naviserver-devel — Developer discussion

Re: [naviserver-devel] nsssl crash

[Vlad S. <vl...@cr...>]

It used to work fine but actually it was some time ago, let me load and 
check it locally, in the meantime backtrack would be helpful as well.

Daniel Stasinski wrote:
> Before I begin hunting down a crash involving nsssl from cvs, are
> there any known issues?
> 
> The server loads with no errors but crashes on the first ssl page access.
> 
> ns_section    ns/server/default/module/nsssl
> ns_param        port                443
> ns_param        address             192.168.0.10
> ns_param        hostname            secure.server.com
> ns_param        certificate
> /home/nsadmin/naviserver/servers/www.server.com/modules/nsssl/server.pem
> ns_param        ciphers
> "ALL:!ADH:!EXPORT56:RC4+RSA:+HIGH:+MEDIUM:+LOW:+SSLv2:+EXP"
> ns_param        protocols           "SSLv2, SSLv3, TLSv1"
> 
> Daniel
> 

Re: [naviserver-devel] nsphp.c error

[Daniel S. <moo...@av...>]

On Mon, May 5, 2008 at 10:16 AM, Vlad Seryakov <vl...@cr...> wrote:
> Just commited the fix, tested with MediaWiki, works fine, does not crash
>  on auth pages anymore.

Will update.  Thank you, Vlad.

Having run a strictly AOLserver shop for the last 10 years, I'll
looking forward to the direction of NaviServer.

Daniel

-- 
| ---------------------------------------------------------------
| Daniel P. Stasinski | http://www.saidsimple.com
| moo...@av... | http://www.disabilities-r-us.com
| XMMP: moo...@av... | http://www.avenues.org
| Google Talk: mooooooo | http://www.scriptkitties.com

[naviserver-devel] nsssl crash

[Daniel S. <moo...@av...>]

Before I begin hunting down a crash involving nsssl from cvs, are
there any known issues?

The server loads with no errors but crashes on the first ssl page access.

ns_section    ns/server/default/module/nsssl
ns_param        port                443
ns_param        address             192.168.0.10
ns_param        hostname            secure.server.com
ns_param        certificate
/home/nsadmin/naviserver/servers/www.server.com/modules/nsssl/server.pem
ns_param        ciphers
"ALL:!ADH:!EXPORT56:RC4+RSA:+HIGH:+MEDIUM:+LOW:+SSLv2:+EXP"
ns_param        protocols           "SSLv2, SSLv3, TLSv1"

Daniel

-- 
| ---------------------------------------------------------------
| Daniel P. Stasinski | http://www.saidsimple.com
| moo...@av... | http://www.disabilities-r-us.com
| XMMP: moo...@av... | http://www.avenues.org
| Google Talk: mooooooo | http://www.scriptkitties.com

[naviserver-devel] tmpnam issues

[Vlad S. <vl...@cr...>]

I guess Zoran uses Windows port in production.
Have you ever encounter any problems with tmpnam?

Do we need to work on thins for naviserver?

Re: [naviserver-devel] nsphp.c error

[Vlad S. <vl...@cr...>]

Just commited the fix, tested with MediaWiki, works fine, does not crash 
on auth pages anymore.

Also, upgraded to php-5.2.6

Daniel Stasinski wrote:
> When using nsphp with CVS version of naviserver, the following crash occurs:
> 
> /root/modules/nsphp/php-5.2.5/main/SAPI.c(424) : Block 0x08523200 status:
> Invalid pointer: ((thread_id=0x00000001) != (expected=0xB53FFB90))
> [04/May/2008:20:58:14][27361.b53ffb90][-conn:default:0] Fatal:
> received fatal signal 11
> Aborted
> 
> To duplicate, load a page that requires authentication, such as:
> http://something.com/nsconf.tcl and then a plain non-authenticated
> page like http://something.com/helloworld.php
> 
> Daniel
> 

Re: [naviserver-devel] nsphp.c error

[Vlad S. <vl...@cr...>]

I am working on it, i think i found the problem, i use ns_strcopy but 
should use estrdup, PHP does freeing. Will test and commit, i've never 
tested PHP with basic auth before.

