naviserver-devel — Developer discussion

Re: [naviserver-devel] IPv6 for NaviServer

[Gustaf N. <ne...@wu...>]

Hi Maurizio,

thanks for the fixes. as such, they could not work, but they point
out things, where one has to watch out. I have fixed these
things in a way it could work under windows.

By looking at the differences, it seems as if you compiled just a
IPv4 version (no IPv6 enabled). i would have expected this
to go more smoothly. In case you tried the sample
nsd-config.tcl script, that could not work with IPv4, since
IPv6 addresses were hard-coded.

all the best
-g

Am 24.02.16 um 18:00 schrieb Maurizio Martignano:
> Dear Gustaf,
> 	Thank you as always for all your work.
>
> About Windows I cannot really talk about MinGW nor Cygwin. I'm just using
> Visual Studio 2015 with 64 bit as target.
> With this configuration I tried to modify the file sockaddr.c (to make it
> compile), but I'm not sure about the changes I made.
> I also had to modify the file nswin32.c as I could not find anywhere the
> function "NsSockGetPort".
> Well the resulting system compiles... but nothing works.
> I need much more time to do a proper testing/tuning and I can't do that at
> the moment.
> As soon as I discover more, I'll let you know.
>
> Back to your delivery of the 16th of February.
>
> Once again thank you,
> Maurizio
>
>
> -----Original Message-----
> From: Gustaf Neumann [mailto:ne...@wu...]
> Sent: 24 February 2016 14:01
> To: Navidevel <nav...@li...>
> Subject: [naviserver-devel] IPv6 for NaviServer
>
> Dear all,
>
> over the last week, i've worked on IPv6 support for NaviServer.
> For this task, i took the following path:
> 1) build a version of NaviServer for IPv6
> 2) build a single code source for compiling either for IPv4 or IPv6
> 3) let the IPv6 version run the setup-files for IPv4 in the IPv6 version
> 4) Run regression test suite under IPv4 and/IPv6
> 5) Allow mixed mode (handle IPv4 and IPv6 by the same binary)
>      (requires some interface changes to handle generic sockaddrs)
>
> Everything seems to work fine, one can work with the same config files as
> before, and everything works as expected, but one can as well listen on IPv6
> addresses (e.g.
> the IPv6 unspecified address "::"). IPv4->6 mapping is activated, or the
> kernel allows to deliver as well
> IPv4 traffic to the IPv6 interface, then one can talk also multiple
> protocols. The mileage may vary depending on the OS and kernel.
>
> It is as well possible to deactivate IPv6 support at compile time via
> --disable-ipv6.
>
> Here is a summary of the necessary changes:
> - Added a amall Abstraction API for IP-version agnostic calls:
>     Ns_LogSockaddr(), Ns_SockaddrGetPort(),
>     Ns_SockaddrGetSockLen(), Ns_SockaddrMask(), Ns_SockaddrMaskBits(),
>     Ns_SockaddrSetPort(), ns_inet_ntop(), ns_inet_pton()
>
> - Replaced in interface more places where "sockaddr_in" (the IPv4
>     version of the sockaddr) with the generic (version agnostic)
>     sockaddr (also in IPv4 code to obtain API consistency when no IPv6
>     is available): Ns_BindSock(), Ns_ConnSetPeer(), Ns_GetSockAddr(),
>     Ns_SockBind(), Ns_SockBindUdp()
>
> - The Ns_Sock structure contained sockaddr_in, which is not sufficient
>     to store IPv6 addresses. Therefore it was necessary to change
>     that to "sockaddr_storage" which guarantees to be large enough
>     to keep all kind of addresses. This affects as well driver modules
>     which communicate via the NaviServer managed sockets. Old drivers
>     might crash with the new structure. Therefore, I've bumped
>     NS_DRIVER_VERSION to 3 such that driver modules (also binaries)
>     can communicate that they support IPv6. In IPv6 mode, NaviServer
>     refuses to work with versions <3.
>
> - Replaced all usages of ns_inet_ntoa() by ns_intet_ntop()
>     (except for legacy an minimal library support)
>
> - Added log file warnings on various places, where NaviServer
>     was ignoring error states in socket communications silently.
>
> - Transformed hash table for open listen ports to string keys
>     (the old code used 32bit IPv4-addresses as keys in hash tables)
> - added test for ns_listencallback
>
> - Ns_HttpParseHost: new abstraction for parsing host and port number.
>     Reason: IPv6 literal notation contains many colons, parsing
>     IP-literal notation according to RFC 3986 section 3.2.2 is necessary
>     (removed places where code was searching for ':' to expect the port)
>
> - Added [ns_info ipv6] to obtain information whether the binary
>     supports IPv6 or not
>
> - Added configure flag --disable-ipv6 to build NaviServer just with
>     IPv4 support.
>
> - provide defines for core differences between IPv4 and IPv6 #ifdef
> HAVE_IPV6
> # define NS_IP_LOOPBACK     "::1"
> # define NS_IP_UNSPECIFIED  "::"
> # define NS_SOCKADDR_IN     sockaddr_storage
> # define NS_IPADDR_SIZE     INET6_ADDRSTRLEN
> #else
> # define NS_IP_LOOPBACK     "127.0.0.1"
> # define NS_IP_UNSPECIFIED  "0.0.0.0"
> # define NS_SOCKADDR_IN     sockaddr_in
> # define NS_IPADDR_SIZE     INET_ADDRSTRLEN
> #endif
>
>    - nsperm:
>     * Aligned implementation with the documentation
>       (hosts.allow/deny have comma-separated entries)
>     * Added submask specification in the form /xxx
>       (e.g. ::1/64)
>     * Use binary IPv4 and IPv6 values for hash and mask lookup
>     * Generalizing masking functions
>     * Refactor code to avoid memory leaks
>     * Updated documentation
>
> - modules:
>     * nsudp and nsssl are now IPv6 aware/compliant
>
> TODOs:
> - more platform testing
>     (probably changes for Windows necessary, help is appreciated)
> - remove commented out debug statements
>
> - make more modules IPv6 compliant
> - documentation updates
>
> I've tested the generic and IPv4 only version via regression test and random
> browsing on Mac OS X and Linux. If you have other platforms, testing would
> be appreciated. There are some adjustments for windows probably necessary
> (in the absence of "configure", one has to add manually "HAVE_IPV4" to the
> compile flags). Any help is welcome.
>
> best regards
> -gustaf neumann
>
>
> ----------------------------------------------------------------------------
> --
> Site24x7 APM Insight: Get Deep Visibility into Application Performance APM +
> Mobile APM + RUM: Monitor 3 App instances at just $35/Month Monitor
> end-to-end web transactions and take corrective actions now Troubleshoot
> faster and improve end-user experience. Signup Now!
> http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
> _______________________________________________
> naviserver-devel mailing list
> nav...@li...
> https://lists.sourceforge.net/lists/listinfo/naviserver-devel
>
>
> ------------------------------------------------------------------------------
> Site24x7 APM Insight: Get Deep Visibility into Application Performance
> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
> Monitor end-to-end web transactions and take corrective actions now
> Troubleshoot faster and improve end-user experience. Signup Now!
> http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
>
>
> _______________________________________________
> naviserver-devel mailing list
> nav...@li...
> https://lists.sourceforge.net/lists/listinfo/naviserver-devel


-- 
Univ.Prof. Dr. Gustaf Neumann
WU Vienna
Institute of Information Systems and New Media
Welthandelsplatz 1, A-1020 Vienna, Austria

Re: [naviserver-devel] IPv6 for NaviServer

[Maurizio M. <Mau...@sp...>]

Dear Gustaf,
	Thank you as always for all your work.

About Windows I cannot really talk about MinGW nor Cygwin. I'm just using
Visual Studio 2015 with 64 bit as target.
With this configuration I tried to modify the file sockaddr.c (to make it
compile), but I'm not sure about the changes I made.
I also had to modify the file nswin32.c as I could not find anywhere the
function "NsSockGetPort".
Well the resulting system compiles... but nothing works.
I need much more time to do a proper testing/tuning and I can't do that at
the moment.
As soon as I discover more, I'll let you know.

Back to your delivery of the 16th of February.

Once again thank you,
Maurizio


-----Original Message-----
From: Gustaf Neumann [mailto:ne...@wu...] 
Sent: 24 February 2016 14:01
To: Navidevel <nav...@li...>
Subject: [naviserver-devel] IPv6 for NaviServer

Dear all,

over the last week, i've worked on IPv6 support for NaviServer.
For this task, i took the following path:
1) build a version of NaviServer for IPv6
2) build a single code source for compiling either for IPv4 or IPv6
3) let the IPv6 version run the setup-files for IPv4 in the IPv6 version
4) Run regression test suite under IPv4 and/IPv6
5) Allow mixed mode (handle IPv4 and IPv6 by the same binary)
    (requires some interface changes to handle generic sockaddrs)

Everything seems to work fine, one can work with the same config files as
before, and everything works as expected, but one can as well listen on IPv6
addresses (e.g.
the IPv6 unspecified address "::"). IPv4->6 mapping is activated, or the
kernel allows to deliver as well
IPv4 traffic to the IPv6 interface, then one can talk also multiple
protocols. The mileage may vary depending on the OS and kernel.

It is as well possible to deactivate IPv6 support at compile time via
--disable-ipv6.

Here is a summary of the necessary changes:
- Added a amall Abstraction API for IP-version agnostic calls:
   Ns_LogSockaddr(), Ns_SockaddrGetPort(),
   Ns_SockaddrGetSockLen(), Ns_SockaddrMask(), Ns_SockaddrMaskBits(),
   Ns_SockaddrSetPort(), ns_inet_ntop(), ns_inet_pton()

- Replaced in interface more places where "sockaddr_in" (the IPv4
   version of the sockaddr) with the generic (version agnostic)
   sockaddr (also in IPv4 code to obtain API consistency when no IPv6
   is available): Ns_BindSock(), Ns_ConnSetPeer(), Ns_GetSockAddr(),
   Ns_SockBind(), Ns_SockBindUdp()

- The Ns_Sock structure contained sockaddr_in, which is not sufficient
   to store IPv6 addresses. Therefore it was necessary to change
   that to "sockaddr_storage" which guarantees to be large enough
   to keep all kind of addresses. This affects as well driver modules
   which communicate via the NaviServer managed sockets. Old drivers
   might crash with the new structure. Therefore, I've bumped
   NS_DRIVER_VERSION to 3 such that driver modules (also binaries)
   can communicate that they support IPv6. In IPv6 mode, NaviServer
   refuses to work with versions <3.

