
> for each function it appears in.)
> cmdline.C: In member function `void cmdline::setConfig()':
> cmdline.C:30: error: `getValue' undeclared (first use this function)
> cmdline.C: In member function `void cmdline::setLocalStateDir()':
> cmdline.C:39: error: `getValue' undeclared (first use this function)
> gmake[2]: *** [cmdline.lo] Error 1
> gmake[2]: Leaving directory `/export/home/seb/sqlrelay-0.39.4/src/util'
> gmake[1]: *** [all] Error 2
> gmake[1]: Leaving directory `/export/home/seb/sqlrelay-0.39.4/src'
> gmake: *** [all] Error 2
> 
> Any ideas ?
> 
> Séb.
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
> Don't miss this year's exciting event. There's still time to save $100. 
> Use priority code J8TL2D2. 
> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
> _______________________________________________
> Sqlrelay-discussion mailing list
> Sql...@li...
> https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion
> 
> 
> __________________________________________________
> D O T E A S Y - "Join the web hosting revolution!"
>              http://www.doteasy.com



__________________________________________________
D O T E A S Y - "Join the web hosting revolution!"
             http://www.doteasy.com

[Sqlrelay-discussion] Running sqlr-scaler locally

[Tim P. <ti...@gm...>]

Hi folks,

I have just successfully setup SQLRelay to connect to a MS SQL db, but
all my efforts are being spoiled by the fact that I can't seem to be
able to run sqlr-scaler ... The problem is that I don't have write
permissions to /var/cache/sqlrelay/tmp/, which is where the scaler
looks for the sqlr-listener-x file. I'm using the "sqlr-start -id ID
-config CONFIGFILE [-localstatedir LOCALSTATEDIR]" form to start the
listener and the connection daemons, and they work fine. I have them
running and writing tmp files as well as debugging output in my home
dir. So when I try to run "query -id ID -config MYCONFIG select ...",
it just hangs forever. I looked in the debug files and saw the
following:

tim@xyz debug: cat sqlr-listener.12775

05/16/2008 22:57:44 PST listener [12775] : creating shared memory and semaphores

 ... blah blah blah ...

05/16/2008 23:01:53 PST listener [12775] : forked a child: 12814

05/16/2008 23:01:53 PST listener [12775] : waiting for client connection...

tim@xyz debug: cat sqlr-listener.12814

05/16/2008 23:01:53 PST listener [12814] : getting authentication...

05/16/2008 23:01:53 PST listener [12814] : 	listener-based
authentication succeeded

05/16/2008 23:01:53 PST listener [12814] : incrementing session count...

and then nothing ... it never gets incremented.

Is there really nothing I can do? I can't believe that there's a
"-localstatedir" option for everything but the scaler. When I try to
run the scaler, I get:

sqlr-scaler error:
	The file /var/cache/sqlrelay/tmp/sqlr-listener-sql1 was not found.
	This usually means that the sqlr-listener is not running.
	The sqlr-listener must be running for the sqlr-scaler to start.

Is there no way to tell it where my local tmp file is?

Thanks!

-Tim

[Sqlrelay-discussion] Python bindings and rowcount

[Jonathan K. <jka...@vo...>]

Hi,

(This is a resend because I didn't see it show up in the mailing list.)

I'm running the latest release of SQLRelay, 0.39.4, which is talking to an Oracle database.

I've looked at the archives of the mailing list and haven't seen my issue mentioned before.

I have a problem with the python db api 2.0-esque bindings, PySQLRDB; cur.rowcount always returns 0 (cur.rowCount() from PySQLClient works fine).



>>> from SQLRelay import PySQLRDB
>>> con = PySQLRDB.connect('localhost', 9200, '', 'username', 'password', 0, 1)
>>> cur = con.cursor()
>>> cur.execute("select 1 from dual")
>>> cur.rowcount
0
>>> cur.fetchall()
[[Decimal("1")]]
>>> cur.rowcount
0



>>> from SQLRelay import PySQLRClient
>>> con = PySQLRClient.sqlrconnection('localhost', 9200, '', 'username', 'password', 0, 1)
>>> cur = PySQLRClient.sqlrcursor(con)
>>> cur.sendQuery("select 1 from dual")
1
>>> cur.rowCount()
1



I looked at PySQLRDB.py, PySQLRClient.py, and CSQLRelay.C.

In rowCount(), in PySQLRClient.py, the value returned comes from CSQLRelay.rowCount(self.cursor), which comes from ((sqlrcursor *)sqlrcur)->rowCount().

In execute(), in PySQLRDB.py, rowcount is assigned to either CSQLRelay.totalRows(self.cursor) or CSQLRelay.affectedRows(self.cursor), which come from ((sqlrcursor *)sqlrcur)->totalRows() or  ((sqlrcursor *)sqlrcur)->affectedRows(), respectively.

So, rowCount returns 1, and both totalRows and affectedRows return 0.

Does anyone know why cur.rowcount, from PySQLRDB.py, would always return 0?

Thank you,
Jonathan Kaczynski

[Sqlrelay-discussion] Building Sqlrelay 0.39.4 on Solaris 10

[Sebastien B. <seb...@do...>]

Hello,

I can't build sqlrelay 0.39.4 on my solaris 10 box.