Stephen Deasey wrote:
> On Mon, May 5, 2008 at 5:02 AM, Daniel Stasinski <moo...@av...> wrote:
>> When using nsphp with CVS version of naviserver, the following crash occurs:
>>
>> /root/modules/nsphp/php-5.2.5/main/SAPI.c(424) : Block 0x08523200 status:
>> Invalid pointer: ((thread_id=0x00000001) != (expected=0xB53FFB90))
>> [04/May/2008:20:58:14][27361.b53ffb90][-conn:default:0] Fatal:
>> received fatal signal 11
>> Aborted
>>
>> To duplicate, load a page that requires authentication, such as:
>> http://something.com/nsconf.tcl and then a plain non-authenticated
>> page like http://something.com/helloworld.php
>>
>> Daniel
> 
> 
> Here's some tips to help narrow down the cause of the crash:
> 
>   http://naviserver.sourceforge.net/w/Running_Tests
> 
> A backtrace would be very helpful.
> 
> 
> So would a short test case (the auth stuff you mentioned above).  You
> could do this by loading nsphp in the naviserver/tests/test.nscfg file
> and adding a new file tests/php.test, but even better would be to add
> this to the nsphp module itself.
> 
> nsphp doesn't already have the test harness set up, but you can use
> nsloopctl as an example of a module which does.
> 
>   http://naviserver.cvs.sourceforge.net/naviserver/modules/nsloopctl/
> 
> Basically you need to cut 'n paste the test targets from the Makefile,
> copy the tests/all.tcl into place, copy the tests/config.tcl and edit,
> then add a php.test file with your single test.
> 
> Send the output of 'cvs diff -u' to the list  (and your backtrace /
> bug hunting results).
> 
> 
> If you're uncomfortable fixing the bug someone else will take a look
> at it, but if you'd like to take a shot yourself we can guide you
> through it.
> 
> 
> Thanks!
> 
> -------------------------------------------------------------------------
> 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
> _______________________________________________
> naviserver-devel mailing list
> nav...@li...
> https://lists.sourceforge.net/lists/listinfo/naviserver-devel
> 

Re: [naviserver-devel] nsphp.c error

[Stephen D. <sd...@gm...>]

On Mon, May 5, 2008 at 5:02 AM, Daniel Stasinski <moo...@av...> wrote:
> When using nsphp with CVS version of naviserver, the following crash occurs:
>
> /root/modules/nsphp/php-5.2.5/main/SAPI.c(424) : Block 0x08523200 status:
> Invalid pointer: ((thread_id=0x00000001) != (expected=0xB53FFB90))
> [04/May/2008:20:58:14][27361.b53ffb90][-conn:default:0] Fatal:
> received fatal signal 11
> Aborted
>
> To duplicate, load a page that requires authentication, such as:
> http://something.com/nsconf.tcl and then a plain non-authenticated
> page like http://something.com/helloworld.php
>
> Daniel


Here's some tips to help narrow down the cause of the crash:

  http://naviserver.sourceforge.net/w/Running_Tests

A backtrace would be very helpful.


So would a short test case (the auth stuff you mentioned above).  You
could do this by loading nsphp in the naviserver/tests/test.nscfg file
and adding a new file tests/php.test, but even better would be to add
this to the nsphp module itself.

nsphp doesn't already have the test harness set up, but you can use
nsloopctl as an example of a module which does.

  http://naviserver.cvs.sourceforge.net/naviserver/modules/nsloopctl/

Basically you need to cut 'n paste the test targets from the Makefile,
copy the tests/all.tcl into place, copy the tests/config.tcl and edit,
then add a php.test file with your single test.

Send the output of 'cvs diff -u' to the list  (and your backtrace /
bug hunting results).


If you're uncomfortable fixing the bug someone else will take a look
at it, but if you'd like to take a shot yourself we can guide you
through it.


Thanks!