- Replaced all usages of ns_inet_ntoa() by ns_intet_ntop()
   (except for legacy an minimal library support)

- Added log file warnings on various places, where NaviServer
   was ignoring error states in socket communications silently.

- Transformed hash table for open listen ports to string keys
   (the old code used 32bit IPv4-addresses as keys in hash tables)
- added test for ns_listencallback

- Ns_HttpParseHost: new abstraction for parsing host and port number.
   Reason: IPv6 literal notation contains many colons, parsing
   IP-literal notation according to RFC 3986 section 3.2.2 is necessary
   (removed places where code was searching for ':' to expect the port)

- Added [ns_info ipv6] to obtain information whether the binary
   supports IPv6 or not

- Added configure flag --disable-ipv6 to build NaviServer just with
   IPv4 support.

- provide defines for core differences between IPv4 and IPv6 #ifdef
HAVE_IPV6
# define NS_IP_LOOPBACK     "::1"
# define NS_IP_UNSPECIFIED  "::"
# define NS_SOCKADDR_IN     sockaddr_storage
# define NS_IPADDR_SIZE     INET6_ADDRSTRLEN
#else
# define NS_IP_LOOPBACK     "127.0.0.1"
# define NS_IP_UNSPECIFIED  "0.0.0.0"
# define NS_SOCKADDR_IN     sockaddr_in
# define NS_IPADDR_SIZE     INET_ADDRSTRLEN
#endif

  - nsperm:
   * Aligned implementation with the documentation
     (hosts.allow/deny have comma-separated entries)
   * Added submask specification in the form /xxx
     (e.g. ::1/64)
   * Use binary IPv4 and IPv6 values for hash and mask lookup
   * Generalizing masking functions
   * Refactor code to avoid memory leaks
   * Updated documentation

- modules:
   * nsudp and nsssl are now IPv6 aware/compliant

TODOs:
- more platform testing
   (probably changes for Windows necessary, help is appreciated)
- remove commented out debug statements

- make more modules IPv6 compliant
- documentation updates

I've tested the generic and IPv4 only version via regression test and random
browsing on Mac OS X and Linux. If you have other platforms, testing would
be appreciated. There are some adjustments for windows probably necessary
(in the absence of "configure", one has to add manually "HAVE_IPV4" to the
compile flags). Any help is welcome.

best regards
-gustaf neumann


----------------------------------------------------------------------------
--
Site24x7 APM Insight: Get Deep Visibility into Application Performance APM +
Mobile APM + RUM: Monitor 3 App instances at just $35/Month Monitor
end-to-end web transactions and take corrective actions now Troubleshoot
faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140
_______________________________________________
naviserver-devel mailing list
nav...@li...
https://lists.sourceforge.net/lists/listinfo/naviserver-devel

[naviserver-devel] IPv6 for NaviServer

[Gustaf N. <ne...@wu...>]

Dear all,

over the last week, i've worked on IPv6 support for NaviServer.
For this task, i took the following path:
1) build a version of NaviServer for IPv6
2) build a single code source for compiling either for IPv4 or IPv6
3) let the IPv6 version run the setup-files for IPv4 in the IPv6 version
4) Run regression test suite under IPv4 and/IPv6
5) Allow mixed mode (handle IPv4 and IPv6 by the same binary)
    (requires some interface changes to handle generic sockaddrs)

Everything seems to work fine, one can work with the same
config files as before, and everything works as expected,
but one can as well listen on IPv6 addresses (e.g.
the IPv6 unspecified address "::"). IPv4->6 mapping is
activated, or the kernel allows to deliver as well
IPv4 traffic to the IPv6 interface, then one can talk
also multiple protocols. The mileage may vary depending on the
OS and kernel.

It is as well possible to deactivate IPv6 support at compile
time via --disable-ipv6.

Here is a summary of the necessary changes:
- Added a amall Abstraction API for IP-version agnostic calls:
   Ns_LogSockaddr(), Ns_SockaddrGetPort(),
   Ns_SockaddrGetSockLen(), Ns_SockaddrMask(), Ns_SockaddrMaskBits(),
   Ns_SockaddrSetPort(), ns_inet_ntop(), ns_inet_pton()

- Replaced in interface more places where "sockaddr_in" (the IPv4
   version of the sockaddr) with the generic (version agnostic)
   sockaddr (also in IPv4 code to obtain API consistency when no IPv6
   is available): Ns_BindSock(), Ns_ConnSetPeer(), Ns_GetSockAddr(),
   Ns_SockBind(), Ns_SockBindUdp()

- The Ns_Sock structure contained sockaddr_in, which is not sufficient
   to store IPv6 addresses. Therefore it was necessary to change
   that to "sockaddr_storage" which guarantees to be large enough
   to keep all kind of addresses. This affects as well driver modules
   which communicate via the NaviServer managed sockets. Old drivers
   might crash with the new structure. Therefore, I've bumped
   NS_DRIVER_VERSION to 3 such that driver modules (also binaries)
   can communicate that they support IPv6. In IPv6 mode, NaviServer
   refuses to work with versions <3.

- Replaced all usages of ns_inet_ntoa() by ns_intet_ntop()
   (except for legacy an minimal library support)

- Added log file warnings on various places, where NaviServer
   was ignoring error states in socket communications silently.

- Transformed hash table for open listen ports to string keys
   (the old code used 32bit IPv4-addresses as keys in hash tables)
- added test for ns_listencallback

- Ns_HttpParseHost: new abstraction for parsing host and port number.
   Reason: IPv6 literal notation contains many colons, parsing
   IP-literal notation according to RFC 3986 section 3.2.2 is necessary
   (removed places where code was searching for ':' to expect the port)

- Added [ns_info ipv6] to obtain information whether the binary
   supports IPv6 or not

- Added configure flag --disable-ipv6 to build NaviServer just with
   IPv4 support.

- provide defines for core differences between IPv4 and IPv6
#ifdef HAVE_IPV6
# define NS_IP_LOOPBACK     "::1"
# define NS_IP_UNSPECIFIED  "::"
# define NS_SOCKADDR_IN     sockaddr_storage
# define NS_IPADDR_SIZE     INET6_ADDRSTRLEN
#else
# define NS_IP_LOOPBACK     "127.0.0.1"
# define NS_IP_UNSPECIFIED  "0.0.0.0"
# define NS_SOCKADDR_IN     sockaddr_in
# define NS_IPADDR_SIZE     INET_ADDRSTRLEN
#endif

  - nsperm:
   * Aligned implementation with the documentation
     (hosts.allow/deny have comma-separated entries)
   * Added submask specification in the form /xxx
     (e.g. ::1/64)
   * Use binary IPv4 and IPv6 values for hash and mask lookup
   * Generalizing masking functions
   * Refactor code to avoid memory leaks
   * Updated documentation

- modules:
   * nsudp and nsssl are now IPv6 aware/compliant

TODOs:
- more platform testing
   (probably changes for Windows necessary, help is appreciated)
- remove commented out debug statements

- make more modules IPv6 compliant
- documentation updates

I've tested the generic and IPv4 only version via regression
test and random browsing on Mac OS X and Linux. If you
have other platforms, testing would be appreciated. There
are some adjustments for windows probably necessary
(in the absence of "configure", one has to add manually
"HAVE_IPV4" to the compile flags). Any help is welcome.

best regards
-gustaf neumann

[naviserver-devel] NaviServer 4.99.10 available

[Gustaf N. <ne...@wu...>]

Dear friends,

NaviServer 4.99.10 is available. The new version is tagged with
naviserver-4.99.10 in mercurial and is as well available at source-forge
(naviserver, modules, documentation pages). This version is just
a bug-fix release. There is no need to make a big trara about this
new version, but it is worthwhile to get the issues fixed soon.
  
all the best

-gustaf neumann


=======================================
NaviServer 4.99.10, released 2016-02-16
=======================================

Changes relative to 4.99.9
  32 files changed, 404 insertions(+), 123 deletions(-)

Bug Fixes:
- fix potential race condition for configure value "connsperthread"
- fix conflicting prototypes in nsdbpg (on some compilers)

Documentation improvements:
  * add list of deprecated tcl-commands to documentation
  * fix spelling errors

Code Changes:
  * added log entries, when deprecated tcl-commands are called
  * remove obsolete .cvsignore files from repo
  * use SCNd64 instead of PRId64 for scanning numbers
  * further cleanup of the compatibility layer of the PRI*PTR macros
  * add parallel checks and inconclusive results to cppcheck options
  * use native bool definition when available
  * add target to makefile for testing under lldb
  * further code cleanup

Important changes in modules:
  * nsdbpg: fix conflicting prototype
  * nsssl: don't crash when fetching obtaining DH parameters from pem file fails

Re: [naviserver-devel] Error installing nsdpg in Mac OS X El Capitan

[Cesáreo G. R. <ce...@ce...>]

Dear Gustaf,

Nice! I've installed it (I had some issues connecting DB but I think is
a problem of my setup).

And, in case it would be useful, this is a complete warnings list in my
setup[1] Sorry if this is not the right medium to send this.

Thanks so much for your support
Cesáreo

----------
[1] List of warnings in
>> uname -v
Darwin Kernel Version 15.2.0: Fri Nov 13 19:56:56 PST 2015;
root:xnu-3248.20.55~2/RELEASE_X86_64


----- tclExecute.c, same error in several lines
/usr/local/src/naviserver/tcl8.5.18/unix/../generic/tclExecute.c
/usr/local/src/naviserver/tcl8.5.18/unix/../generic/tclExecute.c:725:22:
warning: array index -1 is before the beginning of the array
[-Warray-bounds]
    esPtr->tosPtr = &esPtr->stackWords[-1];
                     ^                 ~~
/usr/local/src/naviserver/tcl8.5.18/unix/../generic/tclInt.h:1299:5:
note: array 'stackWords' declared here
    Tcl_Obj *stackWords[1];
    ^
/usr/local/src/naviserver/tcl8.5.18/unix/../generic/tclExecute.c:937:31:
warning: array index -1 is before the beginning of the array
[-Warray-bounds]
        currElems = esPtr->endPtr - &esPtr->stackWords[-1];
                                     ^                 ~~
/usr/local/src/naviserver/tcl8.5.18/unix/../generic/tclInt.h:1299:5:
note: array 'stackWords' declared here
    Tcl_Obj *stackWords[1];
    ^
/usr/local/src/naviserver/tcl8.5.18/unix/../generic/tclExecute.c:938:45:
warning: array index -1 is before the beginning of the array
[-Warray-bounds]
        if (esPtr->markerPtr || (esPtr->tosPtr != &esPtr->stackWords[-1])) {
                                                   ^                 ~~
/usr/local/src/naviserver/tcl8.5.18/unix/../generic/tclInt.h:1299:5:
note: array 'stackWords' declared here
    Tcl_Obj *stackWords[1];
    ^
/usr/local/src/naviserver/tcl8.5.18/unix/../generic/tclExecute.c:950:31:
warning: array index -1 is before the beginning of the array
[-Warray-bounds]
        currElems = esPtr->endPtr - &esPtr->stackWords[-1];
                                     ^                 ~~
/usr/local/src/naviserver/tcl8.5.18/unix/../generic/tclInt.h:1299:5:
note: array 'stackWords' declared here
    Tcl_Obj *stackWords[1];
    ^
/usr/local/src/naviserver/tcl8.5.18/unix/../generic/tclExecute.c:1092:22: warning:
array index -1 is before the beginning of the array [-Warray-bounds]
    esPtr->tosPtr = &esPtr->stackWords[-1];
                     ^                 ~~
/usr/local/src/naviserver/tcl8.5.18/unix/../generic/tclInt.h:1299:5:
note: array 'stackWords' declared here
    Tcl_Obj *stackWords[1];
    ^
5 warnings generated.