# gmake
gmake -C src all
gmake[1]: Entering directory `/export/home/seb/sqlrelay-0.39.4/src'
gmake -C util all
gmake[2]: Entering directory `/export/home/seb/sqlrelay-0.39.4/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 -I/usr/pkg/include -pthread
-I/usr/sfw/include -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 -I/usr/pkg/include -pthread
-I/usr/sfw/include -c cmdline.C  -fPIC -DPIC -o .libs/cmdline.o
g++: unrecognized option `-pthread'
g++: unrecognized option `-pthread'
cmdline.C: In member function `void cmdline::setId()':
cmdline.C:20: error: `getValue' undeclared (first use this function)
cmdline.C:20: error: (Each undeclared identifier is reported only once
for each function it appears in.)
cmdline.C: In member function `void cmdline::setConfig()':
cmdline.C:30: error: `getValue' undeclared (first use this function)
cmdline.C: In member function `void cmdline::setLocalStateDir()':
cmdline.C:39: error: `getValue' undeclared (first use this function)
gmake[2]: *** [cmdline.lo] Error 1
gmake[2]: Leaving directory `/export/home/seb/sqlrelay-0.39.4/src/util'
gmake[1]: *** [all] Error 2
gmake[1]: Leaving directory `/export/home/seb/sqlrelay-0.39.4/src'
gmake: *** [all] Error 2

Any ideas ?

Séb.

Re: [Sqlrelay-discussion] Ruby Binding Issue - SQLRCursor has no explicit "close" method

[Chad J. <cha...@gm...>]

On Thu, May 1, 2008 at 10:17 AM, David Muse <dav...@fi...> wrote:
> Aaah!  You're right.  And it looks like that's not a ruby-specific
>  issue.  Basically any non C/C++ api has the problem.  In C/C++ you can
>  free or delete the cursor, but not in other languages.
>
>  I'll add a feature to close a cursor in the next release.
>
>  David Muse
>  dav...@fi...

Excellent, thanks David!

-Chad Johnson

Re: [Sqlrelay-discussion] Installing sql relay on Mac OS X with MySQL support

[Holger A. <ke...@fe...>]

I found out, that the tweaks are not necessary. configure looks for a  
mysql_config, but MacPorts added a 5 to almost all mysql executables  
and paths. So a ln -s /opt/local/bin/mysql_config5 /opt/local/bin/ 
mysql_config and ./configure --with-mysql-prefix=/opt/local works too.

Re: [Sqlrelay-discussion] Installing sql relay on Mac OS X with MySQL support

[Holger A. <ke...@fe...>]

Hi David!

David Muse wrote:

> Did the MacPorts version of mysql come with a mysql_config script?  If
> so, you should just have to provide the configure script with it's
> prefix.  For example, if it's in /opt/local/bin/mysql_config, then you
> could use:
>
> ./configure --with-mysql-prefix=/opt/local
>
> And it will find the mysql_config script and use it to get the paths  
> for
> libs and include files.  And everything should go well.

Yes, there is an mysql_config5 in /opt/local/bin which prints

Usage: /opt/local/bin/mysql_config5 [OPTIONS]
Options:
         --cflags         [-I/opt/local/include/mysql5/mysql]
         --include        [-I/opt/local/include/mysql5/mysql]
         --libs           [-L/opt/local/lib -L/opt/local/lib/mysql5/ 
mysql -lmysqlclient -L/opt/local/lib -lz -lm  -L/opt/local/lib -lssl - 
lcrypto]
         --libs_r         [-L/opt/local/lib -L/opt/local/lib/mysql5/ 
mysql -lmysqlclient_r -L/opt/local/lib -lz -lm   -L/opt/local/lib - 
lssl -lcrypto]
         --socket         [/opt/local/var/run/mysql5/mysqld.sock]
         --port           [3306]
         --version        [5.0.51]
         --libmysqld-libs [-L/opt/local/lib -L/opt/local/lib/mysql5/ 
mysql -lmysqld -L/opt/local/lib -lz -lm     -L/opt/local/lib -lssl - 
lcrypto]

but running  ./configure --with-mysql-prefix=/opt/local without tweaks  
disables mysql support.


> Otherwise, you're on the right track. It looks like the tweaks to the
> configure script just didn't do the job.  Try adding them around line
> 23992 which is right before the tests are run.  You'll know it  
> worked if
> the various "checking for XXX" lines output by the configure script  
> say
> "yes" sometimes.

23992? Are you sure?

23991   (exit $ac_status); } &&
23992          { ac_try='test -z "$ac_cxx_werror_flag"
23993                          || test ! -s conftest.err'
23994   { (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5

Line 23644 seems better to me and I added lines according to  
mysql_config5

23644 #eval "MYSQLINCLUDES=\"$INCLUDESTRING\""
23645 #eval "MYSQLLIBS=\"$LIBSTRING\""
23646 eval "MYSQLINCLUDES=\"-I/opt/local/include/mysql5/mysql\""
23647 eval "MYSQLLIBS=\"-L/opt/local/lib -L/opt/local/lib/mysql5/mysql  
-lmysqlclient -L/opt/local/lib -lz -lm  -L/opt/local/lib -lssl -lcrypto 
\""