[naviserver-devel] nsphp.c error

[Daniel S. <moo...@av...>]

When using nsphp with CVS version of naviserver, the following crash occurs:

/root/modules/nsphp/php-5.2.5/main/SAPI.c(424) : Block 0x08523200 status:
Invalid pointer: ((thread_id=0x00000001) != (expected=0xB53FFB90))
[04/May/2008:20:58:14][27361.b53ffb90][-conn:default:0] Fatal:
received fatal signal 11
Aborted

To duplicate, load a page that requires authentication, such as:
http://something.com/nsconf.tcl and then a plain non-authenticated
page like http://something.com/helloworld.php

Daniel

-- 
| ---------------------------------------------------------------
| Daniel P. Stasinski | http://www.saidsimple.com
| moo...@av... | http://www.disabilities-r-us.com
| XMMP: moo...@av... | http://www.avenues.org
| Google Talk: mooooooo | http://www.scriptkitties.com

Re: [naviserver-devel] drop CVS

[Vlad S. <vl...@cr...>]

SF in general but CVS in particular, i remember offered to switch to SF 
SVN, but i could not find good enough reasons.

With CVS i know that it does not handle directories, renaming, moving 
and i do not even try to do it, contacting SF everytime to cleanup CVS 
is not very convenient either. So, at least moving to SVN will make it 
much easier to control our repository, we can still stick with SF, i am 
not insisting.

Vasiljevic Zoran wrote:
> On 03.05.2008, at 21:14, Vlad Seryakov wrote:
> 
>> i do not have big
>> enough reason beside my not-liking Sf service.
> 
> 
> Just to be clear... SF supports both CVS and SVN.
> Is the SF in general that you do not like or is
> it the CVS especially?
> 
> 
> -------------------------------------------------------------------------
> 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
> _______________________________________________
> naviserver-devel mailing list
> nav...@li...
> https://lists.sourceforge.net/lists/listinfo/naviserver-devel
> 

-- 
Vlad Seryakov
vl...@cr...
http://www.crystalballinc.com/vlad/

Re: [naviserver-devel] drop CVS

[Vasiljevic Z. <zv...@ar...>]

On 03.05.2008, at 21:14, Vlad Seryakov wrote:

> i do not have big
> enough reason beside my not-liking Sf service.


Just to be clear... SF supports both CVS and SVN.
Is the SF in general that you do not like or is
it the CVS especially?

Re: [naviserver-devel] drop CVS

[Vlad S. <vl...@cr...>]

We've talked about this in the past and nobody could come up with very 
good reason, so i guess this time it is the same, i do not have big 
enough reason beside my not-liking Sf service.

Vasiljevic Zoran wrote:
> On 02.05.2008, at 21:50, Vlad Seryakov wrote:
> 
>> Frankly, i would go from CVS to SVN/Mercurial/Git, whatever.
>> I still think SF CVS sucks, so i would switch anytime if my voice will
>> need to be counted:-)))
> 
> It seems as I'm the only one advocating continued use of CVS...
> And I have a simple reason: I do not see any benefit undertaking
> this (abandond-CVS) effort. What is to be gained? Or, perhaps,
> whar is that we  are loosing by sticking to CVS?
> 
> Zoran
> 
> 
> -------------------------------------------------------------------------
> 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
> _______________________________________________
> naviserver-devel mailing list
> nav...@li...
> https://lists.sourceforge.net/lists/listinfo/naviserver-devel
> 

Re: [naviserver-devel] drop CVS (was: [AOLSERVER] nsdbi database driver interface 0.1)

[Vasiljevic Z. <zv...@ar...>]

On 02.05.2008, at 21:50, Vlad Seryakov wrote:

> Frankly, i would go from CVS to SVN/Mercurial/Git, whatever.
> I still think SF CVS sucks, so i would switch anytime if my voice will
> need to be counted:-)))

It seems as I'm the only one advocating continued use of CVS...
And I have a simple reason: I do not see any benefit undertaking
this (abandond-CVS) effort. What is to be gained? Or, perhaps,
whar is that we  are loosing by sticking to CVS?

Zoran

Re: [naviserver-devel] [AOLSERVER] nsdbi database driver interface 0.1

[Vlad S. <vl...@cr...>]

Frankly, i would go from CVS to SVN/Mercurial/Git, whatever.
I still think SF CVS sucks, so i would switch anytime if my voice will 
need to be counted:-)))

The question i have, is freehg.org reliable and supposed to live long?
On other hand SF is very popular and search engine favorite, not that 
popularity is my big concern though.