-----

----------
/usr/local/src/naviserver/tcl8.5.18/unix/../unix/tclUnixInit.c:836:9:
warning: comparison of function 'CFLocaleCopyCurrent' not equal to a
null pointer is always true [-Wtautological-pointer-compare]
    if (CFLocaleCopyCurrent != NULL && CFLocaleGetIdentifier != NULL &&
        ^~~~~~~~~~~~~~~~~~~    ~~~~
/usr/local/src/naviserver/tcl8.5.18/unix/../unix/tclUnixInit.c:836:9:
note: prefix with the address-of operator to silence this warning
    if (CFLocaleCopyCurrent != NULL && CFLocaleGetIdentifier != NULL &&
        ^
        &
/usr/local/src/naviserver/tcl8.5.18/unix/../unix/tclUnixInit.c:836:9:
note: suffix with parentheses to turn this into a function call
    if (CFLocaleCopyCurrent != NULL && CFLocaleGetIdentifier != NULL &&
        ^
                           ()
/usr/local/src/naviserver/tcl8.5.18/unix/../unix/tclUnixInit.c:836:40:
warning: comparison of function 'CFLocaleGetIdentifier' not equal to a
null pointer is always true [-Wtautological-pointer-compare]
    if (CFLocaleCopyCurrent != NULL && CFLocaleGetIdentifier != NULL &&
                                       ^~~~~~~~~~~~~~~~~~~~~    ~~~~
/usr/local/src/naviserver/tcl8.5.18/unix/../unix/tclUnixInit.c:836:40:
note: prefix with the address-of operator to silence this warning
    if (CFLocaleCopyCurrent != NULL && CFLocaleGetIdentifier != NULL &&
                                       ^
                                       &
2 warnings generated.

--------

--------
/usr/local/src/naviserver/tcl8.5.18/unix/../unix/tclLoadDyld.c:386:17:
warning: 'NSLookupSymbolInImage' is deprecated: first deprecated in OS X
10.5 [-Wdeprecated-declarations]
            nsSymbol = NSLookupSymbolInImage(dyldLoadHandle->dyldLibHeader,
                       ^
/usr/include/mach-o/dyld.h:175:17: note: 'NSLookupSymbolInImage' has
been explicitly marked deprecated here
extern NSSymbol NSLookupSymbolInImage(const struct mach_header* image,
const char* symbolName, uint32_t options)
__OSX_AVAILABLE_BUT_DEPRECATED(__MAC_10_1,__MAC_10_5,__IPHONE_NA,__IPHONE_NA);
                ^
/usr/local/src/naviserver/tcl8.5.18/unix/../unix/tclLoadDyld.c:419:3:
warning: 'NSLinkEditError' is deprecated: first deprecated in OS X 10.5
[-Wdeprecated-declarations]
                NSLinkEditError(&editError, &errorNumber, &errorName,
&errMsg);
                ^
/usr/include/mach-o/dyld.h:210:13: note: 'NSLinkEditError' has been
explicitly marked deprecated here
extern void NSLinkEditError(NSLinkEditErrors *c, int *errorNumber, const
char** fileName, const char** errorString)
__OSX_AVAILABLE_BUT_DEPRECATED(__MAC_10_1,__MAC_10_5,__IPHONE_NA,__IPHONE_NA);
            ^
/usr/local/src/naviserver/tcl8.5.18/unix/../unix/tclLoadDyld.c:423:17:
warning: 'NSLookupSymbolInModule' is deprecated: first deprecated in OS
X 10.5 [-Wdeprecated-declarations]
            nsSymbol = NSLookupSymbolInModule(
                       ^
/usr/include/mach-o/dyld.h:174:17: note: 'NSLookupSymbolInModule' has
been explicitly marked deprecated here
extern NSSymbol NSLookupSymbolInModule(NSModule module, const char*
symbolName)
__OSX_AVAILABLE_BUT_DEPRECATED(__MAC_10_1,__MAC_10_5,__IPHONE_NA,__IPHONE_NA);
                ^
/usr/local/src/naviserver/tcl8.5.18/unix/../unix/tclLoadDyld.c:432:13:
warning: 'NSAddressOfSymbol' is deprecated: first deprecated in OS X
10.5 [-Wdeprecated-declarations]
            proc = NSAddressOfSymbol(nsSymbol);
                   ^
/usr/include/mach-o/dyld.h:181:21: note: 'NSAddressOfSymbol' has been
explicitly marked deprecated here
extern void *       NSAddressOfSymbol(NSSymbol symbol)
__OSX_AVAILABLE_BUT_DEPRECATED(__MAC_10_1,__MAC_10_5,__IPHONE_NA,__IPHONE_NA);
                    ^
/usr/local/src/naviserver/tcl8.5.18/unix/../unix/tclLoadDyld.c:497:15:
warning: 'NSUnLinkModule' is deprecated: first deprecated in OS X 10.5
[-Wdeprecated-declarations]
            result = NSUnLinkModule(modulePtr->module,
                     ^
/usr/include/mach-o/dyld.h:162:13: note: 'NSUnLinkModule' has been
explicitly marked deprecated here
extern bool NSUnLinkModule(NSModule module, uint32_t options)
__OSX_AVAILABLE_BUT_DEPRECATED(__MAC_10_1,__MAC_10_5,__IPHONE_NA,__IPHONE_NA);
            ^
/usr/local/src/naviserver/tcl8.5.18/unix/../unix/tclLoadDyld.c:701:12:
warning: 'NSCreateObjectFileImageFromMemory' is deprecated: first
deprecated in OS X 10.5 [-Wdeprecated-declarations]
            err = NSCreateObjectFileImageFromMemory(buffer, codeSize,
                  ^
/usr/include/mach-o/dyld.h:139:36: note:
'NSCreateObjectFileImageFromMemory' has been explicitly marked
deprecated here
extern NSObjectFileImageReturnCode
NSCreateObjectFileImageFromMemory(const void *address, size_t size,
NSObjectFileImage *objectFileImage)
__OSX_AVAILABLE_BUT_DEPRECATED(__MAC_10_1,__MAC_10_5,__IPHONE_NA,__IPHONE_NA);
                                   ^
/usr/local/src/naviserver/tcl8.5.18/unix/../unix/tclLoadDyld.c:734:14:
warning: 'NSLinkModule' is deprecated: first deprecated in OS X 10.5
[-Wdeprecated-declarations]
    module = NSLinkModule(dyldObjFileImage, "[Memory Based Bundle]",
             ^
/usr/include/mach-o/dyld.h:154:17: note: 'NSLinkModule' has been
explicitly marked deprecated here
extern NSModule NSLinkModule(NSObjectFileImage objectFileImage, const
char* moduleName, uint32_t options)
__OSX_AVAILABLE_BUT_DEPRECATED(__MAC_10_1,__MAC_10_5,__IPHONE_NA,__IPHONE_NA);
                ^
/usr/local/src/naviserver/tcl8.5.18/unix/../unix/tclLoadDyld.c:736:5:
warning: 'NSDestroyObjectFileImage' is deprecated: first deprecated in
OS X 10.5 [-Wdeprecated-declarations]
    NSDestroyObjectFileImage(dyldObjFileImage);
    ^
/usr/include/mach-o/dyld.h:140:36: note: 'NSDestroyObjectFileImage' has
been explicitly marked deprecated here
extern bool
NSDestroyObjectFileImage(NSObjectFileImage objectFileImage)

__OSX_AVAILABLE_BUT_DEPRECATED(__MAC_10_1,__MAC_10_5,__IPHONE_NA,__IPHONE_NA);
                                   ^
/usr/local/src/naviserver/tcl8.5.18/unix/../unix/tclLoadDyld.c:744:2:
warning: 'NSLinkEditError' is deprecated: first deprecated in OS X 10.5
[-Wdeprecated-declarations]
        NSLinkEditError(&editError, &errorNumber, &errorName, &errMsg);
        ^
/usr/include/mach-o/dyld.h:210:13: note: 'NSLinkEditError' has been
explicitly marked deprecated here
extern void NSLinkEditError(NSLinkEditErrors *c, int *errorNumber, const
char** fileName, const char** errorString)
__OSX_AVAILABLE_BUT_DEPRECATED(__MAC_10_1,__MAC_10_5,__IPHONE_NA,__IPHONE_NA);
            ^
9 warnings generated.
-----------