and now I get

***** MySQL ******************
checking if MySQL requires -lz... no
checking for gzopen in -lz... yes
checking for mysql_real_connect... yes
checking for mysql_select_db... yes
checking for mysql_ping... yes
checking for mysql_change_user... yes
checking for mysql_commit... yes
checking for mysql_rollback... yes
checking for mysql_autocommit... yes
checking for mysql_prepare... yes
checking for mysql_next_result... yes
checking for CR_SERVER_GONE_ERROR... yes
checking for CR_SERVER_LOST... yes
checking for FIELD_TYPE_YEAR... yes
checking for FIELD_TYPE_NEWDATE... yes
checking for FIELD_TYPE_ENUM... yes
checking for FIELD_TYPE_SET... yes
checking for FIELD_TYPE_NEWDECIMAL... yes
checking for MYSQL_OPT_RECONNECT... yes
checking for MYSQL_GET_SERVER_VERSION... yes
mysql includes... -I/opt/local/include/mysql5/mysql
mysql libs... -L/opt/local/lib -L/opt/local/lib/mysql5/mysql - 
lmysqlclient -L/opt/local/lib -lz -lm  -L/opt/local/lib -lssl -lcrypto

and sql relay compiles without problems. Thank you very much! :)

Re: [Sqlrelay-discussion] Installing sql relay on Mac OS X with MySQL support

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

It looks like the tweaks to the configure script didn't do the job.
Try adding them around line 23992.

Did the MacPorts version of mysql come with a mysql_config script?  If
so, you should just have to provide the configure script with it's
prefix.  For example, if it's in /opt/local/bin/mysql_config, then you
could use:

./configure --with-mysql-prefix=/opt/local

And it will find the mysql_config script and use it to get the paths for
libs and include files.  And everything should go well.

Otherwise, you're on the right track. It looks like the tweaks to the
configure script just didn't do the job.  Try adding them around line
23992 which is right before the tests are run.  You'll know it worked if
the various "checking for XXX" lines output by the configure script say
"yes" sometimes.

David Muse
dav...@fi...

On Wed, 2008-04-30 at 15:59 +0200, Holger Amann wrote:
> Hi!
> 
> I'm trying to compile sql relay on OS X 10.5.2 with MySQL 5.0.51  
> installed by MacPorts in /opt/local/libs/mysql5/mysql and /opt/local/ 
> include/mysql5/mysql. Because configure didn't find the headers and  
> libs, I played with configure.mk.in and set
> 
> MYSQLINCLUDES = -I/opt/local/include/mysql5/mysql
> MYSQLLIBS = -L/opt/local/libs/mysql5/mysql
> 
> , and I modified configure with
> 
> MYSQLINCLUDES="/opt/local/include/mysql5/mysql"
> MYSQLLIBS="/opt/local/libs/mysql5/mysql"
> 
> This results in
> 
> ...
> ***** MySQL ******************
> checking for mysql_real_connect... no
> checking for mysql_select_db... no
> checking for mysql_ping... no
> checking for mysql_change_user... no
> checking for mysql_commit... no
> checking for mysql_rollback... no
> checking for mysql_autocommit... no
> checking for mysql_prepare... no
> checking for mysql_next_result... no
> checking for CR_SERVER_GONE_ERROR... no
> checking for CR_SERVER_LOST... no
> checking for FIELD_TYPE_YEAR... no
> checking for FIELD_TYPE_NEWDATE... no
> checking for FIELD_TYPE_ENUM... no
> checking for FIELD_TYPE_SET... no
> checking for FIELD_TYPE_NEWDECIMAL... no
> checking for MYSQL_OPT_RECONNECT... no
> checking for MYSQL_GET_SERVER_VERSION... no
> mysql includes... /opt/local/include/mysql5/mysql
> mysql libs... /opt/local/libs/mysql5/mysql
> ...
> 
> and make aborts with
> ...
> make -C connection all
> make[2]: Nothing to be done for `all'.
> make -C connections all
> make -C status all
> make[3]: Nothing to be done for `all'.
> make -C mysql all
> /bin/sh ../../../libtool --mode=compile g++ -Wall -pipe -Wno-long- 
> double -I/opt/local/include/mysql5/mysql -pthread  -I./ -I../../../ - 
> I../../../src/common -I../../../src/util -I../../../src/connection -I/ 
> usr/local/firstworks/include -pthread -I/opt/local/include -I/opt/ 
> local/include/mysql5/mysql -I../../../src/api/c++/include -c  
> mysqlconnection.C -o mysqlconnection.lo
>   g++ -Wall -pipe -Wno-long-double -I/opt/local/include/mysql5/mysql - 
> pthread -I./ -I../../../ -I../../../src/common -I../../../src/util - 
> I../../../src/connection -I/usr/local/firstworks/include -pthread -I/ 
> opt/local/include -I/opt/local/include/mysql5/mysql -I../../../src/api/ 
> c++/include -c mysqlconnection.C  -fno-common -DPIC -o .libs/ 
> mysqlconnection.o
> mysqlconnection.C: In member function 'virtual bool  
> mysqlconnection::logIn(bool)':
> mysqlconnection.C:102: error: 'mysql_connect' was not declared in this  
> scope
> mysqlconnection.C:70: warning: unused variable 'dbval'
> make[3]: *** [mysqlconnection.lo] Error 1
> make[2]: *** [all] Error 2
> make[1]: *** [all] Error 2
> make: *** [all] Error 2
> ...
> 
> I want to write an application based on sql relay, but I have no idea  
> how to resolve this.
> Thanks for any hint or suggestion!
> 
> Holger
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
> Don't miss this year's exciting event. There's still time to save $100. 
> Use priority code J8TL2D2. 
> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
> _______________________________________________
> Sqlrelay-discussion mailing list
> Sql...@li...
> https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion
> 
> 
> __________________________________________________
> D O T E A S Y - "Join the web hosting revolution!"
>              http://www.doteasy.com



__________________________________________________
D O T E A S Y - "Join the web hosting revolution!"
             http://www.doteasy.com

Re: [Sqlrelay-discussion] Ruby Binding Issue - SQLRCursor has no explicit "close" method

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

Aaah!  You're right.  And it looks like that's not a ruby-specific
issue.  Basically any non C/C++ api has the problem.  In C/C++ you can
free or delete the cursor, but not in other languages.

