From max at e-soft.ru Mon Apr 5 05:57:16 2010 From: max at e-soft.ru (Maxim Dementiev) Date: Mon, 05 Apr 2010 13:57:16 +0400 Subject: [FASTCGI] Unit testing for FastCGI. Message-ID: Hi, I wonder if there is any unit testing for FastCGI library. As I can see http://fastcgi.com/cvs/fcgi2/ doesn't contain "ut" or "unittests". What have initial FastCGI developers used to ensure that the library works? Thanks! Regards, Max From jsprenkle at gmail.com Mon Apr 5 08:16:05 2010 From: jsprenkle at gmail.com (Jay Sprenkle) Date: Mon, 5 Apr 2010 07:16:05 -0500 Subject: [FASTCGI] Unit testing for FastCGI. In-Reply-To: References: Message-ID: Don't believe there is any. If you want to contribute that would be very nice. On Mon, Apr 5, 2010 at 4:57 AM, Maxim Dementiev wrote: > Hi, > > I wonder if there is any unit testing for FastCGI library. > As I can see http://fastcgi.com/cvs/fcgi2/ doesn't contain "ut" or > "unittests". > -------------- next part -------------- An HTML attachment was scrubbed... URL: From no-reply at dropbox.com Wed Apr 7 08:11:12 2010 From: no-reply at dropbox.com (Dropbox) Date: Wed, 07 Apr 2010 12:11:12 +0000 Subject: [FASTCGI] Durfos Terminator has invited you to Dropbox Message-ID: <20100407121112.46DCE4616DC@mailman.dropbox.com> We're excited to let you know that Durfos Terminator has invited you to Dropbox! Durfos Terminator has been using Dropbox to sync and share files online and across computers, and thought you might want it too. Visit http://www.dropbox.com/link/20.ZNRhTEcTX7/NjEzMzQ2ODkyNw to get started. - The Dropbox Team ____________________________________________________ To stop receiving invites from Dropbox, please go to http://www.dropbox.com/bl/769ebe0274d8/fastcgi-developers%40fastcgi.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From apriestley at messenger.org Thu Apr 8 18:54:35 2010 From: apriestley at messenger.org (apriestley) Date: Thu, 8 Apr 2010 15:54:35 -0700 Subject: [FASTCGI] freecgi++ Message-ID: <54437c5f$eda2b69$4bd0545c$@com> Hi, Anyone have success with the CGI library freecgi++ as fastcgi? for those of you familiar with the library, off the bat I'm getting errors regarding fstream. is this an instance where one needs to use fcgiapp.h? any help would be appreciated, Auron -------------- next part -------------- An HTML attachment was scrubbed... URL: From max at e-soft.ru Fri Apr 9 02:18:51 2010 From: max at e-soft.ru (Maxim P. Dementiev) Date: Fri, 09 Apr 2010 10:18:51 +0400 Subject: [FASTCGI] freecgi++ In-Reply-To: <54437c5f$eda2b69$4bd0545c$@com> References: <54437c5f$eda2b69$4bd0545c$@com> Message-ID: Hi, Auron, Your question is very concise, it doesn't contain any explanations (e.g. output from compiler), so it doesn't provide any clue. Anyway... On Thu, 8 Apr 2010 15:54:35 -0700 "apriestley" wrote: > Hi, > > Anyone have success with the CGI library freecgi++ as fastcgi? > Yes, we have! > for those of you familiar with the library, off the bat I'm getting errors > regarding fstream. > We are using fcgi_streambuf class from fcgio.h. The application, which uses it, compiles and links without any problem. > is this an instance where one needs to use fcgiapp.h? > > any help would be appreciated, > > Auron > > Best regards, Max ________________________________________________________________ LiveJournal: http://mpd.livejournal.com/ SourceForge: http://maximdementiev.users.sourceforge.net/ Wikipedia: http://en.wikipedia.org/wiki/User:Mpd_en From max at e-soft.ru Fri Apr 9 02:34:34 2010 From: max at e-soft.ru (Maxim P. Dementiev) Date: Fri, 09 Apr 2010 10:34:34 +0400 Subject: [FASTCGI] freecgi++ In-Reply-To: References: <54437c5f$eda2b69$4bd0545c$@com> Message-ID: Excuse me, we don't use _FREE_cgi++, I was wrong... Sorry. Max On Fri, 09 Apr 2010 10:18:51 +0400 "Maxim P. Dementiev" wrote: > Hi, Auron, > > Your question is very concise, it doesn't contain any explanations (e.g. output from compiler), so it doesn't provide any clue. > Anyway... > > On Thu, 8 Apr 2010 15:54:35 -0700 > "apriestley" wrote: >> Hi, >> >> Anyone have success with the CGI library freecgi++ as fastcgi? >> > Yes, we have! > From paulojorgedias at gmail.com Tue Apr 13 04:25:44 2010 From: paulojorgedias at gmail.com (Paulo Jorge Dias) Date: Tue, 13 Apr 2010 09:25:44 +0100 Subject: [FASTCGI] Logging Message-ID: Hi all, I have a fastcgi based service with Apache 2.2 with the configuration bellow. I am facing logging problems because all 50 processes write to the same file (error_log) and the messages are mixed. Is there any configuration in order to have one log file per each fast-cgi process? Httpd.conf ErrorLog "logs/error_log" LoadModule fastcgi_module modules/mod_fastcgi.so Alias /fcgi-bin/ /app/onlinec/applon/exe/ AllowOverride None Allow from all # # Start a "static" server at httpd initialization inside the scope of the SetHandler # FastCgiServer /app/onlinec/applon/exe/elag -processes 50 -idle-timeout 180 -restart-delay 5 -initial-env LD_PRELOAD=/usr/lib/0 at 0.so.1 -initial-env NLS _LANG=AMERICAN_AMERICA.WE8ISO8859P1 -initial-env LD_LIBRARY_PATH=/usr/openwin/lib:/app/fcgi/lib:/app/oracle/instclient -initial-env DBCONNECT=XXXXX/XXXXX at XXXXX:####/XXXXX -initial-env ENVFILE=/app/onlinec/resources/config/applon.env # LogFormat "%P %t %>s %T %a %B \"%r\"" common CustomLog logs/access_log common Best regards, -- Paulo Jorge Dias -------------- next part -------------- An HTML attachment was scrubbed... URL: From gordon at group309.com Tue Apr 13 09:43:27 2010 From: gordon at group309.com (Gordon Colburn) Date: Tue, 13 Apr 2010 09:43:27 -0400 Subject: [FASTCGI] Logging In-Reply-To: References: Message-ID: <001e01cadb0f$4d716570$0138a8c0@mozart> I don't know of any good way to configure one log file per process. Instead, consider using a single log file (like you are now) and define your logging format so that it contains a process identifier of some kind (e.g. PID). If you are running on a Unix-like OS, you can then use the various command-line tools (e.g. awk, grep) to filter out the messages for a particular process if needed. With this approach, by default you have an aggregate view of the logs but it is also easy to view the logs by process (I find myself needing to do both on occasion). Plus, it is less cumbersome (IMO) to have a single log file than 50. BTW, if you are deploying to a Windows machine, you