From robert@pauli.utmb.edu Wed Jul 1 13:06:59 1998 Received: from nmr.utmb.edu (dggin1.utmb.edu [129.109.73.3]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id NAA20324 Wed, 1 Jul 1998 13:06:58 -0400 (EDT) Received: from pauli.utmb.edu (pauli.utmb.edu [129.109.73.25]) by nmr.utmb.edu (950413.SGI.8.6.12/950213.SGI.AUTOCF) via ESMTP id MAA16628 for <@nmr.utmb.edu:CHEMISTRY@www.ccl.net>; Wed, 1 Jul 1998 12:20:40 -0500 Received: (from robert@localhost) by pauli.utmb.edu (980616/Sun/9.16./980616/Sun_Microsystems) id MAA08203 for CHEMISTRY@www.ccl.net; Wed, 1 Jul 1998 12:07:45 -0500 (CDT) From: "Robert Fraczkiewicz" Message-Id: <9807011207.ZM8201@pauli.utmb.edu> Date: Wed, 1 Jul 1998 12:07:44 -0500 X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: CHEMISTRY@www.ccl.net Subject: Summary addendum :: Uploading PDB files to a CGI script Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Dear Netters, It seems that Malcolm Gillies has found a perfect explanation of the "CGI bug" reported to CCL a few days ago. With his permission and for the sake of summary completeness, I am including his recent e-mail below. Robert Fraczkiewicz --- Forwarded mail from Malcolm Gillies >> I just did a quick test connecting to the httpd via telnet and feeding >> in text with copy-and-paste, and I can confirm that the cgi process >> deadlocks in the same way, and in the same place. >> >> So I'm fairly confident it's not a problem with the browser. > > Hm... You are right, it must be the server. But that's very interesting, > because you use server from NCSA, and I use one from Netscape (unless > one is based on the other ;-) I had another look at the NCSA httpd server source code yesterday, and as I suspected, it's a design "feature" of the cgi interface. While I would have to check the CGI spec to see if this is purely an implementation detail, or an actual requirement, the circumstances are fairly predictable. The Netscape server is very indirectly descended from the NCSA sever, I think, so it would also be interesting to see whether this is any different on a server with a different provenance, such as CERN, or one of the W3C reference implementations. As you probably know, communication between the cgi process and the httpd is done with standard unix pipes. The NCSA httpd performs the cgi call by first forking off the cgi process, then reading all the POST content off the network socket, then writing it all to the pipe attached to the cgi process, and only then, reading the output of the cgi off the other pipe and writing it to the network socket. Given that pipes only have a certain amount of buffering, this means that interleaving input and output in the cgi process will result in the buffer of the output pipe filling up (as the httpd process is not reading from it -- it's still busy at the stage of writing the POST data to the cgi's input pipe), and therefore the 'write' system call in the cgi (e.g. called by putchar) blocking. Once the write blocks, then the cgi is stuck, and things don't go any further, until the server times out and tries to kill the cgi process. I can envisage alternative server architectures where I/O to the cgi is multiplexed using the select system call, or where the server forks off separate processes for input and output to the CGI, but given the extra overhead and effort, I have my doubts as to whether these have been implemented. So, the upshot appears to be: it's best to read all input before generating any output, at least if you want to be compatible with the NCSA and Netscape httpds (and quite probably a number of others, I would suspect). cheers, Malcolm ---End of forwarded mail from Malcolm Gillies From gadre@chem.unipune.ernet.in Tue Jun 30 09:24:33 1998 Received: from naveen.ncst.ernet.in (naveen.ncst.ernet.in [202.41.110.35]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id JAA03330 Tue, 30 Jun 1998 09:22:17 -0400 (EDT) Received: from unipune.ernet.in (unipune.ernet.in [196.1.114.1]) by naveen.ncst.ernet.in (8.6.12/8.6.6) with ESMTP id SAA18873 for ; Tue, 30 Jun 1998 18:38:07 +0530 Received: from chem.unipune.ernet.in by unipune.ernet.in (8.8.5/UNIPUNE-MAILHUB(02.04.97)) id SAA11535; Tue, 30 Jun 1998 18:19:05 -0500 (GMT) Received: (from gadre@localhost) by chem.unipune.ernet.in (8.8.7/8.8.7) id SAA04486 for CHEMISTRY@www.ccl.net; Tue, 30 Jun 1998 18:32:52 +0500 Date: Tue, 30 Jun 1998 18:32:52 +0500 From: "Prof. Shridhar R. Gadre" Message-Id: <199806301332.SAA04486@chem.unipune.ernet.in> To: CHEMISTRY@www.ccl.net Subject: Contact Distances Dear Sirs and Madams : Pardon me for this query. Maybe the answer is too well-known. Could you give me references to searches done using crystallographic databases of the following. 