I'll add a feature to close a cursor in the next release.

David Muse
dav...@fi...


On Thu, 2008-04-24 at 17:18 -0500, Chad Johnson wrote:
> Within the Ruby bindings there does not appear to be a way to explicitly
> close an open SQLRCursor object.
> 
> Now this isn't a catastrophic problem because cursors are being closed when
> SQLRCursor objects are garbage collected. But if you run up against the
> configured SQL Relay cursor limit (5 by default) it appears as though your
> code will hang (assuming the garbage collector hasn't had a chance to kick
> in).
> 
> Can a method be added to the Ruby bindings to allow a cursor to be closed
> by hand?
> 
> -Chad Johnson
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
> Don't miss this year's exciting event. There's still time to save $100. 
> Use priority code J8TL2D2. 
> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
> _______________________________________________
> Sqlrelay-discussion mailing list
> Sql...@li...
> https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion
> 
> 
> __________________________________________________
> D O T E A S Y - "Join the web hosting revolution!"
>              http://www.doteasy.com



__________________________________________________
D O T E A S Y - "Join the web hosting revolution!"
             http://www.doteasy.com

Re: [Sqlrelay-discussion] FastCgi and SqlRelay?

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

I'm really not sure.  It probably wouldn't be a C++ thing though, rather
a FastCGI thing.  It's possible that the fastcgi framework might throw
an exception.  I believe that the apache module api provides a way for a
function to be called when the page is aborted.  If you're using fastcgi
as an apache module, then it may implement that function and provide
it's own callback that you'd override or may raise an exception.  I
don't know fastcgi well enough though to say.

Dave


On Wed, 2008-04-30 at 08:29 -0600, TazaTek wrote:
> In C++, what would be the equivalent of a register_shutdown_function()??  
> 
> Would this be a try-catch block? Or do I need to register a signal? Or... ???
> 
> I'm not sure of what happens when a user aborts the connection.... could you shed some light on this?
> 
> Thanks
> 
> Matt
> 
> 
> 
> >> If so, the only issue that I know of is that you really need to make
> >> sure that you handle page aborts carefully.  
> 
> >> See http://sqlrelay.sourceforge.net/sqlrelay/faq.html#php for
> >> PHP-specific info.  The same may apply to fastcgi.
> 
> >> Dave
> >> dav...@fi...
> 
> 
> On Thu, 2008-04-24 at 13:24 -0600, TazaTek wrote:
> >  
> > 
> > Hello all!
> > 
> >  
> > 
> > I was wondering if there were any problems in mixing fastcgi and
> > sqlrelay?  It seems that they both handle some of the speed
> > performance mechanisms that are required of a web server, but I didn’t
> > know if there were any problems in mixing the two.
> > 
> >  
> > 
> > Thanks
> > 
> >  
> > 
> > Matt
> > 
> > --
> > 
> > Matthew Kettlewell
> > 
> > President - TazaTek
> > 
> > www.TazaTek.com
> > 
> > www.Pairs-List.com
> > 
> > 970-985-4192
> 
> 
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
> Don't miss this year's exciting event. There's still time to save $100. 
> Use priority code J8TL2D2. 
> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
> _______________________________________________
> Sqlrelay-discussion mailing list
> Sql...@li...
> https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion
> 
> __________________________________________________
> D O T E A S Y - "Join the web hosting revolution!"
>              http://www.doteasy.com



__________________________________________________
D O T E A S Y - "Join the web hosting revolution!"
             http://www.doteasy.com

Re: [Sqlrelay-discussion] FastCgi and SqlRelay?

[TazaTek <ma...@ta...>]

In C++, what would be the equivalent of a register_shutdown_function()??  

Would this be a try-catch block? Or do I need to register a signal? Or... ???

I'm not sure of what happens when a user aborts the connection.... could you shed some light on this?

Thanks

Matt



>> If so, the only issue that I know of is that you really need to make
>> sure that you handle page aborts carefully.  

>> See http://sqlrelay.sourceforge.net/sqlrelay/faq.html#php for
>> PHP-specific info.  The same may apply to fastcgi.

>> Dave
>> dav...@fi...


On Thu, 2008-04-24 at 13:24 -0600, TazaTek wrote:
>  
> 
> Hello all!
> 
>  
> 
> I was wondering if there were any problems in mixing fastcgi and
> sqlrelay?  It seems that they both handle some of the speed
> performance mechanisms that are required of a web server, but I didn’t
> know if there were any problems in mixing the two.
> 
>  
> 
> Thanks
> 
>  
> 
> Matt
> 
> --
> 
> Matthew Kettlewell
> 
> President - TazaTek
> 
> www.TazaTek.com
> 
> www.Pairs-List.com
> 
> 970-985-4192

[Sqlrelay-discussion] Installing sql relay on Mac OS X with MySQL support

[Holger A. <ke...@fe...>]

Hi!