can still use awk, grep, etc. to filter the logs; just install Cygwin (www.cygwin.com ). Regards, Gordon _____ From: fastcgi-developers-bounces+gordon=group309.com at mailman.fastcgi.com [mailto:fastcgi-developers-bounces+gordon=group309.com at mailman.fastcgi.com] On Behalf Of Paulo Jorge Dias Sent: Tuesday, April 13, 2010 4:26 AM To: fastcgi-developers at mailman.fastcgi.com Subject: [FASTCGI] Logging Hi all, I have a fastcgi based service with Apache 2.2 with the configuration bellow. I am facing logging problems because all 50 processes write to the same file (error_log) and the messages are mixed. Is there any configuration in order to have one log file per each fast-cgi process? Httpd.conf ErrorLog "logs/error_log" LoadModule fastcgi_module modules/mod_fastcgi.so Alias /fcgi-bin/ /app/onlinec/applon/exe/ AllowOverride None Allow from all # # Start a "static" server at httpd initialization inside the scope of the SetHandler # FastCgiServer /app/onlinec/applon/exe/elag -processes 50 -idle-timeout 180 -restart-delay 5 -initial-env LD_PRELOAD=/usr/lib/0 at 0.so.1 -initial-env NLS _LANG=AMERICAN_AMERICA.WE8ISO8859P1 -initial-env LD_LIBRARY_PATH=/usr/openwin/lib:/app/fcgi/lib:/app/oracle/instclient -initial-env DBCONNECT=XXXXX/XXXXX at XXXXX:####/XXXXX -initial-env ENVFILE=/app/onlinec/resources/config/applon.env # LogFormat "%P %t %>s %T %a %B \"%r\"" common CustomLog logs/access_log common Best regards, -- Paulo Jorge Dias -------------- next part -------------- An HTML attachment was scrubbed... URL: From paulojorgedias at gmail.com Tue Apr 13 13:30:24 2010 From: paulojorgedias at gmail.com (Paulo Jorge Dias) Date: Tue, 13 Apr 2010 18:30:24 +0100 Subject: [FASTCGI] Logging In-Reply-To: <001e01cadb0f$4d716570$0138a8c0@mozart> References: <001e01cadb0f$4d716570$0138a8c0@mozart> Message-ID: Unfortunately I have already thought of use that approach but the application I am talking about is a migration from other technology and all the logging is hardcoded and spread over many files. I notice also situations where the mix of logging is not only at line level, but also at string level, like two or more SQL statements mixed. Best regards, 2010/4/13 Gordon Colburn > I don?t know of any good way to configure one log file per process. > Instead, consider using a single log file (like you are now) and define your > logging format so that it contains a process identifier of some kind (e.g. > PID). If you are running on a Unix-like OS, you can then use the various > command-line tools (e.g. awk, grep) to filter out the messages for a > particular process if needed. With this approach, by default you have an > aggregate view of the logs but it is also easy to view the logs by process > (I find myself needing to do both on occasion). Plus, it is less cumbersome > (IMO) to have a single log file than 50. > > > > BTW, if you are deploying to a Windows machine, you can still use awk, > grep, etc. to filter the logs; just install Cygwin (www.cygwin.com). > > > > Regards, > > Gordon > > > ------------------------------ > > *From:* fastcgi-developers-bounces+gordon=group309.com at mailman.fastcgi.com[mailto: > fastcgi-developers-bounces+gordon = > group309.com at mailman.fastcgi.com] *On Behalf Of *Paulo Jorge Dias > *Sent:* Tuesday, April 13, 2010 4:26 AM > *To:* fastcgi-developers at mailman.fastcgi.com > *Subject:* [FASTCGI] Logging > > > > Hi all, > > > I have a fastcgi based service with Apache 2.2 with the configuration > bellow. > > I am facing logging problems because all 50 processes write to the same > file (error_log) and the messages are mixed. > > Is there any configuration in order to have one log file per each fast-cgi > process? > > Httpd.conf > > ErrorLog "logs/error_log" > LoadModule fastcgi_module modules/mod_fastcgi.so > > Alias /fcgi-bin/ /app/onlinec/applon/exe/ > > AllowOverride None > Allow from all > > # > # Start a "static" server at httpd initialization inside the scope of > the SetHandler > # > FastCgiServer /app/onlinec/applon/exe/elag -processes 50 -idle-timeout > 180 -restart-delay 5 -initial-env LD_PRELOAD=/usr/lib/0 at 0.so.1-initial-env NLS > _LANG=AMERICAN_AMERICA.WE8ISO8859P1 -initial-env > LD_LIBRARY_PATH=/usr/openwin/lib:/app/fcgi/lib:/app/oracle/instclient > -initial-env DBCONNECT=XXXXX/XXXXX at XXXXX:####/XXXXX -initial-env > ENVFILE=/app/onlinec/resources/config/applon.env > # > LogFormat "%P %t %>s %T %a %B \"%r\"" common > CustomLog logs/access_log common > > > Best regards, > > -- > Paulo Jorge Dias > -- Paulo Jorge Dias -------------- next part -------------- An HTML attachment was scrubbed... URL: From charles_thomas at mac.com Wed Apr 14 17:03:29 2010 From: charles_thomas at mac.com (Tom Bowden) Date: Wed, 14 Apr 2010 16:03:29 -0500 Subject: [FASTCGI] Logging In-Reply-To: References: <001e01cadb0f$4d716570$0138a8c0@mozart> Message-ID: I typically use syslog() for that sort of thing -- it is extremely flexible and lets me do things like specify rules to transfer logs to other systems/syslog daemons. On Apr 13, 2010, at 12:30 PM, Paulo Jorge Dias wrote: > Unfortunately I have already thought of use that approach but the > application I am talking about is a migration from other technology > and all the logging is hardcoded and spread over many files. > > I notice also situations where the mix of logging is not only at > line level, but also at string level, like two or more SQL > statements mixed. > > Best regards, > > 2010/4/13 Gordon Colburn > I don?t know of any good way to configure one log file per process. > Instead, consider using a single log file (like you are now) and > define your logging format so that it contains a process identifier > of some kind (e.g. PID). If you are running on a Unix-like OS, you > can then use the various command-line tools (e.g. awk, grep) to > filter out the messages for a particular process if needed. With > this approach, by default you have an aggregate view of the logs > but it is also easy to view the logs by process (I find myself > needing to do both on occasion). Plus, it is less cumbersome (IMO) > to have a single log file than 50. > > > BTW, if you are deploying to a Windows machine, you can still use > awk, grep, etc. to filter the logs; just install Cygwin > (www.cygwin.com). > > > Regards, > > Gordon > > > From: fastcgi-developers-bounces > +gordon=group309.com at mailman.fastcgi.com [mailto:fastcgi-developers- > bounces+gordon=group309.com at mailman.fastcgi.com] On Behalf Of Paulo > Jorge Dias > Sent: Tuesday, April 13, 2010 4:26 AM > To: fastcgi-developers at mailman.fastcgi.com > Subject: [FASTCGI] Logging > > > Hi all, > > > > I have a fastcgi based service with Apache 2.2 with the > configuration bellow. > > I am facing logging problems because all 50 processes write to the > same file (error_log) and the messages are mixed. > > Is there any configuration in order to have one log file per each > fast-cgi process? > > Httpd.conf > > ErrorLog "logs/error_log" > LoadModule fastcgi_module modules/mod_fastcgi.so > > Alias /fcgi-bin/ /app/onlinec/applon/exe/ > > AllowOverride None > Allow from all > > # > # Start a "static" server at httpd initialization inside the > scope of the SetHandler > # > FastCgiServer /app/onlinec/applon/exe/elag -processes 50 -idle- > timeout 180 -restart-delay 5 -initial-env LD_PRELOAD=/usr/lib/ > 0 at 0.so.1 -initial-env NLS > _LANG=AMERICAN_AMERICA.WE8ISO8859P1 -initial-env LD_LIBRARY_PATH=/ > usr/openwin/lib:/app/fcgi/lib:/app/oracle/instclient -initial-env > DBCONNECT=XXXXX/XXXXX at XXXXX:####/XXXXX -initial-env ENVFILE=/app/ > onlinec/resources/config/applon.env > # > LogFormat "%P %t %>s %T %a %B \"%r\"" common > CustomLog logs/access_log common > > > Best regards, > > -- > Paulo Jorge Dias > > > > > -- > Paulo Jorge Dias > _______________________________________________ > FastCGI-developers mailing list > FastCGI-developers at mailman.fastcgi.com > http://mailman.pins.net/mailman/listinfo.cgi/fastcgi-developers -------------- next part -------------- An HTML attachment was scrubbed... URL: From rogerdpack2 at gmail.com Fri Apr 16 13:43:36 2010 From: rogerdpack2 at gmail.com (Roger Pack) Date: Fri, 16 Apr 2010 11:43:36 -0600 Subject: [FASTCGI] unable to build in mingw Message-ID: Using msys, I get this: ... gcc -DHAVE_CONFIG_H -I. -I. -I.. -I../include -g -O2 -Wall -c os_unix.c -MT libfcgi_la-os_unix.lo -MD -MP -MF .deps/libfcgi_la-os_unix.TPlo -DDLL_EXPORT -DPIC -o .libs/libfcgi_la-os_unix.lo os_unix.c:31:23: arpa/inet.h: No such file or directory os_unix.c:37:25: netinet/tcp.h: No such file or directory os_unix.c:43:20: sys/un.h: No such file or directory Thoughts? Thanks. -rp From jlsandell at gmail.com Sat Apr 17 14:30:24 2010 From: jlsandell at gmail.com (Jeremy Sandell) Date: Sat, 17 Apr 2010 14:30:24 -0400 Subject: [FASTCGI] unable to build in mingw In-Reply-To: References: Message-ID: Are you running configure? Looking through the script, it has SYSTEM hard coded to "unix" - the C library built here after changing it to "win32" - the C++ library, not so much. Hope this helps. Jeremy Sandell On Fri, Apr 16, 2010 at 1:43 PM, Roger Pack wrote: > Using msys, I get this: > > ... > gcc -DHAVE_CONFIG_H -I. -I. -I.. -I../include -g -O2 -Wall -c > os_unix.c -MT libfcgi_la-os_unix.lo -MD -MP -MF > .deps/libfcgi_la-os_unix.TPlo  -DDLL_EXPORT -DPIC -o > .libs/libfcgi_la-os_unix.lo > os_unix.c:31:23: arpa/inet.h: No such file or directory > os_unix.c:37:25: netinet/tcp.h: No such file or directory > os_unix.c:43:20: sys/un.h: No such file or directory > > Thoughts? > Thanks. > -rp > _______________________________________________ > FastCGI-developers mailing list > FastCGI-developers at mailman.fastcgi.com > http://mailman.pins.net/mailman/listinfo.cgi/fastcgi-developers > From svensvenson3 at gmail.com Tue Apr 20 20:32:12 2010 From: svensvenson3 at gmail.com (Sven Svenson) Date: Tue, 20 Apr 2010 17:32:12 -0700 Subject: [FASTCGI] FastCGI still a smart choice in 2010? Message-ID: What's happened with the attitudes developers have about FastCGI? It seems that FastCGI is not the preferred way of deploying web applications outside the Perl world. Are the Java, PHP, Ruby and Python folks right to be using something other than FastCGI? Is FastCGI still around only because it is a legacy technology (Cobol is still around too!) or is FastCGI still a good choice for a high performance web site? Thanks. /sven From webmaster at cosmicperl.com Tue Apr 20 20:48:13 2010 From: webmaster at cosmicperl.com (Lyle) Date: Wed, 21 Apr 2010 01:48:13 +0100 Subject: [FASTCGI] FastCGI still a smart choice in 2010? In-Reply-To: References: Message-ID: <4BCE4B4D.7000009@cosmicperl.com> Sven Svenson wrote: > What's happened with the attitudes developers have about FastCGI? It > seems that FastCGI is not the preferred way of deploying web > applications outside the Perl world. Are the Java, PHP, Ruby and > Python folks right to be using something other than FastCGI? Is > FastCGI still around only because it is a legacy technology (Cobol is > still around too!) or is FastCGI still a good choice for a high > performance web site? > What do you think the other folks are recommending? As far as I've seen, if you want to have your software written in any of the above languages, fast and portable between different platforms and webservers, then FastCGI is still very much the way to go. (sorry for the long sentence) Lyle From unique at idempot.net Tue Apr 20 21:37:30 2010 From: unique at idempot.net (Matthew Weigel) Date: Tue, 20 Apr 2010 20:37:30 -0500 Subject: [FASTCGI] FastCGI still a smart choice in 2010? In-Reply-To: References: Message-ID: <4BCE56DA.6030809@idempot.net> On 4/20/2010 7:32 PM, Sven Svenson wrote: > What's happened with the attitudes developers have about FastCGI? It > seems that FastCGI is not the preferred way of deploying web > applications outside the Perl world. Are the Java, PHP, Ruby and > Python folks right to be using something other than FastCGI? Java is an interesting case, because trying to stay true to the "compile once, debug^H^H^H^H^Hrun anywhere" idea leads naturally to "the whole stack is in Java." Once you get there, language-generic standards like FastCGI aren't nearly as interesting. You also need a massively complex set of standards and APIs in order to make the whole thing sufficiently "enterprise ready." :-) As for PHP... um, my web server runs quite a bit of PHP via FastCGI. Works great, and lets me isolate PHP applications (file access, privilege, etc.) enough that I feel *comfortable* running PHP applications. Ruby (...on Rails, particularly) is a sad case, there are quite a few completely different and incompatible approaches to handling