-----------
In file included from ./generic/nsf.c:988:
/usr/local/src/naviserver/nsf2.0.0/generic/nsfStack.c:293:10: warning:
nonnull parameter 'framePtr' will evaluate to 'true' on first encounter
[-Wpointer-bool-conversion]
  for (; framePtr; framePtr = Tcl_CallFrame_callerPtr(framePtr)) {
  ~~~    ^~~~~~~~
/usr/local/src/naviserver/nsf2.0.0/generic/nsfStack.c:330:10: warning:
nonnull parameter 'framePtr' will evaluate to 'true' on first encounter
[-Wpointer-bool-conversion]
  for (; framePtr; framePtr = Tcl_CallFrame_callerPtr(framePtr)) {
  ~~~    ^~~~~~~~
./generic/nsf.c:1135:7: warning: nonnull parameter 'clientData' will
evaluate to 'true' on first encounter [-Wpointer-bool-conversion]
  if (clientData && NsfObjectIsClass((NsfObject *)clientData)) {
      ^~~~~~~~~~ ~~
./generic/nsf.c:1637:11: warning: nonnull parameter 'objectPtr' will
evaluate to 'true' on first encounter [-Wpointer-bool-conversion]
      if (objectPtr) *objectPtr = object;
      ~~  ^~~~~~~~~
./generic/nsf.c:1662:9: warning: nonnull parameter 'objectPtr' will
evaluate to 'true' on first encounter [-Wpointer-bool-conversion]
    if (objectPtr) {*objectPtr = object;}
    ~~  ^~~~~~~~~
./generic/nsf.c:2981:10: warning: nonnull parameter 'clPtr' will
evaluate to 'true' on first encounter [-Wpointer-bool-conversion]
  for (; clPtr; clPtr = clPtr->nextPtr) {
  ~~~    ^~~~~
./generic/nsf.c:3604:10: warning: nonnull parameter 'pl' will evaluate
to 'true' on first encounter [-Wpointer-bool-conversion]
  for (; pl;  pl = pl->nextPtr) {
  ~~~    ^~
./generic/nsf.c:3625:10: warning: nonnull parameter 'pl' will evaluate
to 'true' on first encounter [-Wpointer-bool-conversion]
  for (; pl;  pl = pl->nextPtr) {
  ~~~    ^~
./generic/nsf.c:6901:10: warning: nonnull parameter 'l' will evaluate to
'true' on first encounter [-Wpointer-bool-conversion]
  for (; l; l = l->nextPtr) {
  ~~~    ^
./generic/nsf.c:6971:10: warning: nonnull parameter 'list' will evaluate
to 'true' on first encounter [-Wpointer-bool-conversion]
  while (list) {
  ~~~~~  ^~~~
./generic/nsf.c:7885:11: warning: nonnull parameter 'cmdl' will evaluate
to 'true' on first encounter [-Wpointer-bool-conversion]
  for ( ; cmdl; cmdl = cmdl->nextPtr) {
  ~~~     ^~~~
./generic/nsf.c:8500:11: warning: nonnull parameter 'cmdlist' will
evaluate to 'true' on first encounter [-Wpointer-bool-conversion]
  for ( ; cmdlist; cmdlist = cmdlist->nextPtr) {
  ~~~     ^~~~~~~
./generic/nsf.c:19977:42: warning: absolute value function 'abs' given
an argument of type 'long' but has parameter of type 'int' which may
cause truncation of value [-Wabsolute-value]
    if (ForwardArgString == remainder || abs(pos) > totalargs) {
                                         ^
./generic/nsf.c:19977:42: note: use function 'labs' instead
    if (ForwardArgString == remainder || abs(pos) > totalargs) {
                                         ^~~
                                         labs
./generic/nsf.c:28186:7: warning: nonnull parameter 'object' will
evaluate to 'true' on first encounter [-Wpointer-bool-conversion]
  if (object && (object->filterStack || object->mixinStack)) {
      ^~~~~~ ~~
./generic/nsf.c:29671:10: warning: nonnull parameter 'subClasses' will
evaluate to 'true' on first encounter [-Wpointer-bool-conversion]
  for (; subClasses; subClasses = subClasses->nextPtr) {
  ~~~    ^~~~~~~~~~
15 warnings generated.
----------

----------
./generic/nsfUtil.c:120:7: warning: nonnull parameter 'len' will
evaluate to 'true' on first encounter [-Wpointer-bool-conversion]
  if (len) *len = nr_written;
  ~~  ^~~
1 warning generated.
----------

----------
../expat/xmlparse.c:1568:18: warning: implicit conversion from
enumeration type 'enum XML_Status' to different enumeration type 'enum
XML_Error' [-Wenum-conversion]
        result = XML_STATUS_SUSPENDED;
               ~ ^~~~~~~~~~~~~~~~~~~~
../expat/expat.h:51:30: note: expanded from macro 'XML_STATUS_SUSPENDED'
#define XML_STATUS_SUSPENDED XML_STATUS_SUSPENDED
                             ^~~~~~~~~~~~~~~~~~~~
../expat/xmlparse.c:1578:18: warning: implicit conversion from
enumeration type 'enum XML_Status' to different enumeration type 'enum
XML_Error' [-Wenum-conversion]
        result = XML_STATUS_OK;
               ~ ^~~~~~~~~~~~~
../expat/expat.h:49:23: note: expanded from macro 'XML_STATUS_OK'
#define XML_STATUS_OK XML_STATUS_OK
                      ^~~~~~~~~~~~~
../expat/xmlparse.c:1608:12: warning: implicit conversion from
enumeration type 'enum XML_Error' to different enumeration type 'enum
XML_Status' [-Wenum-conversion]
    return result;
    ~~~~~~ ^~~~~~
3 warnings generated.
------------


------------
../generic/domxslt.c:1900:29: warning: equality comparison with
extraneous parentheses [-Wparentheses-equality]
        if ((node->nodeType == ATTRIBUTE_NODE)) {
             ~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~
../generic/domxslt.c:1900:29: note: remove extraneous parentheses around
the comparison to silence this warning
        if ((node->nodeType == ATTRIBUTE_NODE)) {
            ~               ^                ~
../generic/domxslt.c:1900:29: note: use '=' to turn this equality
comparison into an assignment
        if ((node->nodeType == ATTRIBUTE_NODE)) {
                            ^~
                            =
../generic/domxslt.c:7033:14: warning: address of 'xs->topLevelVars'
will always evaluate to 'true' [-Wpointer-bool-conversion]
    if (&xs->topLevelVars) {
    ~~   ~~~~^~~~~~~~~~~~
2 warnings generated.
-----------

-----------
../generic/tcldom.c:594:21: warning: assigning to 'Tcl_ObjType *' (aka
'struct Tcl_ObjType *') from 'const Tcl_ObjType *' (aka 'const struct
Tcl_ObjType *') discards qualifiers
[-Wincompatible-pointer-types-discards-qualifiers]
    objPtr->typePtr = &tdomNodeType;
                    ^ ~~~~~~~~~~~~~
../generic/tcldom.c:661:28: warning: assigning to 'Tcl_ObjType *' (aka
'struct Tcl_ObjType *') from 'const Tcl_ObjType *' (aka 'const struct
Tcl_ObjType *') discards qualifiers
[-Wincompatible-pointer-types-discards-qualifiers]
        resultObj->typePtr = &tdomNodeType;
                           ^ ~~~~~~~~~~~~~
../generic/tcldom.c:696:24: warning: assigning to 'Tcl_ObjType *' (aka
'struct Tcl_ObjType *') from 'const Tcl_ObjType *' (aka 'const struct
Tcl_ObjType *') discards qualifiers
[-Wincompatible-pointer-types-discards-qualifiers]
    resultObj->typePtr = &tdomNodeType;
                       ^ ~~~~~~~~~~~~~
3 warnings generated.
----------


----------
../generic/tclexpat.c:964:15: warning: implicit declaration of function
'close' is invalid in C99 [-Wimplicit-function-declaration]
              close (fd);
              ^
../generic/tclexpat.c:970:19: warning: implicit declaration of function
'read' is invalid in C99 [-Wimplicit-function-declaration]
          nread = read(fd, fbuf, READ_SIZE);
                  ^
2 warnings generated.
----------







El 17/enero/16 a las 13:53, Gustaf Neumann escribió:
> Dear Cesáreo,
> 
> You are right. There is a problem which shows up depending
> on the used compilers. The conflict of prototypes is fixed by [1].
> 
> Since i've fixed also one more probably old issue (setting an
> Ns_Set member from different threads leading to a potential
> race condidtion), we should release a bugfix version soon.
> 
> best regards
> -g
> 
> [1] 
> https://bitbucket.org/naviserver/nsdbpg/commits/8bd4d0a0cc8f3f4a81305dde225b684bfc16d5af
> 
> Am 17.01.16 um 02:24 schrieb Cesáreo García Rodicio:
>> Hi
>>
>> I'm trying to install naviserver (HEAD) in Mac OS X El Capitan[1] and,
>> when installing nsdbpg, I had this error:
>>   tclcmds.c:1041:1: error: conflicting types for 'blob_send_to_stream'
>> (See [2] full log)
>>
>> Any idea?
>>
>> Thanks
>> Cesáreo
>>
>> -----
>> [1] >> uname -v
>> Darwin Kernel Version 15.2.0: Fri Nov 13 19:56:56 PST 2015;
>> root:xnu-3248.20.55~2/RELEASE_X86_64
>>
>> [2]
>> ------------------------ Installing Modules/nsdbpg ----------------------
>> Makefile:61: warning: overriding commands for target `cppcheck'
>> /usr/local/naviserver/include/Makefile.module:199: warning: ignoring old
>> commands for target `cppcheck'
>> gcc -I/usr/local/pgsql/include -Os -DNDEBUG -Wall -fno-common -pipe
>> -I/usr/local/naviserver/include -I"/usr/local/naviserver/include"
>> -DHAVE_CONFIG_H    -c -o nsdbpg.o nsdbpg.c
>> gcc -I/usr/local/pgsql/include -Os -DNDEBUG -Wall -fno-common -pipe
>> -I/usr/local/naviserver/include -I"/usr/local/naviserver/include"
>> -DHAVE_CONFIG_H    -c -o tclcmds.o tclcmds.c
>> tclcmds.c:1041:1: error: conflicting types for 'blob_send_to_stream'
>> blob_send_to_stream(Tcl_Interp *interp, Ns_DbHandle *handle, const char
>> *lob_id,
>> ^
>> tclcmds.c:80:12: note: previous declaration is here
>> static int blob_send_to_stream(Tcl_Interp *interp, Ns_DbHandle *handle,
>> const char *lob_id,
>>             ^
>> 1 error generated.
>> make: *** [tclcmds.o] Error 1
>>
> 
> ------------------------------------------------------------------------------
> Site24x7 APM Insight: Get Deep Visibility into Application Performance
> APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
> Monitor end-to-end web transactions and take corrective actions now
> Troubleshoot faster and improve end-user experience. Signup Now!
> http://pubads.g.doubleclick.net/gampad/clk?id=267308311&iu=/4140
> _______________________________________________
> naviserver-devel mailing list
> nav...@li...
> https://lists.sourceforge.net/lists/listinfo/naviserver-devel
> 

Re: [naviserver-devel] Error installing nsdpg in Mac OS X El Capitan

[Gustaf N. <ne...@wu...>]

Dear Cesáreo,

You are right. There is a problem which shows up depending
on the used compilers. The conflict of prototypes is fixed by [1].