I'm trying to compile sql relay on OS X 10.5.2 with MySQL 5.0.51  
installed by MacPorts in /opt/local/libs/mysql5/mysql and /opt/local/ 
include/mysql5/mysql. Because configure didn't find the headers and  
libs, I played with configure.mk.in and set

MYSQLINCLUDES = -I/opt/local/include/mysql5/mysql
MYSQLLIBS = -L/opt/local/libs/mysql5/mysql

, and I modified configure with

MYSQLINCLUDES="/opt/local/include/mysql5/mysql"
MYSQLLIBS="/opt/local/libs/mysql5/mysql"

This results in

...
***** MySQL ******************
checking for mysql_real_connect... no
checking for mysql_select_db... no
checking for mysql_ping... no
checking for mysql_change_user... no
checking for mysql_commit... no
checking for mysql_rollback... no
checking for mysql_autocommit... no
checking for mysql_prepare... no
checking for mysql_next_result... no
checking for CR_SERVER_GONE_ERROR... no
checking for CR_SERVER_LOST... no
checking for FIELD_TYPE_YEAR... no
checking for FIELD_TYPE_NEWDATE... no
checking for FIELD_TYPE_ENUM... no
checking for FIELD_TYPE_SET... no
checking for FIELD_TYPE_NEWDECIMAL... no
checking for MYSQL_OPT_RECONNECT... no
checking for MYSQL_GET_SERVER_VERSION... no
mysql includes... /opt/local/include/mysql5/mysql
mysql libs... /opt/local/libs/mysql5/mysql
...

and make aborts with
...
make -C connection all
make[2]: Nothing to be done for `all'.
make -C connections all
make -C status all
make[3]: Nothing to be done for `all'.
make -C mysql all
/bin/sh ../../../libtool --mode=compile g++ -Wall -pipe -Wno-long- 
double -I/opt/local/include/mysql5/mysql -pthread  -I./ -I../../../ - 
I../../../src/common -I../../../src/util -I../../../src/connection -I/ 
usr/local/firstworks/include -pthread -I/opt/local/include -I/opt/ 
local/include/mysql5/mysql -I../../../src/api/c++/include -c  
mysqlconnection.C -o mysqlconnection.lo
  g++ -Wall -pipe -Wno-long-double -I/opt/local/include/mysql5/mysql - 
pthread -I./ -I../../../ -I../../../src/common -I../../../src/util - 
I../../../src/connection -I/usr/local/firstworks/include -pthread -I/ 
opt/local/include -I/opt/local/include/mysql5/mysql -I../../../src/api/ 
c++/include -c mysqlconnection.C  -fno-common -DPIC -o .libs/ 
mysqlconnection.o
mysqlconnection.C: In member function 'virtual bool  
mysqlconnection::logIn(bool)':
mysqlconnection.C:102: error: 'mysql_connect' was not declared in this  
scope
mysqlconnection.C:70: warning: unused variable 'dbval'
make[3]: *** [mysqlconnection.lo] Error 1
make[2]: *** [all] Error 2
make[1]: *** [all] Error 2
make: *** [all] Error 2
...

I want to write an application based on sql relay, but I have no idea  
how to resolve this.
Thanks for any hint or suggestion!

Holger

[Sqlrelay-discussion] Ruby Binding Issue - SQLRCursor has no explicit "close" method

[Chad J. <cha...@gm...>]

Within the Ruby bindings there does not appear to be a way to explicitly
close an open SQLRCursor object.

Now this isn't a catastrophic problem because cursors are being closed when
SQLRCursor objects are garbage collected. But if you run up against the
configured SQL Relay cursor limit (5 by default) it appears as though your
code will hang (assuming the garbage collector hasn't had a chance to kick
in).

Can a method be added to the Ruby bindings to allow a cursor to be closed
by hand?

-Chad Johnson

Re: [Sqlrelay-discussion] FastCgi and SqlRelay?

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

If I remember correctly, fastcgi keeps the cgi running persistently
rather than having a cgi process start and stop over and over.

If so, the only issue that I know of is that you really need to make
sure that you handle page aborts carefully.  Similar issues exist when
using mod_perl or mod_php.  If someone hits stop on their browser with a
regular cgi, then the cgi will exit and everything will be cleaned up.
However, with fastcgi, if the entire code doesn't execute, then clean-up
code may not be called and it could leave lingering network connections
between the http server and sqlrelay server as well as general memory
leaks.

See http://sqlrelay.sourceforge.net/sqlrelay/faq.html#php for
PHP-specific info.  The same may apply to fastcgi.

Dave
dav...@fi...


On Thu, 2008-04-24 at 13:24 -0600, TazaTek wrote:
>  
> 
> Hello all!
> 
>  
> 
> I was wondering if there were any problems in mixing fastcgi and
> sqlrelay?  It seems that they both handle some of the speed
> performance mechanisms that are required of a web server, but I didn’t
> know if there were any problems in mixing the two.
> 
>  
> 
> Thanks
> 
>  
> 
> Matt
> 
>  
> 
>  
> 
>  
> 
> --
> 
> Matthew Kettlewell
> 
> President - TazaTek
> 
> www.TazaTek.com
> 
> www.Pairs-List.com
> 
> 970-985-4192
> 
>  
> 
>  
> 
> 
> __________________________________________________
> D O T E A S Y - "Join the web hosting revolution!"
> http://www.doteasy.com 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
> Don't miss this year's exciting event. There's still time to save $100. 
> Use priority code J8TL2D2. 
> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
> _______________________________________________ Sqlrelay-discussion mailing list Sql...@li... https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion



__________________________________________________
D O T E A S Y - "Join the web hosting revolution!"
             http://www.doteasy.com

[Sqlrelay-discussion] FastCgi and SqlRelay?