the web server/RoR communication layer, each one trying to make RoR performance less apparent. I'm not sure there's *that* much consensus on the 'right way' to deploy Ruby on Rails applications, but FastCGI is certainly one of them. Python... I thought the standard way of doing things in Python was WSGI, which can be (and has been) implemented on top of FastCGI. Depending on which web server you choose for the other side of WSGI, FastCGI underneath may or may not be the best choice. Past that, I've seen some interesting work in using FastCGI to provide a low-impact embedded web application in a program that is, overall, not a web application - without having to have an entire HTTP stack embedded. And regarding "high performance" web sites, some of the fastest web servers around - such as lighttpd and nginx - rely on FastCGI to support secure, platform-agnostic, fast web applications. Apache's FastCGI support may not be as good, but they tend to prefer emebedding everything in monstrous insecure processes anyway. :-) (Lots of bombast here, everyone should free to provide more balanced views should they desire!) -- Matthew Weigel hacker unique & idempot . ent From gordon at group309.com Tue Apr 20 22:06:44 2010 From: gordon at group309.com (Gordon Colburn) Date: Tue, 20 Apr 2010 22:06:44 -0400 Subject: [FASTCGI] FastCGI still a smart choice in 2010? In-Reply-To: <4BCE4B4D.7000009@cosmicperl.com> References: <4BCE4B4D.7000009@cosmicperl.com> Message-ID: <000d01cae0f7$4d7c0dc0$04ee1c0a@mozart> There is not much reason to develop fastcgi Java applications as Java has a number of very solid web technologies, the most basic of which is the servlet spec. Servlets have been around for over 10 years, provide similar functionality to fastcgi, are highly scalable, and are portable across application servers. In fact, I'm not aware of anyone who is currently developing Java fastcgi apps. However, fastcgi is still commonly used to deploy PHP apps; on Apache, fastcgi is arguably a more scalable technique for deploying PHP than is mod_php. And, fastcgi is a great way to deploy C/C++ web applications, IMO. Regards, Gordon -----Original Message----- From: fastcgi-developers-bounces+gordon=group309.com at mailman.fastcgi.com [mailto:fastcgi-developers-bounces+gordon=group309.com at mailman.fastcgi.com] On Behalf Of Lyle Sent: Tuesday, April 20, 2010 8:48 PM To: Sven Svenson Cc: fastcgi-developers at mailman.pins.net Subject: Re: [FASTCGI] FastCGI still a smart choice in 2010? Sven Svenson wrote: > What's happened with the attitudes developers have about FastCGI? It > seems that FastCGI is not the preferred way of deploying web > applications outside the Perl world. Are the Java, PHP, Ruby and > Python folks right to be using something other than FastCGI? Is > FastCGI still around only because it is a legacy technology (Cobol is > still around too!) or is FastCGI still a good choice for a high > performance web site? > What do you think the other folks are recommending? As far as I've seen, if you want to have your software written in any of the above languages, fast and portable between different platforms and webservers, then FastCGI is still very much the way to go. (sorry for the long sentence) Lyle _______________________________________________ FastCGI-developers mailing list FastCGI-developers at mailman.fastcgi.com http://mailman.pins.net/mailman/listinfo.cgi/fastcgi-developers From jon+fastcgi-developers at unequivocal.co.uk Wed Apr 21 04:47:20 2010 From: jon+fastcgi-developers at unequivocal.co.uk (Jon Ribbens) Date: Wed, 21 Apr 2010 09:47:20 +0100 Subject: [FASTCGI] FastCGI still a smart choice in 2010? In-Reply-To: <000d01cae0f7$4d7c0dc0$04ee1c0a@mozart> References: <4BCE4B4D.7000009@cosmicperl.com> <000d01cae0f7$4d7c0dc0$04ee1c0a@mozart> Message-ID: <20100421084720.GI31281@snowy.squish.net> On Tue, Apr 20, 2010 at 10:06:44PM -0400, Gordon Colburn wrote: > However, fastcgi is still commonly used to deploy PHP apps; on Apache, > fastcgi is arguably a more scalable technique for deploying PHP than is > mod_php. And, fastcgi is a great way to deploy C/C++ web applications, IMO. It's great for Python too - Python can have quite a slow start-up time, so keeping the process running is great. There are several Python FastCGI libraries, including one that I wrote (jonpy). From andylin02 at qq.com Wed Apr 21 05:14:48 2010 From: andylin02 at qq.com (=?ISO-8859-1?B?YW5keWxpbjAy?=) Date: Wed, 21 Apr 2010 17:14:48 +0800 Subject: [FASTCGI] FastCGI still a smart choice in 2010? Message-ID: all, I am staring fast-cgi I think I love it. ------------------ andylin ------------------ Original ------------------ From: "Jon Ribbens"; Date: Wed, Apr 21, 2010 04:47 PM To: "Gordon Colburn"; Cc: "fastcgi-developers"; Subject: Re: [FASTCGI] FastCGI still a smart choice in 2010? On Tue, Apr 20, 2010 at 10:06:44PM -0400, Gordon Colburn wrote: > However, fastcgi is still commonly used to deploy PHP apps; on Apache, > fastcgi is arguably a more scalable technique for deploying PHP than is > mod_php. And, fastcgi is a great way to deploy C/C++ web applications, IMO. It's great for Python too - Python can have quite a slow start-up time, so keeping the process running is great. There are several Python FastCGI libraries, including one that I wrote (jonpy). _______________________________________________ FastCGI-developers mailing list FastCGI-developers at mailman.fastcgi.com http://mailman.pins.net/mailman/listinfo.cgi/fastcgi-developers -------------- next part -------------- An HTML attachment was scrubbed... URL: From rclemley at yahoo.com Wed Apr 21 10:42:44 2010 From: rclemley at yahoo.com (Rob) Date: Wed, 21 Apr 2010 09:42:44 -0500 Subject: [FASTCGI] FastCGI still a smart choice in 2010? In-Reply-To: <000d01cae0f7$4d7c0dc0$04ee1c0a@mozart> References: <4BCE4B4D.7000009@cosmicperl.com> <000d01cae0f7$4d7c0dc0$04ee1c0a@mozart> Message-ID: <4BCF0EE4.3070800@yahoo.com> C++: Together with djb's daemontools, we have machines running hundreds of FastCgiExternalServers written in C++ connected via Unix Domain Sockets to a single apache2 instance. We can restart apache without disturbing the fastcgi backends and we can restart individual backends without disturbing apache or the other backends. With FastCGI, we could also run the FastCgiExternalServer backends on different