1. Contacts of a phenyl ring with a nearby acidic hydrogen hanging above the ring. 2. Contacts of an acidic H atom with a methyl group. Thanks a lot!.....................................Shridhar Gadre From gmercier@mail.med.upenn.edu Tue Jun 30 15:03:11 1998 Received: from mail.med.upenn.edu (MAIL.MED.UPENN.EDU [165.123.11.72]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id PAA07018 Tue, 30 Jun 1998 15:03:11 -0400 (EDT) Received: (from gmercier@localhost) by mail.med.upenn.edu (8.8.8/8.8.8) id PAA16352 for chemistry@www.ccl.net; Tue, 30 Jun 1998 15:03:12 -0400 (EDT) From: "Gustavo A. Mercier Jr" Message-Id: <199806301903.PAA16352@mail.med.upenn.edu> Subject: Summary: Fortran compiler in Linux Box To: chemistry@www.ccl.net Date: Tue, 30 Jun 1998 15:03:12 -0400 (EDT) X-Mailer: ELM [version 2.4 PL23-upenn3.1] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit HI! Here is the summary of responses to my query about fortran compilers in Linux boxes... --- From erik@theophys.kth.se Mon Jun 8 02:34:18 1998 My first recommendation would be to go with another hardware platform if you have any possibility. Looking at C or C++ compilers, there are a couple of all right ones for PC (look at KAI for instance, I don't know the url, but have a search at altavista), but I've never seen any fortran compiler for PC's as good as the digital ones for digital alpha. Nevertheless, if you absolutely need to stick with PC, microsoft are selling a digital-made fortran 90 compiler for windows. (I really hate recommending a windows product, but I think it's the best pentium fortran compiler there is.) Sincerely, Erik Lindahl --------------------------------------------------------------------- Erik Lindahl, MSc erik@theophys.kth.se Theoretical Physics, Royal Institute of Technology, STOCKHOLM, SWEDEN Phone: +46 8 7907172 Fax: +46 8 104879 ------ From york@brel.u-strasbg.fr Mon Jun 8 04:10:51 1998 Gustavo, Your question is of interest to me as well, so I hope you post a summary. I would like to point that recently there has become available a free f90 compiler for Linux from the Pacific-Sierra Research Corporation. It looks encouraging, for example see: http://www.psrv.com/ I have no first-hand assessment yet of the quality and performance of this compiler on large production code; however, I would be happy to communicate to you anything I find out. You might also be interested in some f90 compiler comparisons on http://www.cs.rpi.edu/~szymansk/oof90.html. Best of luck, Darrin York ------------------------------------------------------------------------------ : Darrin M. York : : Department of Chemistry and *Laboratoire de Chimie Biophysique : : Chemical Biology Institut Le Bel : : Harvard University Universite Louis Pasteur : : 12 Oxford Street 4, rue Blaise Pascal : : Cambridge, MA USA 02138 67000 Strasbourg, FRANCE : : Tel. (617)495-1782 Tel. 03.88.41.62.85 : : Fax. (617)496-3204 Fax. 03.88.60.63.83 : : e-mail: york@tammy.harvard.edu e-mail: york@brel.u-strasbg.fr : : : : (*) current address : ------------------------------------------------------------------------------ From wws20@cus.cam.ac.uk Mon Jun 8 07:19:07 1998 Gustavo, there is a compiler out from Pacific Sierra. Chout their web-site at www.psrv.com. They have a free linux version (needs g77) and a rather inexpensive optimising compiler for about USD 250. Cheers, WWS ----------------------------------------------------------------------- | Werner W Schulz | | Dept of Chemistry email: wws20@cam.ac.uk | | University of Cambridge Phone: (+44) (0)1223 336 502 | | Lensfield Road Secretary: 1223 336 338 | | Cambridge CB2 1EW Fax: 1223 336 536 | | United Kingdom WWW: | ----------------------------------------------------------------------- From n9644867@cc.wwu.edu Tue Jun 30 17:41:04 1998 Received: from cc.wwu.edu (titan.cc.wwu.edu [140.160.240.18]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id RAA08622 Tue, 30 Jun 1998 17:41:03 -0400 (EDT) Received: from localhost (n9644867@localhost) by cc.wwu.edu (SMI-8.6/SMI-SVR4) with SMTP id OAA09412 for ; Tue, 30 Jun 1998 14:41:04 -0700 Date: Tue, 30 Jun 1998 14:41:04 -0700 (PDT) From: Patrick Brian Kohl Reply-To: Patrick Brian Kohl To: ccl Subject: ccl: Data visualization software for MD sims Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII My group is running a few molecular dynamics sims involving a few hundred molecules each, and we need to invest in some data visualization software to allow us to display images and animations of our sims. Something that would accept as input the coordinates of the atoms involved would be preferable, as would be a program that is easily modified. Has anyone had any good/bad experiences