[TazaTek <ma...@ta...>]

 

Hello all!

 

I was wondering if there were any problems in mixing fastcgi and sqlrelay?
It seems that they both handle some of the speed performance mechanisms that
are required of a web server, but I didn't know if there were any problems
in mixing the two.

 

Thanks

 

Matt

 

 

 

--

Matthew Kettlewell

President - TazaTek

www.TazaTek.com

www.Pairs-List.com

970-985-4192

 

 

[Sqlrelay-discussion] BFILE?

[John K. <kl...@gm...>]

The list archive is mum on the topic of BFILE data...  A google search
within the sqlrelay.sourceforge.net site is also mum.

[ I saw what looked like a mirror of the SQL Relay pages, but on a
foreign server (http://doc.dvgu.ru/db/sql_relay/faq.html), that said
something about BFILE support.  I'm guessing that's old/invalid
somehow, though. ]

Anyway, when I use the non-DBI perl bindings with SQL Relay, and query
for a BFILE column, I get back what appears to be the right number of
bytes (might not be -- not sure), but all of the characters in the
data are nulls (they look like ^@ in Emacs :-)).  When I use the DBD
perl bindings with SQL Relay, and do the same thing, I get a one- or
two-character chunk of data back (despite long-read-length set high,
and truncate set to false).

If I use regular DBD::Oracle, without SQL Relay, I get the whole BFILE back.

So is BFILE supported, and if so, is there a trick? :-)

Thanks.

Re: [Sqlrelay-discussion] sqlrelay redirect question

[Peter C. <pc...@sa...>]

Hello David,

Many thanks for your reply, much appreciated. For real world usage I 
think I may be able to make a few generic rules for filtering as they 
will be database revision based (fingers crossed there :-) )

I had a play with the router example but hit the following errors. Ones 
of particular interest were the unknown dbase router and it didn't like 
an instance section without a router entry. I've attached a v. 
simplistic version of the config and output below (apologies if I'm 
missing something obvious here). On the plus side the connection pooling 
is working elsewhere like a charm (so many thanks there :-) )

sqlrelay version:
0.37.1-3.1 (with Debian Etch)

config file:
see attached

output:
Starting SQL Relay ... /etc/sqlrelay/sqlrelay.conf:14: element instance: 
validity error : Element instance content does not follow the DTD, 
expecting (users , connections , router), got (users connections )
        </instance>
                   ^
/etc/sqlrelay/sqlrelay.conf:24: element instance: validity error : 
Element instance content does not follow the DTD, expecting (users , 
connections , router), got (users connections )
        </instance>
                   ^
/etc/sqlrelay/sqlrelay.conf:27: element instance: validity error : Value 
"router" for attribute dbase of instance is not among the enumerated set
tringbindvaluelength="4000" maxlobbindvaluelength="71680" 
idleclienttimeout="-1"
                                                                               
^
/etc/sqlrelay/sqlrelay.conf:40: parser error : expected '>'
</instances>
^
SQL Relay not started/etc/sqlrelay/sqlrelay.conf:14: element instance: 
validity error : Element instance content does not follow the DTD, 
expecting (users , connections , router), got (users connections )
        </instance>
                   ^
/etc/sqlrelay/sqlrelay.conf:24: element instance: validity error : 
Element instance content does not follow the DTD, expecting (users , 
connections , router), got (users connections )
        </instance>
                   ^
/etc/sqlrelay/sqlrelay.conf:27: element instance: validity error : Value 
"router" for attribute dbase of instance is not among the enumerated set
tringbindvaluelength="4000" maxlobbindvaluelength="71680" 
idleclienttimeout="-1"
                                                                               
^
/etc/sqlrelay/sqlrelay.conf:40: parser error : expected '>'
</instances>
^
SQL Relay not started.

Many thabks again for your assistance

Pete


> SQL Relay can do this.  You need to set up 4 instances of SQL Relay, one
> to talk to each database.  Then you need to set up a 5th instance to
> "route" queries to the other 4.
>
> There's a doc at: http://sqlrelay.sourceforge.net/sqlrelay/router.html
> that describes this in detail.  The section "Provide a Single Point of
> Access to Multiple Databases" is the section that applies in this case.
> In that example, one of the db's is mysql and the other is postgresql,
> but you could just as easily use several mysql databases.
>
> The trick is that you need to come up with regular expressions so that
> the router can look at a query and know which database to send it to by
> which expression it matches.  That could be a long list of regular
> expressions for each db.
>
> Dave
>
> On Thu, 2008-04-17 at 11:45 +0100, Peter Clapham wrote:
>   
>> Dear all,
>>
>> Having just started playing with sqlrealy I apologise if this is a daft 
>> Q, I've browsed the archives but I'm not sure I've found how best to 
>> tackle the following problem.
>>
>> We have a number of mysql instances running on various hosts and ports. 
>> We would like to be able to rediect queries to a given instance 
>> depending upon the database requested. So in essence we'd have:
>>
>> server1  port 3306  database1
>> server2  port 3307  database2
>> server2  port 5307  database3
>> server3  port 5304  database4
>>
>> So user points connection at sqlrelay for a given database connection 
>> this is then (hopefully) redirected transparently to the required server .
>>
>> Pointers of best way forward would be much appreciated.
>>
>> Many thanks
>>
>> Pete
>>
>>
>>     
>
>
>
> __________________________________________________
> D O T E A S Y - "Join the web hosting revolution!"
>              http://www.doteasy.com
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
> Don't miss this year's exciting event. There's still time to save $100. 
> Use priority code J8TL2D2. 
> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
> _______________________________________________
> Sqlrelay-discussion mailing list
> Sql...@li...
> https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion
>   