Since i've fixed also one more probably old issue (setting an
Ns_Set member from different threads leading to a potential
race condidtion), we should release a bugfix version soon.

best regards
-g

[1] 
https://bitbucket.org/naviserver/nsdbpg/commits/8bd4d0a0cc8f3f4a81305dde225b684bfc16d5af

Am 17.01.16 um 02:24 schrieb Cesáreo García Rodicio:
> Hi
>
> I'm trying to install naviserver (HEAD) in Mac OS X El Capitan[1] and,
> when installing nsdbpg, I had this error:
>   tclcmds.c:1041:1: error: conflicting types for 'blob_send_to_stream'
> (See [2] full log)
>
> Any idea?
>
> Thanks
> Cesáreo
>
> -----
> [1] >> uname -v
> Darwin Kernel Version 15.2.0: Fri Nov 13 19:56:56 PST 2015;
> root:xnu-3248.20.55~2/RELEASE_X86_64
>
> [2]
> ------------------------ Installing Modules/nsdbpg ----------------------
> Makefile:61: warning: overriding commands for target `cppcheck'
> /usr/local/naviserver/include/Makefile.module:199: warning: ignoring old
> commands for target `cppcheck'
> gcc -I/usr/local/pgsql/include -Os -DNDEBUG -Wall -fno-common -pipe
> -I/usr/local/naviserver/include -I"/usr/local/naviserver/include"
> -DHAVE_CONFIG_H    -c -o nsdbpg.o nsdbpg.c
> gcc -I/usr/local/pgsql/include -Os -DNDEBUG -Wall -fno-common -pipe
> -I/usr/local/naviserver/include -I"/usr/local/naviserver/include"
> -DHAVE_CONFIG_H    -c -o tclcmds.o tclcmds.c
> tclcmds.c:1041:1: error: conflicting types for 'blob_send_to_stream'
> blob_send_to_stream(Tcl_Interp *interp, Ns_DbHandle *handle, const char
> *lob_id,
> ^
> tclcmds.c:80:12: note: previous declaration is here
> static int blob_send_to_stream(Tcl_Interp *interp, Ns_DbHandle *handle,
> const char *lob_id,
>             ^
> 1 error generated.
> make: *** [tclcmds.o] Error 1
>

[naviserver-devel] Error installing nsdpg in Mac OS X El Capitan

[Cesáreo G. R. <ce...@ce...>]

Hi

I'm trying to install naviserver (HEAD) in Mac OS X El Capitan[1] and,
when installing nsdbpg, I had this error:
 tclcmds.c:1041:1: error: conflicting types for 'blob_send_to_stream'
(See [2] full log)

Any idea?

Thanks
Cesáreo

-----
[1] >> uname -v
Darwin Kernel Version 15.2.0: Fri Nov 13 19:56:56 PST 2015;
root:xnu-3248.20.55~2/RELEASE_X86_64

[2]
------------------------ Installing Modules/nsdbpg ----------------------
Makefile:61: warning: overriding commands for target `cppcheck'
/usr/local/naviserver/include/Makefile.module:199: warning: ignoring old
commands for target `cppcheck'
gcc -I/usr/local/pgsql/include -Os -DNDEBUG -Wall -fno-common -pipe
-I/usr/local/naviserver/include -I"/usr/local/naviserver/include"
-DHAVE_CONFIG_H    -c -o nsdbpg.o nsdbpg.c
gcc -I/usr/local/pgsql/include -Os -DNDEBUG -Wall -fno-common -pipe
-I/usr/local/naviserver/include -I"/usr/local/naviserver/include"
-DHAVE_CONFIG_H    -c -o tclcmds.o tclcmds.c
tclcmds.c:1041:1: error: conflicting types for 'blob_send_to_stream'
blob_send_to_stream(Tcl_Interp *interp, Ns_DbHandle *handle, const char
*lob_id,
^
tclcmds.c:80:12: note: previous declaration is here
static int blob_send_to_stream(Tcl_Interp *interp, Ns_DbHandle *handle,
const char *lob_id,
           ^
1 error generated.
make: *** [tclcmds.o] Error 1

Re: [naviserver-devel] NaviServer 4.99.9 available

[Gustaf N. <ne...@wu...>]

Am 12.01.16 um 11:21 schrieb David Osborne:
> I noticed there is a recent intent-to-package ticket with Debian.
> Are there moves afoot to have Naviserver officially included in Debian?
Héctor Romojaro is working on that (he is also maintaining the OpenACS 
debian package).
There is a official request to create an official Debian package, but as 
it looks as
the turn-around times of mail with the master maintainer in the Tcl context
are long. We keep you updated.

I've as well submitted a port to macports

    https://trac.macports.org/ticket/48917
    https://trac.macports.org/ticket/48918
    https://trac.macports.org/ticket/48919

which is still in the ticket-queue. if there is interest, it might help
to add a "cc me!" or a short hint "that port would be helpful" to the
ticket, to raise the urgency...

all the best
-g

Re: [naviserver-devel] NaviServer 4.99.9 available

[David O. <da...@qc...>]

Thanks Gustaf - that looks great.
On-the-fly minification in particular will be very helpful.

I noticed there is a recent intent-to-package ticket with Debian.
Are there moves afoot to have Naviserver officially included in Debian?

[naviserver-devel] NaviServer 4.99.9 available

[Gustaf N. <ne...@wu...>]

Dear friends,

NaviServer 4.99.9 is available. The new version is tagged with
naviserver-4.99.9 in mercurial and is as well available at source-forge
(naviserver, modules, documentation pages). I'll prepare as well
an announcement for c.l.tcl and OpenACS.

Below is the section for 4.99.9 from the NEWS file.

all the best

-gustaf neumann

======================================
NaviServer 4.99.6, released 2016-01-11
======================================

Changes relative to 4.99.8
  146 files changed, 5595 insertions(+), 4389 deletions(-)

New Features:

   * New command "ns_urlspace": this command allows to attach and
     retrieve arbitrary data to URLs in a hierarchical manner form the
     Tcl level. The command allows e.g.  to write ns-perm like access
     control in Tcl without much hassle.

        ns_urlspace set ?-id i? ?-key k? ?-noinherit? /url/ string
        ns_urlspace get ?-exact? ?-id i? ?-key k? ?-noinherit? /url/
        ns_urlspace list
        ns_urlspace new
        ns_urlspace unset ?-id i? ?-key k? ?-noinherit? ?-recurse? /url/

   * Added ability to optionally minify css and js files on the fly
     when gzipped content is requested. This is an extension of the
     gzip_static/gzip_refresh feature (for details see "fastpath" section
     in "ns_return" man page.

   * New statistics from nsdb interface: "ns_db stats" returns
     number and time of handle and sql operations (can be used
     in connection with e.g. munin for monitoring db health)

   * Added ability to pass env to ns_proxy (for details see
     man page of ns_proxy)

   * Added ability to keep "Host:" header in http client requests
     (for details, see man page of ns_http)

   * Improved logging support:

     - Added log colorization based on ANSI color codes. This feature
       makes it easier to spot error and warning in the error log
       file. Log colorization is turned on via configuration parameter
       "logcolorize" in the "ns/parameters" section in the NaviServer
       config file. Fine-tuning is possible via "logprefixcolor",
       "logprefixintensity" and per severity via
       
           ns_logctl severity severity-level ?-color color? \
              ?-intensity intensity? ?bool?

       When viewing a colorized error.log with less, use "less -r".

     - Added ability to log selectively SQL statements depending on
       pool and duration of the SQL command (configuration parameter
       "logMinDuration" for per-pool configuration sections,
       new command "ns_db logminduration ...").
       
     - Provide context in error log entries, when Tcl eval raises
       an error (this allows better distinction from where the
       exception was raised)

     - Added TCP_FASTOPEN support for Mac OS X 10.11

Bug Fixes:
   * the urlspace code now can be compiled with and without
     __URLSPACE_OPTIMIZE__. The performance difference is
     probably for most application little.
   * fix "floating-point" format for Ns_ObjvTime():
     1.05 was interpreted as 1.5 before
   * fix binary handling and content-type in http client
     requests (e.g. for -body in ns_http command)
   * fix bad interaction between chunked encoding and gzipped
     content (when closing gzip stream)
   * Added config option for disabling directorylisting
     (fastpath config option "directorylisting none")
   * fix quoted entries in access log, when quotes are contained
     in the content
   * Don't use TCP_CORK on UDP sockets
   * Ensure list nature of errorCode is preserved in proxy calls
   * Remove potential xss attack (mostly false positives) on error
     messages that are directly generated from user strings by NaviServer
   * Improved "signal lost" handling (timeout happens, but connPtr is
     provided). On some sites the previous approach was leading to a
     state stuck in a while in the error state.
   * Fixed off-by-one error in ns_server.  NsTclServerObjCmd was
     incorrectly checking argc after parsing options. As a result,
     "ns_server -pool {}" (pool given, no subcommand) would segfault.


Documentation improvements:
   * fix obsolete documentation for ns_register_proc
   * new man page for "ns_urlspace"
   * improve documentation for main-features, ns_cache, ns_connchan,
     ns_db, ns_http, ns_job, ns_log, ns_normalizepath, ns_proxy,
     ns_register, ns_return, ns_sleep, ns_time, nsv,
   * include logo of generated html docs via css to ease Debian
     packaging (silence lintian)


Tcl API Changes:
   * improve compatibility of ns_eval with Tcl's eval.
   * new API functions
     - ns_urlspace set|get|list|new|unset ...
     - ns_logctl severity severity-level ?-color color? ?-intensity intensity? ?bool?
     - ns_db logminduration ...
     - ns_db stats
     - ns_proxy config ... -env ns-set ...
     - ns_http queue .... -keep_host_header
     
   
C API Changes:
   * Added Ns_GetTimeFromString: provide API callable from config file to
     convert strings to ns_time (supporting "sec:usec" and "sec.fraction"
     formats)
   * Mark broken and unused function Ns_UrlSpecificGetFast() as deprecated
   * Added Ns_(No)CloseOnExec implementation for Win32
   * Added Ns_SetIUpdate()
   * Ns_Cancel(), Ns_Pause(), Ns_Resume() return "bool" instead of "int"


Incompatible API Changes:
   * Potential incompatibility: "ns_eval" behaves now closer to
     Tcl "eval" (e.g. allows to define procs with comments)


Configuration Changes:
   * Improved sample configuration for OpenACS, sample-config
     and nsssl
   * Added cppcheck target to Makefile.module