Stephen Deasey wrote:
> (sorry for the delay)
> 
> 
> On Thu, Apr 17, 2008 at 3:49 PM, Vlad Seryakov <vl...@cr...> wrote:
>> This is very cool,
>>
>> Can you commit it into primary Naviserver CVS?
> 
> 
> Hmm, I hadn't thought about it much, I just wanted to get it off my
> hard drive. But now that you mention it, it would kind of suck to move
> it out of mercurial (and I'm not even sure there's a downward
> conversion tool...)
> 
> Btw. some time ago I also converted the aolserver and naviserver
> histories to mercurial.  I cleaned up the commits quite a lot so that
> logical changes appear in one changeset with comments in the right
> place in the right format, attributed to the right people etc.  I
> collapsed about 30% of the commits IIRC.
> 
> I did the same for naviserver then stuck it on to the end of the
> history of aolserver 4.0.10.  So, there's now complete history going
> back 8 years which is browsable and annotatable.  Check it out:
> 
> http://freehg.org/u/groks/naviserver/
> http://freehg.org/u/groks/aolserver/
> 
> 
> I don't want to agitate for change too strongly, but maybe it's worth
> thinking about?
> 
> 
> http://www.selenic.com/mercurial/wiki/
> 
> -------------------------------------------------------------------------
> 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
> _______________________________________________
> naviserver-devel mailing list
> nav...@li...
> https://lists.sourceforge.net/lists/listinfo/naviserver-devel
> 

Re: [naviserver-devel] [AOLSERVER] nsdbi database driver interface 0.1

[Stephen D. <sd...@gm...>]

(sorry for the delay)


On Thu, Apr 17, 2008 at 3:49 PM, Vlad Seryakov <vl...@cr...> wrote:
> This is very cool,
>
> Can you commit it into primary Naviserver CVS?


Hmm, I hadn't thought about it much, I just wanted to get it off my
hard drive. But now that you mention it, it would kind of suck to move
it out of mercurial (and I'm not even sure there's a downward
conversion tool...)

Btw. some time ago I also converted the aolserver and naviserver
histories to mercurial.  I cleaned up the commits quite a lot so that
logical changes appear in one changeset with comments in the right
place in the right format, attributed to the right people etc.  I
collapsed about 30% of the commits IIRC.

I did the same for naviserver then stuck it on to the end of the
history of aolserver 4.0.10.  So, there's now complete history going
back 8 years which is browsable and annotatable.  Check it out:

http://freehg.org/u/groks/naviserver/
http://freehg.org/u/groks/aolserver/


I don't want to agitate for change too strongly, but maybe it's worth
thinking about?


http://www.selenic.com/mercurial/wiki/

Re: [naviserver-devel] [naviserver-commits] naviserver/nsd driver.c, 1.97, 1.98

[Vasiljevic Z. <zv...@ar...>]

On 26.04.2008, at 23:46, Vlad Seryakov wrote:

> I think it was working perfectly fine, new buffer size is set on
> accepted socket, new one, not the one we listened.
>
>
> I used to play with these options for one of the modules i wrote and  
> at
> least on Linux it was working fine.

OK. I will back this off. I was unable to get that working
on my platforms. So I thought to comment this and move
close to the place where it *might* work.

Good to know that this works on Linux at least...

Zoran

Re: [naviserver-devel] [naviserver-commits] naviserver/nsd driver.c, 1.97, 1.98

[Vlad S. <vl...@cr...>]

I think it was working perfectly fine, new buffer size is set on 
accepted socket, new one, not the one we listened.


I used to play with these options for one of the modules i wrote and at 
least on Linux it was working fine.

Also i found this about this option:
You can increase a stream socket's buffer size at any time, but decrease 
it only prior to establishing a connection. The maximum buffer size for 
stream sockets is 262144 bytes.