-- 
 The Wellcome Trust Sanger Institute is operated by Genome Research 
 Limited, a charity registered in England with number 1021457 and a 
 company registered in England with number 2742969, whose registered 
 office is 215 Euston Road, London, NW1 2BE. 

Re: [Sqlrelay-discussion] sqlrelay redirect question

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

SQL Relay can do this.  You need to set up 4 instances of SQL Relay, one
to talk to each database.  Then you need to set up a 5th instance to
"route" queries to the other 4.

There's a doc at: http://sqlrelay.sourceforge.net/sqlrelay/router.html
that describes this in detail.  The section "Provide a Single Point of
Access to Multiple Databases" is the section that applies in this case.
In that example, one of the db's is mysql and the other is postgresql,
but you could just as easily use several mysql databases.

The trick is that you need to come up with regular expressions so that
the router can look at a query and know which database to send it to by
which expression it matches.  That could be a long list of regular
expressions for each db.

Dave

On Thu, 2008-04-17 at 11:45 +0100, Peter Clapham wrote:
> Dear all,
> 
> Having just started playing with sqlrealy I apologise if this is a daft 
> Q, I've browsed the archives but I'm not sure I've found how best to 
> tackle the following problem.
> 
> We have a number of mysql instances running on various hosts and ports. 
> We would like to be able to rediect queries to a given instance 
> depending upon the database requested. So in essence we'd have:
> 
> server1  port 3306  database1
> server2  port 3307  database2
> server2  port 5307  database3
> server3  port 5304  database4
> 
> So user points connection at sqlrelay for a given database connection 
> this is then (hopefully) redirected transparently to the required server .
> 
> Pointers of best way forward would be much appreciated.
> 
> Many thanks
> 
> Pete
> 
> 



__________________________________________________
D O T E A S Y - "Join the web hosting revolution!"
             http://www.doteasy.com

[Sqlrelay-discussion] sqlrelay redirect question

[Peter C. <pc...@sa...>]

Dear all,

Having just started playing with sqlrealy I apologise if this is a daft 
Q, I've browsed the archives but I'm not sure I've found how best to 
tackle the following problem.

We have a number of mysql instances running on various hosts and ports. 
We would like to be able to rediect queries to a given instance 
depending upon the database requested. So in essence we'd have:

server1  port 3306  database1
server2  port 3307  database2
server2  port 5307  database3
server3  port 5304  database4

So user points connection at sqlrelay for a given database connection 
this is then (hopefully) redirected transparently to the required server .

Pointers of best way forward would be much appreciated.

Many thanks

Pete


-- 
 The Wellcome Trust Sanger Institute is operated by Genome Research 
 Limited, a charity registered in England with number 1021457 and a 
 company registered in England with number 2742969, whose registered 
 office is 215 Euston Road, London, NW1 2BE. 

Re: [Sqlrelay-discussion] "Error was: No such file or directory" on freeBSD

[Culley H. <har...@gm...>]

My latest attempt this morning is to avoid the port and install from scratch.
During the make process (I had to use gmake) I get this error.  Any ideas?

====

g++ -o .libs/sqlr-listener .libs/sqlrlistener.o .libs/main.o -pthread
-L/usr/home/culley/Desktop/sqlrelay-0.39.4/src/util
/usr/home/culley/Desktop/sqlrelay-0.39.4/src/util/.libs/libsqlrutil.so
-L/usr/local/lib /usr/local/lib/librudiments.so -pthread -L/usr/lib -lcrypt
-Wl,--rpath -Wl,/usr/local/firstworks/lib -Wl,--rpath -Wl,/usr/local/lib
/usr/local/lib/librudiments.so: undefined reference to `SSL_shutdown'
/usr/local/lib/librudiments.so: undefined reference to `SSL_get_error'
/usr/local/lib/librudiments.so: undefined reference to `SSL_read'
/usr/local/lib/librudiments.so: undefined reference to `SSL_new'
/usr/local/lib/librudiments.so: undefined reference to `SSL_pending'
/usr/local/lib/librudiments.so: undefined reference to `SSL_set_bio'
/usr/local/lib/librudiments.so: undefined reference to `SSL_accept'
/usr/local/lib/librudiments.so: undefined reference to `SSL_connect'
/usr/local/lib/librudiments.so: undefined reference to `SSL_free'
/usr/local/lib/librudiments.so: undefined reference to `BIO_new_fd'
/usr/local/lib/librudiments.so: undefined reference to `SSL_write'
/usr/local/lib/librudiments.so: undefined reference to `BIO_new_socket'
gmake[2]: *** [sqlr-listener] Error 1
gmake[2]: Leaving directory
`/usr/home/culley/Desktop/sqlrelay-0.39.4/src/listener'
gmake[1]: *** [all] Error 2
gmake[1]: Leaving directory `/usr/home/culley/Desktop/sqlrelay-0.39.4/src'
gmake: *** [all] Error 2