Code Changes:
   * Updated list of known HTTP status codes
   * Code cleanup:
     - use real boolean type when C99 is available
       (necessary for modules implemented in C++)
     - add const declarations
     - reduce implicit signed/unsigned conversions
     - reduce variable scopes
     - make code more uniform
     - better handling of nonnull asserts
       (don't raise warnings for these cases
        for most-recent gcc6 versions, when -pedantic is used)
     - prefer single returns from functions when possible
     - use NaviServer argv passing Ns_ParseObjv() on more
       occasions
     - replaced obsolete function 'asctime' by strftime
     - added fallback for definition of L_tmpnam, since the fallback
       from Tcl was removed on 2015-07-15
     - various changes for further silencing of static checkers
       (cppcheck and clang static checker)
  * Extended regression tests


Important changes in modules:
  * nsdbpg: fixes for "ns_pg blob*"
  * nsstats: added db-pool statistics, improved formatting
  * nsssl:
    - improved binary handling for "-body" (client procs),
    - ability to keep "Host:" header field
  * nsdbpg, nssmtpd, nsdns, nssnmp, nsudp, nsphp,nsdhcpd, nsssl, nsdbi:
    - code cleanup
  * changes in modules:
     13 files changed, 116 insertions(+), 856 deletions(-)

[naviserver-devel] Towards NaviServer 4.99.9

[Gustaf N. <ne...@wu...>]

Dear friends of NaviServer,

below is a list of changes between NaviServer 4.99.8 and 4.99.9.
In case nobody objects, or there are some other urgent changes
from you, I'll finish a release of the current tip version of NaviServer
the next days.

all the best
-gustaf neumann



======================================
NaviServer 4.99.6, released 2016-xx-xx
======================================

Changes relative to 4.99.8
  145 files changed, 5504 insertions(+), 4323 deletions(-)

New Features:

   * New command "ns_urlspace": this command allows to attach and
     retrieve arbitrary data to URLs in a hierarchical manner form the
     Tcl level. The command allows e.g.  to write ns-perm like access
     control in Tcl without much hassle.

        ns_urlspace set ?-id i? ?-key k? ?-noinherit? /url/ string
        ns_urlspace get ?-exact? ?-id i? ?-key k? ?-noinherit? /url/
        ns_urlspace list
        ns_urlspace new
        ns_urlspace unset ?-id i? ?-key k? ?-noinherit? ?-recurse? /url/

   * Added ability to optionally minify css and js files on the fly
     when gzipped content is requested. This is an extension of the
     gzip_static/gzip_refresh feature (for details see "fastpath" section
     in "ns_return" man page.

   * New statistics from nsdb interface: "ns_db stats" returns
     number and time of handle and sql operations (can be used
     in connection with e.g. munin for monitoring db health)

   * Added ability to pass env to ns_proxy (for details see
     man page of ns_proxy)

   * Added ability to keep "Host:" header in http client requests
     (for details, see man page of ns_http)

   * Improved logging support:

     - Added log colorization based on ANSI color codes. This feature
       makes it easier to spot error and warning in the error log
       file. Log colorization is turned on via configuration parameter
       "logcolorize" in the "ns/parameters" section in the NaviServer
       config file. Fine-tuning is possible via "logprefixcolor",
       "logprefixintensity" and per severity via
       
           ns_logctl severity severity-level ?-color color? \
              ?-intensity intensity? ?bool?

       When viewing a colorized error.log with less, use "less -r".

     - Added ability to log selectively SQL statements depending on
       pool and duration of the SQL command (configuration parameter
       "logMinDuration" for per-pool configuration sections,
       new command "ns_db logminduration ...").
       
     - Provide context in error log entries, when Tcl eval raises
       an error (this allows better distinction from where the
       exception was raised)

Bug Fixes:
   * the urlspace code now can be compiled with and without
     __URLSPACE_OPTIMIZE__. The performance difference is
     probably for most application little.
   * fix "floating-point" format for Ns_ObjvTime():
     1.05 was interpreted as 1.5 before
   * fix binary handling and content-type in http client
     requests (e.g. for -body in ns_http command)
   * fix bad interaction between chunked encoding and gzipped
     content (when closing gzip stream)
   * Added config option for disabling directorylisting
     (fastpath config option "directorylisting none")
   * fix quoted entries in access log, when quotes are contained
     in the content
   * Don't use TCP_CORK on UDP sockets
   * Ensure list nature of errorCode is preserved in proxy calls
   * Remove potential xss attack (mostly false positives) on error
     messages that are directly generated from user strings by NaviServer
   * Improved "signal lost" handling (timeout happens, but connPtr is
     provided). On some sites the previous approach was leading to a
     state stuck in a while in the error state.
   * Fixed off-by-one error in ns_server.  NsTclServerObjCmd was
     incorrectly checking argc after parsing options. As a result,
     "ns_server -pool {}" (pool given, no subcommand) would segfault.


Documentation improvements:
   * fix obsolete documentation for ns_register_proc
   * new man page for "ns_urlspace"
   * improve documentation for main-features, ns_cache, ns_connchan,
     ns_db, ns_http, ns_job, ns_log, ns_normalizepath, ns_proxy,
     ns_register, ns_return, ns_sleep, ns_time, nsv,
   * include logo of generated html docs via css to ease Debian
     packaging (silence lintian)

Tcl API Changes:
   * improve compatibility of ns_eval with Tcl's eval.
   * new API functions
     - ns_urlspace set|get|list|new|unset ...
     - ns_logctl severity severity-level ?-color color? ?-intensity intensity? ?bool?
     - ns_db logminduration ...
     - ns_db stats
     - ns_proxy config ... -env ns-set ...
     - ns_http queue .... -keep_host_header
     
   
C API Changes:
   * Added Ns_GetTimeFromString: provide API callable from config file to
     convert strings to ns_time (supporting "sec:usec" and "sec.fraction"
     formats)
   * Mark broken and unused function Ns_UrlSpecificGetFast() as deprecated
   * Added Ns_(No)CloseOnExec implementation for Win32
   * Added Ns_SetIUpdate()


Incompatible API Changes:
   * Potential incompatibility: "ns_eval" behaves now closer to
     Tcl "eval" (e.g. allows to define procs with comments)


Configuration Changes:
   * Improved sample configuration for OpenACS, sample-config
     and nsssl


Code Changes:
   * Updated list of known HTTP status codes
   * Code cleanup:
     - use real boolean type when C99 is available
       (necessary for modules implemented in C++)
     - add const declarations
     - reduce implicit signed/unsigned conversions
     - make code more uniform
     - better handling of nonnull asserts
       (don't raise warnings for these cases
        for most-recent gcc6 versions, when -pedantic is used)
     - prefer single returns from functions when possible
     - use NaviServer argv passing Ns_ParseObjv() on more
       occasions
     - replaced obsolete function 'asctime' by strftime
     - added fallback for definition of L_tmpnam, since the fallback
       from Tcl was removed on 2015-07-15
     - various changes for further silencing of static checkers
       (cppcheck and clang static checker)
  * Extended regression tests

Re: [naviserver-devel] commit/naviserver: 3 new changesets

[Gustaf N. <ne...@wu...>]

yes, this has pretty much potential for various kinds of applications, 
also for templating, redirects, caching (meta data), etc. However, i 
would not recommend it for logging, since the associated values are 
server-wide values that require mutex locking (somewhat like 
hierarchical nsvs). A higher level fronted for logging should be doable 
on the tcl layer and would run essentially lock-free.

One other value of the ns_urlspace command is that it makes the urlspace 
behavior easier regression-test-able. The behavior of the urlspace is 
non-trivial, especially the interactions with the various flags. I've 
already cleaned up some broken stuff (*FAST* flag + call). There is also 
some potential to improve the code performance-wise, but first the 
detailed behavior has to be "documented" in the regression test to suite 
to avoid potentially unexpected side effects.

I've committed a first version of ns_urlspace.test, which shows already 
3 somewhat unexpected cases (marked with ###). These cases should be 
fixed (or documented), but i'll add some more test cases soon. so treat 
the code as experimental...

-g


Am 19.11.15 um 08:21 schrieb Jeff Rogers:
> Bitbucket wrote:
>> 3 new commits in naviserver:
>> https://bitbucket.org/naviserver/naviserver/commits/2758b9f8392f/
>> Changeset:   2758b9f8392f
>> User:        gustafn
>> Date:        2015-11-18 19:35:25+00:00
>> Summary:     - The new command "ns_urlspace" that allows to attach data to URLs in
>>     a hierarchical manner form the Tcl level. The command allows e.g. to
>>     write ns-perm like access control in Tcl without much hassle.
> This is very cool; it's something I've wanted to expose as a tcl api for
> a long time.
>
> A thought for a further extension to this is to make the separator
> character changeable instead of only "/", so strings represented as
> "a.b.c.d" or "a:b:c:d" could be handled too.  However, that would
> probably add a good deal of complexity that is unjustified lacking an
> actual use case.
>
> I was also thinking that the trie structure could be applicable to
> logging configurations: enabling logging for "debug/sql" and "info/*"
> but not "debug/*" would be really nice.  The "/" as a separator
> character seems odd at first, but ultimately unimportant as it conveys
> the same hierarchical information as "." or ":" would.
>
> -J
>

Re: [naviserver-devel] commit/naviserver: 3 new changesets

[Jeff R. <dv...@di...>]

Bitbucket wrote:
> 3 new commits in naviserver:

> https://bitbucket.org/naviserver/naviserver/commits/2758b9f8392f/
> Changeset:   2758b9f8392f
> User:        gustafn
> Date:        2015-11-18 19:35:25+00:00
> Summary:     - The new command "ns_urlspace" that allows to attach data to URLs in
>    a hierarchical manner form the Tcl level. The command allows e.g. to
>    write ns-perm like access control in Tcl without much hassle.

This is very cool; it's something I've wanted to expose as a tcl api for 
a long time.

A thought for a further extension to this is to make the separator 
character changeable instead of only "/", so strings represented as 
"a.b.c.d" or "a:b:c:d" could be handled too.  However, that would 
probably add a good deal of complexity that is unjustified lacking an 
actual use case.

I was also thinking that the trie structure could be applicable to 
logging configurations: enabling logging for "debug/sql" and "info/*" 
but not "debug/*" would be really nice.  The "/" as a separator 
character seems odd at first, but ultimately unimportant as it conveys 
the same hierarchical information as "." or ":" would.

-J

Re: [naviserver-devel] C-Api / Configuration Docs

[Gustaf N. <ne...@wu...>]

Dear Rodrigo,

The C-API of NaviServer manifests itself through the external function 
definitions of libnsd other the other core libraries. The best sources 
are the naviserver/include/ns.h [1] and the source files (e.g. [2]). All 
public functions are described based on a standardized template 
(function description, results, side effects). However, so far there is 
currently no single document, that summarizes the C-API functions.

 From my experience, C-developer get around quite well with the existing 
API descriptions. A good source for usage cases of the C-API are the 49 
extra modules from [3], which are mostly written in C.

Concerning the possible configuration parameters: NaviServer follows an 
open model, where the entries of the config files fill a kind of a 
database, where NaviServer itself or some NaviServer modules or 
applications (such as OpenACS) can read its config values from it. 
NaviServer offers in good Tcl tradition an introspection interface that 
allows you to see, what parameters with what values are currently in 
use. I would recommend to use the nsstats module [4], which offers a web 
interface to many of the configuration options and runtime statistics.

The valid parameter options and their semantics are described in the 
various modules, such as e.g. in [5,6] and as well in the provided 
sample config files (e.g. [7,8]).

hope this helps
-gn

[1] 
https://bitbucket.org/naviserver/naviserver/src/906257d0d90c8b71d323e61261c6515268397f8b/include/ns.h?at=default&fileviewer=file-view-default#ns.h-611
[2] 
https://bitbucket.org/naviserver/naviserver/src/906257d0d90c8b71d323e61261c6515268397f8b/nsd/set.c?at=default&fileviewer=file-view-default#set.c-442
[3] https://bitbucket.org/naviserver/
[4] https://bitbucket.org/naviserver/nsstats
[5] 
http://naviserver.sourceforge.net/n/manual/files/admin-db.html#subsection2
[6] http://naviserver.sourceforge.net/n/nssock/files/nssock.html
[7] 
https://bitbucket.org/naviserver/naviserver/src/906257d0d90c8b71d323e61261c6515268397f8b/sample-config.tcl.in?at=default&fileviewer=file-view-default
[8] 
https://bitbucket.org/naviserver/naviserver/src/906257d0d90c8b71d323e61261c6515268397f8b/openacs-config.tcl?at=default&fileviewer=file-view-default

Am 12.10.15 um 18:53 schrieb hr...@gm...:
> Dear Sirs!
>
> I just installed naviserver and have some questions. I hope this is
> the right place.
>
> http://naviserver.sourceforge.net/n/manual/files/main-features.html
>
> Speaks about a C api, but I dont find the documentation about it.
> Where is it?
>
> And a much more elementary question about configuration: I dont know
> where to get information about what parameters are possible for each
> section. For example: what are the configurable parameters for the
> section "ns/parameters"? Where I find it in the documentation?
>
> Thanks
> Rodrigo.
>

Re: [naviserver-devel] ns_sendmail's use of ns_sockopen

[David O. <da...@qc...>]

Thanks Gustaf - that's perfect

On 16 October 2015 at 12:56, Gustaf Neumann <ne...@wu...> wrote:

> Dear David,
>
> from reading the source code, a test for error results was broken (see
> [1]).
> -g
>
> PS: We use currently tcllib since it supports as well STARTTLS.
>
> [1]
>
> https://bitbucket.org/naviserver/naviserver/commits/6657f9b1b41aaefc5b452ed0c869659a682813af
>
> Am 15.10.15 um 18:27 schrieb David Osborne:
> > Hi,
> >
> > Been trying to use Naviserver's ns_sendmail for the first time.
> >
> > By default, speaking to localhost:25 (under Debian) I'm getting a
> > "Operation now in progress" error at the point where ns_sockopen is
> > called against localhost:25.
> >
> > I modified the code to use the -nonblock switch (just as an experiment
> > rather than anything else) and it now works for me.
> >
> > Is this something that's just peculiar to my environment?
> >
>
>
>
> ------------------------------------------------------------------------------
> _______________________________________________
> naviserver-devel mailing list
> nav...@li...
> https://lists.sourceforge.net/lists/listinfo/naviserver-devel
>

Re: [naviserver-devel] ns_sendmail's use of ns_sockopen

[Gustaf N. <ne...@wu...>]

Dear David,

from reading the source code, a test for error results was broken (see [1]).
-g

PS: We use currently tcllib since it supports as well STARTTLS.

[1] 
https://bitbucket.org/naviserver/naviserver/commits/6657f9b1b41aaefc5b452ed0c869659a682813af

Am 15.10.15 um 18:27 schrieb David Osborne:
> Hi,
>
> Been trying to use Naviserver's ns_sendmail for the first time.
>
> By default, speaking to localhost:25 (under Debian) I'm getting a 
> "Operation now in progress" error at the point where ns_sockopen is 
> called against localhost:25.
>
> I modified the code to use the -nonblock switch (just as an experiment 
> rather than anything else) and it now works for me.
>
> Is this something that's just peculiar to my environment?
>

[naviserver-devel] ns_sendmail's use of ns_sockopen

[David O. <da...@qc...>]

Hi,

Been trying to use Naviserver's ns_sendmail for the first time.

By default, speaking to localhost:25 (under Debian) I'm getting a
"Operation now in progress" error at the point where ns_sockopen is called
against localhost:25.

I modified the code to use the -nonblock switch (just as an experiment
rather than anything else) and it now works for me.

Is this something that's just peculiar to my environment?

<--------->

diff -r 7dd17a2cf070 tcl/sendmail.tcl
--- a/tcl/sendmail.tcl Mon Aug 03 19:02:11 2015 +0200
+++ b/tcl/sendmail.tcl Thu Oct 15 17:21:02 2015 +0100
@@ -254,7 +254,7 @@
         set smtphost [ns_config -set ns/parameters mailhost "localhost"]
     }

-    set fds [ns_sockopen -timeout $timeout $smtphost $smtpport]
+    set fds [ns_sockopen -nonblock $smtphost $smtpport]
     set rfd [lindex $fds 0]
     set wfd [lindex $fds 1]

<------------>

-- 
David

[naviserver-devel] C-Api / Configuration Docs

[<hr...@gm...>]

Dear Sirs!

I just installed naviserver and have some questions. I hope this is
the right place.

http://naviserver.sourceforge.net/n/manual/files/main-features.html

Speaks about a C api, but I dont find the documentation about it.
Where is it?

And a much more elementary question about configuration: I dont know
where to get information about what parameters are possible for each
section. For example: what are the configurable parameters for the
section "ns/parameters"? Where I find it in the documentation?

Thanks
Rodrigo.

Re: [naviserver-devel] Compilation error NaviServer 4.99.8

[Gustaf N. <ne...@wu...>]

Am 25.09.15 um 15:58 schrieb Zoran Vasiljevic:
> On 25 Sep 2015, at 15:44, Gustaf Neumann <ne...@wu...> wrote:
>
>> kind of. the fix was addressing a problem in rivet under Linux,
>> the consequences were not easy to spot from the code but hit
>> all multi-threaded tcl applications, including NaviServer.
>> It also lead to bug-reports from bsd about "massive slowdowns".
>>
> I see... so you had to fix that mess...
> At this point I believe you've done the
> fix so that it will not harm us!
it was already causing harm. i started to look at the problem,
when one quite large NaviServer based site  was
suddenly running our of memory. This site was
using [exec ...] without nsproxy (for no good reason).
The [exec] call hang sometimes in a fork(), after duplicating the
memory. This site had a maximum of 90 connection threads
configured, using  a footprint of several GB. When forking
such a monster multiple times, and the fork hangs,
one runs soon out of memory... The obvious fix was
to use nsproxy for that site, but still, the Tcl behavior
was bad enough to investigate further..
> I guess they wanted to make [fork] working
> which simply cannot be done MT-safe(ly).
A few people tried to argue for that, but not long :).

The problem was that rivet requires after a fork() a
notifier thread running, such that [vwait ...] etc. works.
Therefore, it was necessary to use a hook that starts
the notifier thread after a fork() operation. That bad thing
is that the implementation in 2013 created a notifier thread
after every fork, (even for [exec], where it is
completely useless).

Furthermore, starting a notifier threads
requires Tcl_MutexLock(), and even Tcl_MutexLock() requires
a (master) lock to function (Tcl maintains a list of mutexes). If one
locks this master mutex before fork (which takes a while),
all parallelism is gone, since every mutex operation
is serialized ... and that happened for every exec!

Finally i implemented the notifier thread creation in a lazy
manner (create the notifier thread when needed, i.e. not during [exec]),
and used the pthread-API for the notifier thread management
  instead of the Tcl_Mutex* interface.
> I've been telling this for years but people
> won't listen. And there is some substantional
> portion in politics there, which I regret :-(
i don't think that politics was a main blocker. It was rather
missing problem awareness (most application are not
using threads), and the hope for a general [fork] call in
Tcl, which makes me shudder when i think about
heavy multi-threading applications.

-g

Re: [naviserver-devel] Compilation error NaviServer 4.99.8

[Clinciu A. <the...@ya...>]

Hi all
The recompilation of naviserver 4.99.8 and modules did not seem to solve the crashes
nsd: set.c:61: Ns_SetUpdate: Assertion `value != ((void *)0)' failed.
It now seems to crash more often.
 However after days of extensive searching and going through ALL naviserver settings i did find the truncation culprit which not only truncated images but everything from .txt,.zip, .pdf,.css and .jpg files whenever the server was accesed from non localhost. This is because testing on localhost did not seem to affect it and I have no idea why.

I've found  that just one setting seemed to screw everything up (including total loading time of page extending to keepwait time):
The fastpath settings of cachemaxentry: 
   ns_param    cachemaxentry        [expr 70*1024];#8192

I've tried to set it to multipliers of 2 like 256*1024 but no luck. So once i've  set it back to 8*1024 it seemed to solve the issue completly. 
The site loads faster than with fastpath on.
Even the bloated ckeditor loads faster, before this i had to refresh the page 3-5 times to just see all the buttons.

Is this something worth looking into? Because JS and CSS today are way bigger than 8 kb the fastpath would be valuable (that's why i wanted to use it) in my point of view.

Hope this helps anyone who ever has the same problem as I did for the past months.

I can't believe how happy I am right now, i just thought i must have been doing something wrong during compilation or setup that caused naviserver to render so slow.
  
With regards,
Clinciu Andrei George

"Vorba buna, zambetul si fapta binefacatoare sunt raze ale soarelui rasfrante in sufletul omului."
"A good word, a smile and a good deed are just like rays of the sun reflected in man's soul."   by Nicolae Iorga
 


     On Friday, September 25, 2015 4:58 PM, Zoran Vasiljevic <zv...@ar...> wrote:
   

 
On 25 Sep 2015, at 15:44, Gustaf Neumann <ne...@wu...> wrote:

> kind of. the fix was addressing a problem in rivet under Linux,
> the consequences were not easy to spot from the code but hit
> all multi-threaded tcl applications, including NaviServer.
> It also lead to bug-reports from bsd about "massive slowdowns".
> 

I see... so you had to fix that mess... 
At this point I believe you've done the
fix so that it will not harm us!