Zoran Vasiljevic wrote:
> Update of /cvsroot/naviserver/naviserver/nsd
> In directory sc8-pr-cvs16.sourceforge.net:/tmp/cvs-serv8949/nsd
> 
> Modified Files:
> 	driver.c 
> Log Message:
> nsd/driver.c: Socket send/recv buffer setup moved to NsStartDrivers()
> and commented as real NO-OP, as too late in the startup sequence.
> TCP honours the buffer-size setup prior to listen() or connect() only.
> Without significant API plumbing, the per-driver values for send and
> receive buffers are just useless.
> 
> 
> Index: driver.c
> ===================================================================
> RCS file: /cvsroot/naviserver/naviserver/nsd/driver.c,v
> retrieving revision 1.97
> retrieving revision 1.98
> diff -C2 -d -r1.97 -r1.98
> *** driver.c	18 Dec 2007 18:21:39 -0000	1.97
> --- driver.c	26 Apr 2008 19:47:52 -0000	1.98
> ***************
> *** 551,627 ****
>   NsStartDrivers(void)
>   {
> !     Driver *drvPtr;
> !     SpoolerQueue *queuePtr;
>   
>       /*
> !      * Listen on all drivers.
>        */
>   
>       drvPtr = firstDrvPtr;
>   
>       while (drvPtr != NULL) {
> -         if (drvPtr->opts & NS_DRIVER_UNIX) {
> -             drvPtr->sock = Ns_SockListenUnix(drvPtr->bindaddr,
> -                                              drvPtr->opts & NS_DRIVER_UDP ? 0 :
> -                                                             drvPtr->backlog, 0);
>   
> !         } else if (drvPtr->opts & NS_DRIVER_UDP) {
> !             drvPtr->sock = Ns_SockListenUdp(drvPtr->bindaddr,
> !                                             drvPtr->port);
>   
>           } else {
> !             drvPtr->sock = Ns_SockListenEx(drvPtr->bindaddr,
> !                                            drvPtr->port, drvPtr->backlog);
>           }
>           if (drvPtr->sock == INVALID_SOCKET) {
> !             Ns_Log(Error, "%s: failed to listen on %s:%d: %s",
> !                    drvPtr->name, drvPtr->address, drvPtr->port,
> !                    ns_sockstrerror(ns_sockerrno));
>           } else {
>   
>               Ns_SockSetNonBlocking(drvPtr->sock);
> -             Ns_Log(Notice, "%s: listening on %s:%d",
> -                    drvPtr->name, drvPtr->address, drvPtr->port);
>   
>               /*
> !              * Create the spooler thread(s).
>                */
>   
> !             queuePtr = drvPtr->spooler.firstPtr;
> !             while (queuePtr) {
> !                 if (ns_sockpair(queuePtr->pipe) != 0) {
> !                     Ns_Fatal("driver: ns_sockpair() failed: %s",
> !                              ns_sockstrerror(ns_sockerrno));
> !                 }
> !                 Ns_ThreadCreate(SpoolerThread, queuePtr, 0,
> !                                 &queuePtr->thread);
> !                 queuePtr = queuePtr->nextPtr;
>               }
>   
>               /*
> !              * Create the writer thread(s)
>                */
>   
> !             queuePtr = drvPtr->writer.firstPtr;
> !             while (queuePtr) {
> !                 if (ns_sockpair(queuePtr->pipe) != 0) {
> !                     Ns_Fatal("driver: ns_sockpair() failed: %s",
> !                              ns_sockstrerror(ns_sockerrno));
>                   }
> -                 Ns_ThreadCreate(WriterThread, queuePtr, 0,
> -                                 &queuePtr->thread);
> -                 queuePtr = queuePtr->nextPtr;
>               }
>           }
>           drvPtr = drvPtr->nextPtr;
>       }
>   
>       /*
> !      * Create the driver thread.
>        */
>   
> !     if (firstDrvPtr != NULL) {
> !         if (ns_sockpair(drvPipe) != 0) {
> !             Ns_Fatal("driver: ns_sockpair() failed: %s",
>                        ns_sockstrerror(ns_sockerrno));
>           }
> --- 551,648 ----
>   NsStartDrivers(void)
>   {
> !     int          i, bl, ncommd;
> !     Driver       *drvPtr;
> !     SpoolerQueue *qPtr;
> ! 
> !     struct {
> !         Ns_ThreadProc *tproc;
> !         SpoolerQueue  *queue;
> !     } thr[] = {{SpoolerThread, NULL}, {WriterThread,  NULL}};
>   
>       /*
> !      * Listen on all communication drivers.
>        */
>   
>       drvPtr = firstDrvPtr;
> +     ncommd = 0;
>   
>       while (drvPtr != NULL) {
>   
> !         bl = (drvPtr->opts & NS_DRIVER_UDP) ? 0 : drvPtr->backlog;
>   
> +         if (drvPtr->opts & NS_DRIVER_UNIX) {
> +             drvPtr->sock = Ns_SockListenUnix(drvPtr->bindaddr, bl, 0);
> +         } else if (drvPtr->opts & NS_DRIVER_UDP) {
> +             drvPtr->sock = Ns_SockListenUdp(drvPtr->bindaddr, drvPtr->port);
>           } else {
> !             drvPtr->sock = Ns_SockListenEx(drvPtr->bindaddr, drvPtr->port, bl);
>           }
>           if (drvPtr->sock == INVALID_SOCKET) {
> !             Ns_Log(Error, "%s: failed to listen on %s:%d: %s", drvPtr->name,
> !                    drvPtr->address, drvPtr->port, ns_sockstrerror(ns_sockerrno));
>           } else {
>   
>               Ns_SockSetNonBlocking(drvPtr->sock);
>   
>               /*
> !              * Set the send/recv socket bufsizes if required.
> !              * Note: it's too late to set them here, as TCP manual says:
> !              *
> !              *     On individual connections, the socket buffer size 
> !              *     must be set prior to the listen() or connect() calls
> !              *     in order to have it take effect.
> !              *
> !              * So now what? The binder process already did all of that for
> !              * us early on startup! So whatever we do here below does not
> !              * really matter.  We must move this early in the binding process
> !              * somehow, but that requires significant API plumbing... Phew.
> !              *
>                */
>   
> !             if (drvPtr->sndbuf > 0) {
> !                 setsockopt(drvPtr->sock, SOL_SOCKET, SO_SNDBUF,
> !                            (char *) &drvPtr->sndbuf, sizeof(drvPtr->sndbuf));
> !             }
> !             if (drvPtr->rcvbuf > 0) {
> !                 setsockopt(drvPtr->sock, SOL_SOCKET, SO_RCVBUF,
> !                            (char *) &drvPtr->rcvbuf, sizeof(drvPtr->rcvbuf));
>               }
>   
> +             Ns_Log(Notice, "%s: listening on %s:%d", drvPtr->name,
> +                    drvPtr->address, drvPtr->port);
> + 
> +             ncommd++; /* Yet another successfuly started comm driver */
> + 
>               /*
> !              * Create the spooler/writer thread(s).
>                */
>   
> !             thr[0].queue = drvPtr->spooler.firstPtr;
> !             thr[1].queue = drvPtr->writer.firstPtr;
> ! 
> !             for (i = 0; i < 2; i++) {
> !                 qPtr = thr[i].queue;
> !                 while (qPtr != NULL) {
> !                     if (ns_sockpair(qPtr->pipe)) {
> !                         Ns_Fatal("ns_sockpair() failed: %s",
> !                                  ns_sockstrerror(ns_sockerrno));
> !                     }
> !                     Ns_ThreadCreate(thr[i].tproc, qPtr, 0, &qPtr->thread);
> !                     qPtr = qPtr->nextPtr;
>                   }
>               }
>           }
> + 
>           drvPtr = drvPtr->nextPtr;
>       }
>   
>       /*
> !      * Create the driver thread itself if we managed to
> !      * get at least one of the communication drivers.
>        */
>   
> !     if (ncommd > 0) {
> !         if (ns_sockpair(drvPipe)) {
> !             Ns_Fatal("ns_sockpair() failed: %s",
>                        ns_sockstrerror(ns_sockerrno));
>           }
> ***************
> *** 1678,1694 ****
>       Ns_SockSetNonBlocking(sockPtr->sock);
>   
> -     /*
> -      * Set the send/recv socket bufsizes if required.
> -      */
> - 
> -     if (drvPtr->sndbuf > 0) {
> -         setsockopt(sockPtr->sock, SOL_SOCKET, SO_SNDBUF,
> -                    (char *) &drvPtr->sndbuf, sizeof(drvPtr->sndbuf));
> -     }
> -     if (drvPtr->rcvbuf > 0) {
> -         setsockopt(sockPtr->sock, SOL_SOCKET, SO_RCVBUF,
> -                    (char *) &drvPtr->rcvbuf, sizeof(drvPtr->rcvbuf));
> -     }
> - 
>       drvPtr->queuesize++;
>   
> --- 1699,1702 ----
> 
> 
> -------------------------------------------------------------------------
> 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
> _______________________________________________
> naviserver-commits mailing list
> nav...@li...
> https://lists.sourceforge.net/lists/listinfo/naviserver-commits
> 