machines if necessary. These tools are simple, modular and flexible. On 04/20/2010 09:06 PM, Gordon Colburn wrote: > There is not much reason to develop fastcgi Java applications as Java has a > number of very solid web technologies, the most basic of which is the > servlet spec. Servlets have been around for over 10 years, provide similar > functionality to fastcgi, are highly scalable, and are portable across > application servers. In fact, I'm not aware of anyone who is currently > developing Java fastcgi apps. > > However, fastcgi is still commonly used to deploy PHP apps; on Apache, > fastcgi is arguably a more scalable technique for deploying PHP than is > mod_php. And, fastcgi is a great way to deploy C/C++ web applications, IMO. > > Regards, > Gordon > > -----Original Message----- > From: fastcgi-developers-bounces+gordon=group309.com at mailman.fastcgi.com > [mailto:fastcgi-developers-bounces+gordon=group309.com at mailman.fastcgi.com] > On Behalf Of Lyle > Sent: Tuesday, April 20, 2010 8:48 PM > To: Sven Svenson > Cc: fastcgi-developers at mailman.pins.net > Subject: Re: [FASTCGI] FastCGI still a smart choice in 2010? > > Sven Svenson wrote: > >> What's happened with the attitudes developers have about FastCGI? It >> seems that FastCGI is not the preferred way of deploying web >> applications outside the Perl world. Are the Java, PHP, Ruby and >> Python folks right to be using something other than FastCGI? Is >> FastCGI still around only because it is a legacy technology (Cobol is >> still around too!) or is FastCGI still a good choice for a high >> performance web site? >> >> > > What do you think the other folks are recommending? As far as I've seen, > if you want to have your software written in any of the above languages, > fast and portable between different platforms and webservers, then > FastCGI is still very much the way to go. (sorry for the long sentence) > > > Lyle > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From rclemley at yahoo.com Wed Apr 21 14:37:02 2010 From: rclemley at yahoo.com (Rob) Date: Wed, 21 Apr 2010 13:37:02 -0500 Subject: [FASTCGI] FastCGI still a smart choice in 2010? In-Reply-To: <547558.35828.qm@web35301.mail.mud.yahoo.com> References: <4BCE4B4D.7000009@cosmicperl.com> <000d01cae0f7$4d7c0dc0$04ee1c0a@mozart> <4BCF0EE4.3070800@yahoo.com> <547558.35828.qm@web35301.mail.mud.yahoo.com> Message-ID: <4BCF45CE.207@yahoo.com> On 04/21/2010 11:21 AM, Jean-Christophe Roux wrote: > Are you using a C++ fastcgi library? Which one(s)? We use an in-house c++ library wrapped around fcgiapp.h that's rather dependent on other lower level libraries in our app. Theres a "server" class which calls FCGX_Init() and then calls the "loop" (or listener) class which calls FCGX_OpenSocket(), and then loops, in each iteration creating one of our FastCGIRequest(socket) objects, and then calling the request object's Accept() method. If the request object successfully accepts a connection, then the "loop" object puts the "request" object on a queue object constructed using an STL container and boost::threads, boost::mutexes, and boost::conditions. There's a pool of "server threads" waiting for fastcgi requests to be placed on the synchronized queue. A "server thread" then retrieves a request off the queue and responds to the client. The server thread sends all output back to the client through methods on the request object. In this scheme, the "FastCGIRequest" object is the hub of activity and data, along with the synchronized queue. The class library was written circa 2004 according to the FastCGI spec and has run largely unchanged since that time. To accomplish this, the following functions are referenced from fcgiapp.h: server object: FCGX_Init loop object: FCGX_OpenSocket fcgi request object: FCGX_InitRequest FCGX_Accept_r FCGX_GetError FCGX_GetStr FCGX_PutStr FCGX_Finish_r On 04/21/2010 11:21 AM, Jean-Christophe Roux wrote: > Hi Rob, > > Are you using a C++ fastcgi library? Which one(s)? > > Thanks > > ------------------------------------------------------------------------ > *From:* Rob > *To:* Gordon Colburn > *Cc:* fastcgi-developers at mailman.pins.net > *Sent:* Wed, April 21, 2010 10:42:44 AM > *Subject:* Re: [FASTCGI] FastCGI still a smart choice in 2010? > > C++: Together with djb's daemontools, we have machines running > hundreds of FastCgiExternalServers written in C++ connected via Unix > Domain Sockets to a single apache2 instance. > > We can restart apache without disturbing the fastcgi backends and we > can restart individual backends without disturbing apache or the other > backends. > > With FastCGI, we could also run the FastCgiExternalServer backends on > different machines if necessary. > > These tools are simple, modular and flexible. > > > On 04/20/2010 09:06 PM, Gordon Colburn wrote: >> There is not much reason to develop fastcgi Java applications as Java has a >> number of very solid web technologies, the most basic of which is the >> servlet spec. Servlets have been around for over 10 years, provide similar >> functionality to fastcgi, are highly scalable, and are portable across >> application servers. In fact, I'm not aware of anyone who is currently >> developing Java fastcgi apps. >> >> However, fastcgi is still commonly used to deploy PHP apps; on Apache, >> fastcgi is arguably a more scalable technique for deploying PHP than is >> mod_php. And, fastcgi is a great way to deploy C/C++ web applications, IMO. >> >> Regards, >> Gordon >> >> -----Original Message----- >> From: fastcgi-developers-bounces+gordon=group309.com at mailman.fastcgi.com >> [mailto:fastcgi-developers-bounces+gordon=group309.com at mailman.fastcgi.com] >> On Behalf Of Lyle >> Sent: Tuesday, April 20, 2010 8:48 PM >> To: Sven Svenson >> Cc: fastcgi-developers at mailman.pins.net >> Subject: Re: [FASTCGI] FastCGI still a smart choice in 2010? >> >> Sven Svenson wrote: >> >>> What's happened with the attitudes developers have about FastCGI? It >>> seems that FastCGI is not the preferred way of deploying web >>> applications outside the Perl world. Are the Java, PHP, Ruby and >>> Python folks right to be using something other than FastCGI? Is >>> FastCGI still around only because it is a legacy technology (Cobol is >>> still around too!) or is FastCGI still a good choice for a high >>> performance web