On Wed, 09 Apr 2008, David Muse wrote:

> Usually this is permissions-related.  The user that sqlrelay is
> configured to run as can't write files to the directory that the ipc
> files are stored in.  In this case, probably /var/sqlrelay/tmp/ipc.
> Or possibly that directory doesn't exist.
> 
> If that's not the case, then you may need to do some stracing to see
> what's causing ftok to generate the -1 id.
> 
> Dave
> dav...@fi...
> 
> On Mon, 2008-04-07 at 22:01 -0700, Culley Harrelson wrote:
> > Hello, 
> > 
> > I am trying to configure sqlrelay on freebsd-- pc-bsd actually, my
> > development machine and eventually on my production freebsd servers.
> > Sqlrelay was installed from the port-- version sqlrelay-0.38_2
> > 
> > When I try to start it I get the below error.  Any ideas?  It seems like it
> > is looking for /var/sqlrelay.  In this directory there was no
> > /var/sqlrelay/var so I created it but it didn't fix the problem.  
> > 
> > ch
> > 
> > ===================
> > 
> > 
> > [root@pcbsd /usr/local/etc]# sqlr-start -id flashex
> > 
> > Starting listener:
> >   sqlr-listener -id flashex -config /usr/local/etc/sqlrelay.conf
> > 
> > sqlr-listener error:
> >         Unable to create a semaphore set.  This is usally because an
> >         sqlr-listener is already running for the flashex instance.
> > 
> >         If it is not running, something may have crashed and left an old semaphore set
> >         lying around.  Use the ipcs command to inspect existing semaphore sets
> >         and the ipcrm command to remove the semaphore set with
> >         id -1.
> > 
> >         Error was: No such file or directory
> > 
> > unsuccessful completion
> > 
> > sqlr-listener failed to start.
> > 
> > 
> > -------------------------------------------------------------------------
> > This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
> > Register now and save $200. Hurry, offer ends at 11:59 p.m., 
> > Monday, April 7! Use priority code J8TLD2. 
> > http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
> > _______________________________________________
> > Sqlrelay-discussion mailing list
> > Sql...@li...
> > https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion
> > 
> > 
> > __________________________________________________
> > D O T E A S Y - "Join the web hosting revolution!"
> >              http://www.doteasy.com
> 
> 
> 
> __________________________________________________
> D O T E A S Y - "Join the web hosting revolution!"
>              http://www.doteasy.com
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
> Don't miss this year's exciting event. There's still time to save $100. 
> Use priority code J8TL2D2. 
> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
> _______________________________________________
> Sqlrelay-discussion mailing list
> Sql...@li...
> https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion

Re: [Sqlrelay-discussion] "Error was: No such file or directory" on freeBSD

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

Usually this is permissions-related.  The user that sqlrelay is
configured to run as can't write files to the directory that the ipc
files are stored in.  In this case, probably /var/sqlrelay/tmp/ipc.
Or possibly that directory doesn't exist.

If that's not the case, then you may need to do some stracing to see
what's causing ftok to generate the -1 id.

Dave
dav...@fi...

On Mon, 2008-04-07 at 22:01 -0700, Culley Harrelson wrote:
> Hello, 
> 
> I am trying to configure sqlrelay on freebsd-- pc-bsd actually, my
> development machine and eventually on my production freebsd servers.
> Sqlrelay was installed from the port-- version sqlrelay-0.38_2
> 
> When I try to start it I get the below error.  Any ideas?  It seems like it
> is looking for /var/sqlrelay.  In this directory there was no
> /var/sqlrelay/var so I created it but it didn't fix the problem.  
> 
> ch
> 
> ===================
> 
> 
> [root@pcbsd /usr/local/etc]# sqlr-start -id flashex
> 
> Starting listener:
>   sqlr-listener -id flashex -config /usr/local/etc/sqlrelay.conf
> 
> sqlr-listener error:
>         Unable to create a semaphore set.  This is usally because an
>         sqlr-listener is already running for the flashex instance.
> 
>         If it is not running, something may have crashed and left an old semaphore set
>         lying around.  Use the ipcs command to inspect existing semaphore sets
>         and the ipcrm command to remove the semaphore set with
>         id -1.
> 
>         Error was: No such file or directory
> 
> unsuccessful completion
> 
> sqlr-listener failed to start.
> 
> 
> -------------------------------------------------------------------------
> This SF.net email is sponsored by the 2008 JavaOne(SM) Conference 
> Register now and save $200. Hurry, offer ends at 11:59 p.m., 
> Monday, April 7! Use priority code J8TLD2. 
> http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone
> _______________________________________________
> Sqlrelay-discussion mailing list
> Sql...@li...
> https://lists.sourceforge.net/lists/listinfo/sqlrelay-discussion
> 
> 
> __________________________________________________
> D O T E A S Y - "Join the web hosting revolution!"
>              http://www.doteasy.com



__________________________________________________
D O T E A S Y - "Join the web hosting revolution!"
             http://www.doteasy.com