-- 
Vlad Seryakov
vl...@cr...
http://www.crystalballinc.com/vlad/

Re: [naviserver-devel] ns_job cancellation

[Vlad S. <vl...@cr...>]

Very nice feature, very useful

Vasiljevic Zoran wrote:
> Hi !
> 
> I have improved semantics of the [ns_job cancel] to cancel
> the script evaluation in the job-thread by means of
> Tcl_AsyncMark et.al. facility. In addition to marking the
> job ticket as cancelled, that is.
> 
> I hope nobody will have problems with that. If yes, we can
> change the command to add "-force" option and cancel the
> script only if -force option is specified.
> 
> Overall... I was pretty busy in last few months so I was not
> really able to add much to the code. Hope this will change
> in near future. Also, I believe we have *superb* code and tons
> of features. This simply deserves V5 release! Great job that
> Vlad and Stephen have done in the meantime. Thanks guys!
> 
> Cheers,
> Zoran 
> 
> 
> -------------------------------------------------------------------------
> 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
> _______________________________________________
> naviserver-devel mailing list
> nav...@li...
> https://lists.sourceforge.net/lists/listinfo/naviserver-devel
> 

-- 
Vlad Seryakov
vl...@cr...
http://www.crystalballinc.com/vlad/

[naviserver-devel] ns_job cancellation

[Vasiljevic Z. <zv...@ar...>]

Hi !

I have improved semantics of the [ns_job cancel] to cancel
the script evaluation in the job-thread by means of
Tcl_AsyncMark et.al. facility. In addition to marking the
job ticket as cancelled, that is.

I hope nobody will have problems with that. If yes, we can
change the command to add "-force" option and cancel the
script only if -force option is specified.

Overall... I was pretty busy in last few months so I was not
really able to add much to the code. Hope this will change
in near future. Also, I believe we have *superb* code and tons
of features. This simply deserves V5 release! Great job that
Vlad and Stephen have done in the meantime. Thanks guys!

Cheers,
Zoran 

[naviserver-devel] nsdbi database driver interface 0.1

[Stephen D. <sd...@gm...>]

nsdbi is a database driver interface for naviserver. It provides
native bind variable support, transparent prepared queries and handle
management, runtime configuration, statistics, and a few other things.

There are drivers for postgres, mysql and sqlite. There's also a stub
driver included for testing, fault injection etc.

It relies on a few utility routines in naviserver, but it wouldn't be
too hard to port those to aolserver, should anyone want to.


Here's a snapshot of the man page:

  http://www.groks.org/nsdbi.n.html

Tarballs:

  http://www.groks.org/nsdbi-0.1.tgz
  http://www.groks.org/nsdbipg-0.1.tgz
  http://www.groks.org/nsdbimy-0.1.tgz
  http://www.groks.org/nsdbilite-0.1.tgz

Mercurial repositories:

  http://freehg.org/u/groks/nsdbi/
  http://freehg.org/u/groks/nsdbipg/
  http://freehg.org/u/groks/nsdbimy/
  http://freehg.org/u/groks/nsdbilite/

Re: [naviserver-devel] Win32 version released

[Vlad S. <vl...@cr...>]

SF admin page says Wiki is disabled

Bernd Eidenschink wrote:
>> Pre-complied 4.99.2 version released on SF download page.
> 
> Thanks Vlad!
> 
> I added a link on the Wiki and deleted some spam on various pages. We should 
> consider turning of self-registration, otherwise this is only a temporary 
> fix. Maybe someone with admin-privileges on the wiki can do that?
> 
> Bernd.
> 
> -------------------------------------------------------------------------
> 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/
> _______________________________________________
> naviserver-devel mailing list
> nav...@li...
> https://lists.sourceforge.net/lists/listinfo/naviserver-devel
> 

Re: [naviserver-devel] Win32 version released

[Bernd E. <eid...@we...>]

> Pre-complied 4.99.2 version released on SF download page.

Thanks Vlad!

I added a link on the Wiki and deleted some spam on various pages. We should 
consider turning of self-registration, otherwise this is only a temporary 
fix. Maybe someone with admin-privileges on the wiki can do that?

Bernd.

[naviserver-devel] Win32 version released

[Vlad S. <vl...@cr...>]

Pre-complied 4.99.2 version released on SF download page.

-- 
Vlad Seryakov
vl...@cr...
http://www.crystalballinc.com/vlad/

Re: [naviserver-devel] mingw makefile problem

[Vlad S. <vl...@cr...>]

Problem solved, naviserver now compiles and runs successfully under 
Windows when built using mingw compiler.

All modules work, also i tested Postgres driver, works fined as well.

I think publishing somewhere binaries for win32, this way it is very 
easy to download and test the server, just not sure where to put the archive

Vlad Seryakov wrote:
> to the makefile gurus out there, i almost pulled all my remaining hair 
> trying to figure out this:
> 
> i am compailing naviserver under mingw on window, all compiles fin until 
> i get into any modules.
> 
> here is the problem
> 
> # cd nssock
> 
> # gmake -d
> GNU Make 3.81
> Copyright (C) 2006  Free Software Foundation, Inc.
> This is free software; see the source for copying conditions.
> There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
> PARTICULAR PURPOSE.
> 
> This program built for i686-pc-mingw32
> find_and_set_shell path search set default_shell = C:/msys/bin/sh.exe
> Reading makefiles...
> Reading makefile `makefile'...
> Reading makefile `../include/Makefile.build' (search path) (no ~ 
> expansion)...
> Reading makefile `../include/Makefile.module' (search path) (no ~ 
> expansion)...
> Reading makefile `../include/Makefile.global' (search path) (no ~ 
> expansion)...
> CreateProcess(C:\msys\bin\uname.exe,uname -a,...)
> Main thread handle = 0x000007a0
> CreateProcess(C:\msys\bin\uname.exe,uname -a,...)
> ../include/Makefile.module:126: *** multiple target patterns.  Stop.
> 
> # gmake -v
> GNU Make 3.81
> 
> Makefile.module is our standard makefile, works fine under Linux.
> 

-- 
Vlad Seryakov
vl...@cr...
http://www.crystalballinc.com/vlad/

[naviserver-devel] mingw makefile problem

[Vlad S. <vl...@cr...>]

to the makefile gurus out there, i almost pulled all my remaining hair 
trying to figure out this:

i am compailing naviserver under mingw on window, all compiles fin until 
i get into any modules.

here is the problem

# cd nssock

# gmake -d
GNU Make 3.81
Copyright (C) 2006  Free Software Foundation, Inc.
This is free software; see the source for copying conditions.
There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A
PARTICULAR PURPOSE.

This program built for i686-pc-mingw32
find_and_set_shell path search set default_shell = C:/msys/bin/sh.exe
Reading makefiles...
Reading makefile `makefile'...
Reading makefile `../include/Makefile.build' (search path) (no ~ 
expansion)...
Reading makefile `../include/Makefile.module' (search path) (no ~ 
expansion)...
Reading makefile `../include/Makefile.global' (search path) (no ~ 
expansion)...
CreateProcess(C:\msys\bin\uname.exe,uname -a,...)
Main thread handle = 0x000007a0
CreateProcess(C:\msys\bin\uname.exe,uname -a,...)
../include/Makefile.module:126: *** multiple target patterns.  Stop.

# gmake -v
GNU Make 3.81

Makefile.module is our standard makefile, works fine under Linux.

Re: [naviserver-devel] naviserver wiki defaced

[Bernd E. <eid...@we...>]

> The NaviServer wiki has been defaced so bad by spammers that it is
> unusable.

We should consider turning off the self-registration feature - before undoing 
the changes, i don't think it's necessary on the Naviserver wiki...

Bernd.