site? >>> >>> >> What do you think the other folks are recommending? As far as I've seen, >> if you want to have your software written in any of the above languages, >> fast and portable between different platforms and webservers, then >> FastCGI is still very much the way to go. (sorry for the long sentence) >> >> >> Lyle >> >> > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From wollman at csail.mit.edu Wed Apr 21 17:30:16 2010 From: wollman at csail.mit.edu (Garrett Wollman) Date: Wed, 21 Apr 2010 17:30:16 -0400 Subject: [FASTCGI] FastCGI still a smart choice in 2010? In-Reply-To: References: Message-ID: <19407.28264.868218.229168@khavrinen.csail.mit.edu> < said: > What's happened with the attitudes developers have about FastCGI? It > seems that FastCGI is not the preferred way of deploying web > applications outside the Perl world. Are the Java, PHP, Ruby and > Python folks right to be using something other than FastCGI? Is > FastCGI still around only because it is a legacy technology (Cobol is > still around too!) or is FastCGI still a good choice for a high > performance web site? As always, it depends on what your requirements are. We stick with FastCGI because it's programming-environment-independent. We can't tell our users that they have to write in (PHP, Perl, Python, Ruby, Java, ...) so we need to support all of those systems (and the environments that layer on top of them) with a single mechanism. The one that comes closest to fitting the bill is FastCGI. (If anyone can show me a way to run Java crudlets under FastCGI, I'd love to see it, as we occasionally get requests for that.) -GAWollman From charles_thomas at mac.com Thu Apr 22 14:04:31 2010 From: charles_thomas at mac.com (Tom Bowden) Date: Thu, 22 Apr 2010 13:04:31 -0500 Subject: [FASTCGI] FastCGI still a smart choice in 2010? In-Reply-To: <19407.28264.868218.229168@khavrinen.csail.mit.edu> References: <19407.28264.868218.229168@khavrinen.csail.mit.edu> Message-ID: Personally - I would get really upset if fastCGI went away -- for the exact reason below. In our particular case I am not going to expose our tomcat server to the outside world or rely on java to process several thousand requests a minute (not that it can't -- but I am old and hornery). I LOVE fastcgi because it is not related to php/perl or java. Tom On Apr 21, 2010, at 4:30 PM, Garrett Wollman wrote: > < said: > >> What's happened with the attitudes developers have about FastCGI? It >> seems that FastCGI is not the preferred way of deploying web >> applications outside the Perl world. Are the Java, PHP, Ruby and >> Python folks right to be using something other than FastCGI? Is >> FastCGI still around only because it is a legacy technology (Cobol is >> still around too!) or is FastCGI still a good choice for a high >> performance web site? > > As always, it depends on what your requirements are. We stick with > FastCGI because it's programming-environment-independent. We can't > tell our users that they have to write in (PHP, Perl, Python, Ruby, > Java, ...) so we need to support all of those systems (and the > environments that layer on top of them) with a single mechanism. The > one that comes closest to fitting the bill is FastCGI. (If anyone can > show me a way to run Java crudlets under FastCGI, I'd love to see it, > as we occasionally get requests for that.) > > -GAWollman > > _______________________________________________ > FastCGI-developers mailing list > FastCGI-developers at mailman.fastcgi.com > http://mailman.pins.net/mailman/listinfo.cgi/fastcgi-developers From chapmanm at pixia.com Thu Apr 22 14:08:35 2010 From: chapmanm at pixia.com (Martin Chapman) Date: Thu, 22 Apr 2010 12:08:35 -0600 Subject: [FASTCGI] FastCGI still a smart choice in 2010? In-Reply-To: References: <19407.28264.868218.229168@khavrinen.csail.mit.edu> Message-ID: <313F8A5DFBB94893B3B7A5EE0C205E88@chapmanm64> Agreed, long live FastCGI!!! Martin -----Original Message----- From: fastcgi-developers-bounces+chapmanm=pixia.com at mailman.fastcgi.com [mailto:fastcgi-developers-bounces+chapmanm=pixia.com at mailman.fastcgi.com] On Behalf Of Tom Bowden Sent: Thursday, April 22, 2010 12:05 PM To: Garrett Wollman Cc: fastcgi-developers at mailman.pins.net Subject: Re: [FASTCGI] FastCGI still a smart choice in 2010? Personally - I would get really upset if fastCGI went away -- for the exact reason below. In our particular case I am not going to expose our tomcat server to the outside world or rely on java to process several thousand requests a minute (not that it can't -- but I am old and hornery). I LOVE fastcgi because it is not related to php/perl or java. Tom On Apr 21, 2010, at 4:30 PM, Garrett Wollman wrote: > < said: > >> What's happened with the attitudes developers have about FastCGI? It >> seems that FastCGI is not the preferred way of deploying web >> applications outside the Perl world. Are the Java, PHP, Ruby and >> Python folks right to be using something other than FastCGI? Is >> FastCGI still around only because it is a legacy technology (Cobol is >> still around too!) or is FastCGI still a good choice for a high >> performance web site? > > As always, it depends on what your requirements are. We stick with > FastCGI because it's programming-environment-independent. We can't > tell our users that they have to write in (PHP, Perl, Python, Ruby, > Java, ...) so we need to support all of those systems (and the > environments that layer on top of them) with a single mechanism. The > one that comes closest to fitting the bill is FastCGI. (If anyone can > show me a way to run Java crudlets under FastCGI, I'd love to see it, > as we occasionally get requests for that.) > > -GAWollman > > _______________________________________________ > FastCGI-developers mailing list > FastCGI-developers at mailman.fastcgi.com > http://mailman.pins.net/mailman/listinfo.cgi/fastcgi-developers _______________________________________________ FastCGI-developers mailing list FastCGI-developers at mailman.fastcgi.com http://mailman.pins.net/mailman/listinfo.cgi/fastcgi-developers From webmaster at cosmicperl.com Fri Apr 23 10:07:52 2010 From: webmaster at cosmicperl.com (Lyle) Date: Fri, 23 Apr 2010 15:07:52 +0100 Subject: [FASTCGI] FastCGI still a smart choice in 2010? In-Reply-To: References: <19407.28264.868218.229168@khavrinen.csail.mit.edu> Message-ID: <4BD1A9B8.4030208@cosmicperl.com> Not to mention Microsoft releasing FastCGI for IIS 6 & 7. Lyle Tom Bowden wrote: > Personally - I would get really upset if fastCGI went away -- for the > exact reason below. > > In our particular case I am not going to expose our tomcat server to > the outside world or rely on java to process several thousand requests > a minute (not that it can't -- but I am old and hornery). I LOVE > fastcgi because it is not related to php/perl or java. > > Tom > > On Apr 21, 2010, at 4:30 PM, Garrett Wollman wrote: > >> <> said: >> >>> What's happened with the attitudes developers have about FastCGI? It >>> seems that FastCGI is not the preferred way of deploying web >>> applications outside the Perl world. Are the Java, PHP, Ruby and >>> Python folks right to be using something other than FastCGI? Is >>> FastCGI still around only because it is a legacy technology (Cobol is >>> still around too!) or is FastCGI still a good choice for a high >>> performance web site? >> >> As always, it depends on what your requirements are. We stick with >> FastCGI because it's programming-environment-independent. We can't >> tell our users that they have to write in (PHP, Perl, Python, Ruby, >> Java, ...) so we need to support all of those systems (and the >> environments that layer on top of them) with a single mechanism. The >> one that comes closest to fitting the bill is FastCGI. (If anyone can >> show me a way to run Java crudlets under FastCGI, I'd love to see it, >> as we occasionally get requests for that.) >> >> -GAWollman >> >> _______________________________________________ >> FastCGI-developers mailing list >> FastCGI-developers at mailman.fastcgi.com >> http://mailman.pins.net/mailman/listinfo.cgi/fastcgi-developers > > _______________________________________________ > FastCGI-developers mailing list > FastCGI-developers at mailman.fastcgi.com > http://mailman.pins.net/mailman/listinfo.cgi/fastcgi-developers > From contato at fernandomarcelo.com Sat Apr 24 18:27:40 2010 From: contato at fernandomarcelo.com (Fernando Morgenstern) Date: Sat, 24 Apr 2010 19:27:40 -0300 Subject: [FASTCGI] (4)Interrupted system call: unable to connect to cgi daemon after multiple tries Message-ID: <4930D926-4B85-4419-8B36-CB8AAA862F07@fernandomarcelo.com> Hi, I'm running the following setup under a high traffic website: apache 2.2.15 PHP 5.2.13 mod_fastcgi 1.4.6 I have lots of messages like this one on my error log: (4)Interrupted system call: unable to connect to cgi daemon after multiple tries: /usr/local/apache2/cgi-bin/php.fcgi And users complaining about 500 http error. It starts right after apache is up. Could someone tell me why this happens and if there is a way to fix it? Best Regards, Fernando Morgenstern contato at fernandomarcelo.com From ysj.ray at gmail.com Sun Apr 25 23:18:58 2010 From: ysj.ray at gmail.com (Ray Allen) Date: Mon, 26 Apr 2010 11:18:58 +0800 Subject: [FASTCGI] comm with (dynamic) server "xxx" idle timeout (30 sec) Message-ID: Hi, all: In my fastcgi program, sometimes there an error occurred when processing a request. The apache error log showed: FastCGI: comm with (dynamic) server "/home/project/guide_csa/gm-cgi/csa_gm_account_control.py" aborted: (first read) idle timeout (30 sec), referer: ...... FastCGI: incomplete headers (0 bytes) received from server "/home/project/guide_csa/gm-cgi/csa_gm_account_control.py", referer:..... I don't know what caused this error, because in most cases it's OK. Can any body explains this for me? Thanks! -- Ray Allen Best wishes! -------------- next part -------------- An HTML attachment was scrubbed... URL: From copyfilenames at gmail.com Tue Apr 27 05:32:37 2010 From: copyfilenames at gmail.com (Copyfilenames Cole) Date: Tue, 27 Apr 2010 17:32:37 +0800 Subject: [FASTCGI] Can FastCgi in c language manipulate MS office file(word, excel), is there a c API for Microsoft Documents? Message-ID: I can easily find poi in Apache foundation Apache POI - the Java API for Microsoft Documents Is there a c api for Microsoft Documents I do not want to switch to java from fastcgi. From Ben at redsnapper.net Thu Apr 29 09:07:43 2010 From: Ben at redsnapper.net (Ben Griffin) Date: Thu, 29 Apr 2010 14:07:43 +0100 Subject: [FASTCGI] dlopen support for fcgi_streambuf Message-ID: Having spent the last few days attempting to use dlopen with fastcgi, I suddenly notice that there are no dlsym accessors for initialising fcgi_streambufs. Have I missed something? -B. From copyfilenames at gmail.com Thu Apr 29 09:54:19 2010 From: copyfilenames at gmail.com (Copyfilenames Cole) Date: Thu, 29 Apr 2010 21:54:19 +0800 Subject: [FASTCGI] I got error FastCGI: can't start server "/usr/www/fcgi-bin/echo.fcg" (pid 8500), execle() failed: Exec format error Message-ID: FastCGI: can't start server "/usr/www/fcgi-bin/echo.fcg" (pid 8500), execle() failed: Exec format error [Thu Apr 29 18:31:42 2010] [warn] FastCGI: server "/usr/www/fcgi-bin/echo.fcg" (pid 8500) terminated by calling exit with status '255' [Thu Apr 29 18:31:42 2010] [warn] FastCGI: (dynamic) server "/usr/www/fcgi-bin/hello.fcg" has failed to remain running for 30 seconds given 3 attempts, its restart interval has been backed off to 600 seconds [Thu Apr 29 18:31:42 2010] [warn] FastCGI: server "/usr/www/fcgi-bin/echo.fcg" has failed to remain running for 30 seconds given 3 attempts, its restart interval has been backed off to 600 seconds How to fix it? From jon+fastcgi-developers at unequivocal.co.uk Thu Apr 29 09:57:13 2010 From: jon+fastcgi-developers at unequivocal.co.uk (Jon Ribbens) Date: Thu, 29 Apr 2010 14:57:13 +0100 Subject: [FASTCGI] I got error FastCGI: can't start server "/usr/www/fcgi-bin/echo.fcg" (pid 8500), execle() failed: Exec format error In-Reply-To: References: Message-ID: <20100429135713.GO25603@snowy.squish.net> On Thu, Apr 29, 2010 at 09:54:19PM +0800, Copyfilenames Cole wrote: > FastCGI: can't start server "/usr/www/fcgi-bin/echo.fcg" (pid 8500), > execle() failed: Exec format error What happens if you run /usr/www/fcgi-bin/echo.fcg manually? What file type is echo.fcg? From ben at redsnapper.net Thu Apr 29 11:15:48 2010 From: ben at redsnapper.net (Ben Griffin) Date: Thu, 29 Apr 2010 16:15:48 +0100 Subject: [FASTCGI] License query Message-ID: Just had legal over the license. We were wondering how Open