with situations like this? Is there any notable software that people have found especially useful for data viz? Thanks. From shenkin@still3.chem.columbia.edu Wed Jul 1 16:46:03 1998 Received: from mailrelay1.cc.columbia.edu (cu41754@mailrelay1.cc.columbia.edu [128.59.35.143]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id QAA24188 Wed, 1 Jul 1998 16:46:02 -0400 (EDT) Received: from still3.chem.columbia.edu (still3.chem.columbia.edu [128.59.112.36]) by mailrelay1.cc.columbia.edu (8.8.5/8.8.5) with SMTP id QAA04943 for <@smtp.columbia.edu:chemistry@www.ccl.net>; Wed, 1 Jul 1998 16:46:01 -0400 (EDT) Received: by still3.chem.columbia.edu (950413.SGI.8.6.12/930416.SGI.AUTO) for chemistry@www.ccl.net id QAA10316; Wed, 1 Jul 1998 16:46:01 -0400 From: "Peter Shenkin" Message-Id: <9807011646.ZM10314@still3.chem.columbia.edu> Date: Wed, 1 Jul 1998 16:46:01 -0400 In-Reply-To: "Gustavo A. Mercier Jr" "CCL:Summary: Fortran compiler in Linux Box" (Jun 30, 3:03pm) References: <199806301903.PAA16352@mail.med.upenn.edu> X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: chemistry@www.ccl.net Subject: Re: CCL:Summary: Fortran compiler in Linux Box Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii On Jun 30, 3:03pm, Gustavo A. Mercier Jr wrote: > Subject: CCL:Summary: Fortran compiler in Linux Box > Here is the summary of responses to my query about fortran > compilers in Linux boxes... To these I will add that our group has had excellent experience with the Absoft f77 and f90 compilers for LINUX. I believe they also have a Windows version, so if you want to use both OS's, this might be the way to go. It's not cheap (ca. $900, as I recall), but one of my group members who uses the LINUX version is extremely happy. -P. > >From erik@theophys.kth.se Mon Jun 8 02:34:18 1998 > ... if you absolutely need to stick with PC, microsoft are selling > a digital-made fortran 90 compiler for windows. Actually, Digital is selling their own compiler for Windows which replaces MicroSoft PowerStation Fortran. ... >From york@brel.u-strasbg.fr Mon Jun 8 04:10:51 1998 > I would like to point that recently there has become > available a free f90 compiler for Linux from the Pacific-Sierra > Research Corporation. It looks encouraging, for example see: > > http://www.psrv.com/ > > I have no first-hand assessment yet ... -- *********** How can we have ethnic music without ethnic hatred? *********** * Peter S. Shenkin; Chemistry, Columbia U.; 3000 Broadway, Mail Code 3153 * ** NY, NY 10027; shenkin@columbia.edu; (212)854-5143; FAX: 678-9039 *** *MacroModel WWW page: http://www.columbia.edu/cu/chemistry/mmod/mmod.html * From rvenable@deimos.cber.nih.gov Wed Jul 1 19:58:18 1998 Received: from deimos.cber.nih.gov (deimos.cber.nih.gov [128.231.52.2]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id TAA25681 Wed, 1 Jul 1998 19:58:18 -0400 (EDT) Received: from localhost by deimos.cber.nih.gov with SMTP (1.37.109.14/16.2) id AA053857422; Wed, 1 Jul 1998 19:57:02 -0400 Date: Wed, 1 Jul 1998 19:57:02 -0400 (EDT) From: Rick Venable To: "Gustavo A. Mercier Jr" Cc: chemistry@www.ccl.net Subject: Re: CCL:Summary: Fortran compiler in Linux Box In-Reply-To: <199806301903.PAA16352@mail.med.upenn.edu> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Tue, 30 Jun 1998, Gustavo A. Mercier Jr wrote: > Here is the summary of responses to my query about fortran > compilers in Linux boxes... > --- > >From erik@theophys.kth.se Mon Jun 8 02:34:18 1998 > My first recommendation would be to go with another hardware platform Not mine, I can tell you that. The Portland group makes an excellent Fortran compiler for Linux, optimized for PPro and PII machines. AbSoft also makes a decent one, and may have a new version coming out soon. Also, the f2c/gcc combination is really pretty good as well. All did very well on CHARMM benchmarks. > >From erik@theophys.kth.se Mon Jun 8 02:34:18 1998 > Nevertheless, if you absolutely need to stick with PC, microsoft are selling > a digital-made fortran 90 compiler for windows. Strike two. The Microsoft compilers, especially the "Powerstation" series, get routinely roasted in the comp.lang.fortran newsgroup as being buggy as the Maine woods in June, with the typical MS extremely poor customer support. For those afflicted with Windoze, there are at least 4 or 5 other compilers better than anything MS has to offer. (Lahey, Salford, Watcom, AbSoft, ...) -- Rick Venable =====\ |=| "Eschew Obfuscation" FDA/CBER Biophysics Lab |____/ |=| Bethesda, MD U.S.A. | \ / |=| ( Not an official statement or rvenable@deimos.cber.nih.gov | \ / |=| position of the FDA; for that, http://nmr1.cber.nih.gov/ \/ |=| see http://www.fda.gov ) http://www.erols.com/rvenable