I guess they wanted to make [fork] working 
which simply cannot be done MT-safe(ly). 
I've been telling this for years but people
won't listen. And there is some substantional
portion in politics there, which I regret :-(




------------------------------------------------------------------------------
_______________________________________________
naviserver-devel mailing list
nav...@li...
https://lists.sourceforge.net/lists/listinfo/naviserver-devel


  

Re: [naviserver-devel] Compilation error NaviServer 4.99.8

[Zoran V. <zv...@ar...>]

On 25 Sep 2015, at 15:44, Gustaf Neumann <ne...@wu...> wrote:

> kind of. the fix was addressing a problem in rivet under Linux,
> the consequences were not easy to spot from the code but hit
> all multi-threaded tcl applications, including NaviServer.
> It also lead to bug-reports from bsd about "massive slowdowns".
> 

I see... so you had to fix that mess... 
At this point I believe you've done the
fix so that it will not harm us!

I guess they wanted to make [fork] working 
which simply cannot be done MT-safe(ly). 
I've been telling this for years but people
won't listen. And there is some substantional
portion in politics there, which I regret :-(

Re: [naviserver-devel] Compilation error NaviServer 4.99.8

[Gustaf N. <ne...@wu...>]

Am 25.09.15 um 14:46 schrieb Zoran Vasiljevic:
> On 25 Sep 2015, at 14:38, Gustaf Neumann <ne...@wu...> wrote:
>
>> which will be fixed in 8.5.19, as well as the bug mentioned below (which can
>> cause problems for all kinds of forks, when e.g. nsproxy is not used for "exec",
>> or for nscgi).
> Can it be that somebody in Tcl was "too clever" and
> short-sightedly wanted to solve the un-solvable?
kind of. the fix was addressing a problem in rivet under Linux,
the consequences were not easy to spot from the code but hit
all multi-threaded tcl applications, including NaviServer.
It also lead to bug-reports from bsd about "massive slowdowns".
> So Tcl 8.5.18 is really useless for NS?
for production, yes.

The forking/exec/cgi problem can be avoided
by avoiding exec and nscgi (probably nsphp). The bug was introduced
in Tcl 8.5.15 and Tcl 8.6.1. All later Tcl versions are affected.
Jan Nijtmans has back-ported the changes yesterday  to the
Tcl core-8-5tcl-8-5 branch targeted for the next Tcl 8.5 release.

So, if one requires a functionality based on fork, the best
approach is to stick to Tcl 8.5.14 for the time being.

-g

Re: [naviserver-devel] Compilation error NaviServer 4.99.8

[Zoran V. <zv...@ar...>]

On 25 Sep 2015, at 14:38, Gustaf Neumann <ne...@wu...> wrote:

> which will be fixed in 8.5.19, as well as the bug mentioned below (which can 
> cause problems for all kinds of forks, when e.g. nsproxy is not used for "exec",
> or for nscgi).

Can it be that somebody in Tcl was "too clever" and
short-sightedly wanted to solve the un-solvable?

So Tcl 8.5.18 is really useless for NS? 

Re: [naviserver-devel] Compilation error NaviServer 4.99.8

[Gustaf N. <ne...@wu...>]

I would not recommend using 8.5.18, using something in the range of 
8.5.15 to 8.5.17
should be ok. We were hit not long ago by the following bug in Tcl 8.5.18

  * http://core.tcl.tk/tcl/info/879a0747bee593e2
  * http://core.tcl.tk/tcl/info/4b964e7afb811898

which will be fixed in 8.5.19, as well as the bug mentioned below (which 
can
cause problems for all kinds of forks, when e.g. nsproxy is not used for 
"exec",
or for nscgi).

-g

http://core.tcl.tk/tcl/tktview/57945b574a6df0332efc4ac96b066f7c347b28f7




Am 25.09.15 um 14:04 schrieb Clinciu Andrei:
> Dear Gustaf,
>
> Thanks for the extra make flags, it now compiles successfully.
> I never imagined 8.6 would be a culprit or that there could be so many 
> differences. But it's worth checking it out.
> I'll give it a try locally to see if nothing breaks from downgrading 
> to 8.6. Although i'm sure i have to find some compatibility layers for 
> some commands like try,zlib, etc and see if tdbc works properly with 8.5.
>
> Is tcl 8.5.18 also good or is it better safer to stick to 8.5.15 you 
> mentioned?
> With regards,
> Clinciu Andrei George
>
> "Vorba buna, zambetul si fapta binefacatoare sunt raze ale soarelui 
> rasfrante in sufletul omului."
> "A good word, a smile and a good deed are just like rays of the sun 
> reflected in man's soul."   by Nicolae Iorga
>
>
>
> On Friday, September 25, 2015 12:42 PM, Gustaf Neumann 
> <ne...@wu...> wrote:
>
>
> Dear Andrei,
>
> Thanks for the compile-log. The problem is that the compiler flags 
> that are
> picked up in your environment include a "
> -fvisibility=hidden", most likely
> picked up from tcl 8.6.1. Using "-fvisibility=hidden" should work 
> certainly,
> an was fixed in the tip versions.
>
> The compilation works well with 4.99.8, if you either compile against 
> Tcl 8.5.* or
> when you use
>
>    make "CFLAGS_EXTRA=-fPIC -pipe"
>
> It might be, that the problems you are experiencing have to do with 
> tcl8.6.
> We use on all our production systems tcl 8.5.* (e.g. on openacs.org: 
> Tcl 8.5.15).
> If you have the chance to use tcl 8.5, please try.
>
> all the best
> -g
>
>
> Am 25.09.15 um 10:43 schrieb Clinciu Andrei:
>> Hi
>>
>> I've tried compiling the 4.99.8 and I remembered why I even went to 
>> get the bitbucket version, because of failure to compile.
>>
>> Doing:
>> *./configure --prefix=/opt/ns --enable-symbols --enable-threads*
>> *make -j8 *
>> *or simply 'make' *
>> Also trying with autoconf (maybe the configure missed something?)
>> This however was not an issue for the bitbucket version at that time
>>
>> Error:
>> libnsd.so: undefined reference to `NS_finalshutdown'
>> collect2: error: ld returned 1 exit status
>> make[1]: *** [nsd] Error 1
>> make[1]: Leaving directory 
>> `/home/lostone/naviserver/naviserver-4.99.8/nsd'
>> make: *** [all] Error 1
>>
>> It's worthy to note that I've also compiled the 4.99.7 version 
>> *without any problem*
>>
>> make.log : Paste2.org - Viewing Paste G2a9BEmz 
>> <http://paste2.org/G2a9BEmz>
>>
>> I sure hope i'm not bugging you guys:)
>>
>> With regards,
>> Clinciu Andrei George
>>
>>
>> "Vorba buna, zambetul si fapta binefacatoare sunt raze ale soarelui 
>> rasfrante in sufletul omului."
>> "A good word, a smile and a good deed are just like rays of the sun 
>> reflected in man's soul."   by Nicolae Iorga
>
>
> -- 
> Univ.Prof. Dr. Gustaf Neumann
> WU Vienna
> Institute of Information Systems and New Media
> Welthandelsplatz 1, A-1020 Vienna, Austria
>
>


-- 
Univ.Prof. Dr. Gustaf Neumann
WU Vienna
Institute of Information Systems and New Media
Welthandelsplatz 1, A-1020 Vienna, Austria

Re: [naviserver-devel] Compilation error NaviServer 4.99.8

[Clinciu A. <the...@ya...>]

Dear Gustaf,
Thanks for the extra make flags, it now compiles successfully.
I never imagined 8.6 would be a culprit or that there could be so many differences. But it's worth checking it out. 
I'll give it a try locally to see if nothing breaks from downgrading to 8.6. Although i'm sure i have to find some compatibility layers for some commands like try,zlib, etc and see if tdbc works properly with 8.5.
Is tcl 8.5.18 also good or is it better safer to stick to 8.5.15 you mentioned?
 With regards,
Clinciu Andrei George

"Vorba buna, zambetul si fapta binefacatoare sunt raze ale soarelui rasfrante in sufletul omului."
"A good word, a smile and a good deed are just like rays of the sun reflected in man's soul."   by Nicolae Iorga
 


     On Friday, September 25, 2015 12:42 PM, Gustaf Neumann <ne...@wu...> wrote:
   

  Dear Andrei, 
 
 Thanks for the compile-log. The problem is that the compiler flags that are 
 picked up in your environment include a "  -fvisibility=hidden", most likely 
 picked up from tcl 8.6.1. Using  "-fvisibility=hidden" should work certainly, 
 an was fixed in the tip versions.
 
 The compilation works well with 4.99.8, if you either compile against Tcl 8.5.* or 
 when you use
 
    make "CFLAGS_EXTRA=-fPIC -pipe" 
 
 It might be, that the problems you are experiencing have to do with tcl8.6. 
 We use on all our production systems tcl 8.5.* (e.g. on openacs.org: Tcl 8.5.15). 
 If you have the chance to use tcl 8.5, please try.
 
 all the best
 -g
 
 
 Am 25.09.15 um 10:43 schrieb Clinciu Andrei:
 
  Hi 
  I've tried compiling the 4.99.8 and I remembered why I even went to get the bitbucket version, because of failure to compile.  
  Doing: ./configure --prefix=/opt/ns --enable-symbols --enable-threads
  make -j8  or simply 'make'  Also trying with autoconf (maybe the configure missed something?) This however was not an issue for the bitbucket version at that time 
  Error: libnsd.so: undefined reference to `NS_finalshutdown' collect2: error: ld returned 1 exit status make[1]: *** [nsd] Error 1 make[1]: Leaving directory `/home/lostone/naviserver/naviserver-4.99.8/nsd' make: *** [all] Error 1 
  It's worthy to note that I've also compiled the 4.99.7 version without any problem 
  make.log : Paste2.org - Viewing Paste G2a9BEmz 
  I sure hope i'm not bugging you guys:)  
   With regards,
 Clinciu Andrei George
  
  
  "Vorba buna, zambetul si fapta binefacatoare sunt raze ale soarelui rasfrante in sufletul omului."
  "A good word, a smile and a good deed are just like rays of the sun reflected in man's soul."   by Nicolae Iorga
    
 
 
 -- 
Univ.Prof. Dr. Gustaf Neumann
WU Vienna
Institute of Information Systems and New Media
Welthandelsplatz 1, A-1020 Vienna, Austria