Market, Inc. envisage the following term as being enforcable: "If modifications to this Software and Documentation have new licensing terms, the new terms must be clearly indicated on the first page of each file where they apply." Clearly if modifications to the software/documentation are copyrighted by the new authors (remember that of course changing the license text counts as modification) then Open Market, Inc. no longer has any right or claim regarding the way in which any terms are to be indicated, which rather nullifies the aforementioned term. I guess it makes sense as a polite request. But that's about it. Am I missing something? From jon+fastcgi-developers at unequivocal.co.uk Thu Apr 29 11:33:40 2010 From: jon+fastcgi-developers at unequivocal.co.uk (Jon Ribbens) Date: Thu, 29 Apr 2010 16:33:40 +0100 Subject: [FASTCGI] License query In-Reply-To: References: Message-ID: <20100429153340.GC20094@snowy.squish.net> On Thu, Apr 29, 2010 at 04:15:48PM +0100, Ben Griffin wrote: > Just had legal over the license. > > We were wondering how Open Market, Inc. envisage the following term > as being enforcable: > > "If modifications to this Software and Documentation have new > licensing terms, the new terms must be clearly indicated on the > first page of each file where they apply." > > Clearly if modifications to the software/documentation are > copyrighted by the new authors (remember that of course changing the > license text counts as modification) then Open Market, Inc. no > longer has any right or claim regarding the way in which any terms > are to be indicated, which rather nullifies the aforementioned term. > > I guess it makes sense as a polite request. But that's about it. > > Am I missing something? Yes - just because you modify a file slightly that doesn't magically make the copyright ownership of the rest of the file suddenly go away. From unique at idempot.net Thu Apr 29 12:43:00 2010 From: unique at idempot.net (Matthew Weigel) Date: Thu, 29 Apr 2010 11:43:00 -0500 Subject: [FASTCGI] License query In-Reply-To: References: Message-ID: <207bc9c8841e51f341d1e979f35a439f@patience.idempot.net> On Thu, 29 Apr 2010 16:15:48 +0100, Ben Griffin wrote: > Just had legal over the license. > > We were wondering how Open Market, Inc. envisage the following term as > being enforcable: > > "If modifications to this Software and Documentation have new licensing > terms, the new terms must be clearly indicated on the first page of each > file where they apply." > > Clearly if modifications to the software/documentation are copyrighted by > the new authors (remember that of course changing the license text counts > as modification) Modifying the license text is not permitted by the BSD license, nor the FastCGI license. See here: "Open Market permits you to use, copy, modify, distribute, and license this Software and the Documentation for any purpose, provided that existing copyright notices are retained in all copies and that this notice is included verbatim in any distributions." > then Open Market, Inc. no longer has any right or claim > regarding the way in which any terms are to be indicated, which rather > nullifies the aforementioned term. Your right to use, copy, modify, distribute, or license the FastCGI software comes to you *only* through the license. When you make changes, you hold copyright on the changes, not the entire combined work. If you use, copy, modify, distribute, or license the combined work, the portions on which you do not hold copyright can only be used, copied, modified, distributed, or licensed under the terms of the FastCGI license, which require you to clearly indicate on the first page of each file on which you hold partial copyright that different license terms apply to portions of that file. It's more than a polite request, it's a legal requirement. > Am I missing something? Everyone makes mistakes, but you should reconsider which legal you have over the license. -- Matthew Weigel hacker, not a lawyer, but this is pretty basic unique & idempot . ent From rogerdpack2 at gmail.com Thu Apr 29 13:51:08 2010 From: rogerdpack2 at gmail.com (Roger Pack) Date: Thu, 29 Apr 2010 11:51:08 -0600 Subject: [FASTCGI] unable to build in mingw In-Reply-To: References: Message-ID: On Sat, Apr 17, 2010 at 12:30 PM, Jeremy Sandell wrote: > Are you running configure?  Looking through the script, it has SYSTEM > hard coded to "unix" - the C library built here after changing it to > "win32" - the C++ library, not so much. Cool thanks. Consider this thread a request to allow building to work with mingw. Thanks! -rp From jsprenkle at gmail.com Thu Apr 29 13:59:48 2010 From: jsprenkle at gmail.com (Jay Sprenkle) Date: Thu, 29 Apr 2010 12:59:48 -0500 Subject: [FASTCGI] unable to build in mingw In-Reply-To: References: Message-ID: That's going to be difficult. Windows does not treat file handles the same as *nix does. A network handle is different than a file handle in Windows. You need to conditionally compile code that only applies to windows based on the operating system flags and not the compiler flags. Unfortunately most code seems to be written based on the MSC compiler flag and doesn't detect if it's compiled in gcc under windows. mingw does not offer work arounds for the file handles issue either. On Thu, Apr 29, 2010 at 12:51 PM, Roger Pack wrote: > On Sat, Apr 17, 2010 at 12:30 PM, Jeremy Sandell > wrote: > > Are you running configure? Looking through the script, it has SYSTEM > > hard coded to "unix" - the C library built here after changing it to > > "win32" - the C++ library, not so much. > > Cool thanks. > > Consider this thread a request to allow building to work with mingw. > Thanks! > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ysj.ray at gmail.com Fri Apr 30 05:23:50 2010 From: ysj.ray at gmail.com (Ray Allen) Date: Fri, 30 Apr 2010 17:23:50 +0800 Subject: [FASTCGI] Can I do some cleanup work when the request is timeout? Message-ID: Hi, all: When I was using mod_fast cgi, I want to do some cleanup work when the current request is timeout(the -idle-timeout configuration). How can I do that in my applications? For example, I want to kill a long running mysql connection just before the timeout. Thanks! -- Ray Allen Best wishes! -------------- next part -------------- An HTML attachment was scrubbed... URL: