From danapple Wed Nov 2 23:21:38 1988 Received: by FLOTSAM.MIT.EDU (5.45/4.8) id AA04085; Wed, 2 Nov 88 23:21:38 EST From: Daniel I. Applebaum Message-Id: <8811030421.AA04085@FLOTSAM.MIT.EDU> To: galatea-bugs Date: Wed, 02 Nov 88 23:21:33 EST ------- Forwarded Message Return-Path: hkbirke Received: by FLOTSAM.MIT.EDU (5.45/4.8) id AA04050; Wed, 2 Nov 88 23:17:20 EST Date: Wed, 2 Nov 88 23:17:20 EST From: Hal Birkeland Message-Id: <8811030417.AA04050@FLOTSAM.MIT.EDU> To: galatea-bugs vdp-50s and gsearch... if one is GSearching on a vdp-50, occasionally it seems to do the following - - this probably is the same problem diagnosed earlier... but this is more common and should be fixed. (ie the Switch needs an artificial delay after in in the driver - --hal (I would be willing to fix this if Dan wants) ------- End of Forwarded Message From hkbirke@MOVIES.MIT.EDU Sun Jan 15 23:18:59 1989 Received: by FLOTSAM.MIT.EDU (5.45/4.8) id AA04087; Sun, 15 Jan 89 23:18:59 EST Received: by MOVIES.MIT.EDU (5.45/4.8) id AA00821; Sun, 15 Jan 89 23:20:19 EST From: Hal Birkeland Message-Id: <8901160420.AA00821@MOVIES.MIT.EDU> To: galatea-bugs@FLOTSAM.MIT.EDU Subject: reverse playsegs on SONY LDP1000A Date: Sun, 15 Jan 89 23:20:17 EST don't work... I try doing a 1 frame reverse (speed = -30) playseg on a Sony and it don't do squat! it works on a vdp50 tho. hmm . --hal From danapple Sun Jan 15 23:28:02 1989 Received: by FLOTSAM.MIT.EDU (5.45/4.8) id AA04127; Sun, 15 Jan 89 23:28:02 EST From: Daniel I. Applebaum Message-Id: <8901160428.AA04127@FLOTSAM.MIT.EDU> To: Hal Birkeland Cc: galatea-bugs@FLOTSAM.MIT.EDU Subject: Re: reverse playsegs on SONY LDP1000A In-Reply-To: Your message of Sun, 15 Jan 89 23:20:17 EST. <8901160420.AA00821@MOVIES.MIT.EDU> Date: Sun, 15 Jan 89 23:27:59 EST Hardware limitation. I think an error message is returned though. The alternative is to have the Sony simply to a reverse play and figure how long in microseconds the section should be, and stop it at the appropriate place. This wouldn't be frame accurate, but might be acceptable. Sound good to you? Dan. From hkbirke@MOVIES.MIT.EDU Mon Jan 16 00:29:22 1989 Received: by FLOTSAM.MIT.EDU (5.45/4.8) id AA04166; Mon, 16 Jan 89 00:29:22 EST Received: by MOVIES.MIT.EDU (5.45/4.8) id AA00860; Mon, 16 Jan 89 00:30:41 EST From: Hal Birkeland Message-Id: <8901160530.AA00860@MOVIES.MIT.EDU> To: Daniel I. Applebaum Cc: Hal Birkeland , galatea-bugs@FLOTSAM.MIT.EDU, hkbirke@MOVIES.MIT.EDU Subject: Re: reverse playsegs on SONY LDP1000A In-Reply-To: Your message of Sun, 15 Jan 89 23:27:59 EST. <8901160428.AA04127@FLOTSAM.MIT.EDU> Date: Mon, 16 Jan 89 00:30:39 EST having NO playsegs on a SONY is a BAD idea... it becomes too difficult for the client application. the reverse guess sounds cool... I am using it to "view" frame-by-frame for editting... --hal From danapple Mon Jan 16 01:13:25 1989 Received: by FLOTSAM.MIT.EDU (5.45/4.8) id AA04243; Mon, 16 Jan 89 01:13:25 EST From: Daniel I. Applebaum Message-Id: <8901160613.AA04243@FLOTSAM.MIT.EDU> To: Hal Birkeland Cc: Daniel I. Applebaum , galatea-bugs@FLOTSAM.MIT.EDU Subject: Re: reverse playsegs on SONY LDP1000A In-Reply-To: Your message of Mon, 16 Jan 89 00:30:39 EST. <8901160530.AA00860@MOVIES.MIT.EDU> Date: Mon, 16 Jan 89 01:13:23 EST I think that the thing to do is build the extra timer into the sony driver. The only problem I foresee is dealing with RETURN_NOW playseg's To do this thing with the timers will require the program to hang around in the driver until the segment is done and interupting is tough. One possibly good idea is to somehow "program" the sony to play backwards for a certain amount of time. I don't know if sony programs are interuptable. I'll check tomorrow. Dan. From danapple Wed Feb 22 18:23:19 1989 Received: by FLOTSAM.MIT.EDU (5.45/4.8) id AA08094; Wed, 22 Feb 89 18:23:19 EST From: Daniel I. Applebaum Message-Id: <8902222323.AA08094@FLOTSAM.MIT.EDU> To: Hal Birkeland Cc: galatea-bugs@FLOTSAM.MIT.EDU, time-out@MOVIES.MIT.EDU, danapple In-Reply-To: Your message of Mon, 20 Feb 89 19:57:21 EST. <8902210057.AA03652@MOVIES.MIT.EDU> Date: Wed, 22 Feb 89 18:23:15 EST I'd like to see a CLV player that can do frame accurate searches. Otherwise, it wouldn't fit the model, anyway. This timeout will be increased to 50 mins. Note that sites can customize galatead's throught the use of the tunables.h file in the SERVER directory. Dan. From drmorris@ATHENA.MIT.EDU Thu Mar 2 22:25:04 1989 Received: by FLOTSAM.MIT.EDU (5.45/4.8) id AA03163; Thu, 2 Mar 89 22:25:04 EST From: drmorris@ATHENA.MIT.EDU Received: by ATHENA.MIT.EDU (5.45/4.7) id AA23029; Thu, 2 Mar 89 22:24:59 EST Received: by M11-115-4.MIT.EDU (5.45/4.7) id AA11616; Thu, 2 Mar 89 22:24:35 EST Message-Id: <8903030324.AA11616@M11-115-4.MIT.EDU> To: galatea-bugs@FLOTSAM.MIT.EDU Subject: cannot get gctl to work Date: Thu, 02 Mar 89 22:24:30 EST Hello! Many users and consultant have been having problems using gctl. I tried to use it an couldn't get it work under any cirsumstance. On M11-113-3, gctl returned: could not open galatea server on M11-113-3 % gshutdown [3] 21043 % gstart [4] 21049 % gctl & % could not open galatea server on M11-113-3 I tried these remidies in the Athenadoc guide and that did not help. I tried it as root and from another workstation. To no avail. Other consultants have not been able to make it work as well. Could you please look at Section 2.1 of the Video Workstation guide and make sure that gctl runs as stated. Thank you. Athena Consultant Dave Morrison p.s. Here is information on how to read the Video Workstation guide on-line. From an Athena workstation: attach athenadoc cd /mit/athenadoc viewdoc video.workstation.PS & After you type the 'viewdoc' command, move the mouse to the upper left hand corner of your screen, and then press the left mouse button. Inside the document previewer, you can go forward a page, back a page, or exit. You do this either by clicking a mouse button on the appropriate box at the upper left of the window, or by pressing: Left Button: previous page Middle Button: exit (that's both buttons simultaneously on a two-button mouse) Right Button: next page From danapple Thu Mar 2 22:44:47 1989 Received: by FLOTSAM.MIT.EDU (5.45/4.8) id AA03235; Thu, 2 Mar 89 22:44:47 EST From: Daniel I. Applebaum Message-Id: <8903030344.AA03235@FLOTSAM.MIT.EDU> To: drmorris@ATHENA.MIT.EDU Cc: galatea-bugs Subject: Re: cannot get gctl to work In-Reply-To: Your message of Thu, 02 Mar 89 22:24:30 EST. <8903030324.AA11616@M11-115-4.MIT.EDU> Date: Thu, 02 Mar 89 22:44:44 EST Thanks for the note about the failure of gctl in the fishbowl. I do not have an immediate idea as to the cause of this failure but will look into it early this morning. I did not install Galatea for the fishbowl, so no immediate fix comes to mind. Dan. From danapple Mon Apr 3 11:37:51 1989 Received: by FLOTSAM.MIT.EDU (5.45/4.8) id AA04623; Mon, 3 Apr 89 11:37:51 EDT From: Daniel I. Applebaum Message-Id: <8904031537.AA04623@FLOTSAM.MIT.EDU> To: paul@athena Cc: jnrees@athena, hkbirke, galatea-bugs Subject: Galatea and clean_tmp_areas interaction Date: Mon, 03 Apr 89 10:37:46 EST Look, I hate clean_tmp_areas on a normal Athena workstation, which is why I don't have flotsam run it. See, it keeps blowing away the UNIX domain socket for galatea. So, as a simple possible solution, I suggest starting galatead when the workstation activates, and kill it when the workstation deactivates. I'd blow away the current /etc/rc.local on the fishbowl workstations, and replace /etc/athena/{de,}activate.local with /mit/danapple/FISHBOWL/{de,}activate.local I haven't tried this solution on any machine, but maybe this'll fix things. Dan. From danapple Tue Apr 4 10:39:40 1989 Received: by FLOTSAM.MIT.EDU (5.45/4.8) id AA07226; Tue, 4 Apr 89 10:39:40 EDT From: Daniel I. Applebaum Message-Id: <8904041439.AA07226@FLOTSAM.MIT.EDU> To: paul@athena, jnrees@athena, galatea-bugs, hkbirke Subject: Correction on G<->clean_temp_areas interaction Date: Tue, 04 Apr 89 09:39:36 EST Win Treese pointed out that clean_tmp_areas does not destroy sockets, so I had to find another victim for my flames and found a: /bin/rm -rf /tmp in /etc/athena/deactivate This may be what causes Galatea's UNIX socket to be destroyed. The solution is same, which is to have galatead started in /etc/athena/activate.local and killed (with a SIGTERM) in /etc/athena/deactivate.local Machines like flotsam and movies never had a problem with the UNIX socket being destroyed because they don't run toehold. Dan. From treese@crl.dec.com Tue Apr 4 11:48:45 1989 Received: by FLOTSAM.MIT.EDU (5.45/4.8) id AA07323; Tue, 4 Apr 89 11:48:45 EDT Received: from crl.dec.com by decwrl.dec.com (5.54.5/4.7.34) id AA10228; Tue, 4 Apr 89 08:48:08 PDT Received: from crl.dec.com by decwrl.dec.com (5.54.5/4.7.34) for hkbirke@flotsam.mit.edu; id AA10228; Tue, 4 Apr 89 08:48:08 PDT Received: by crl.crl.dec.com (5.57/Ultrix2.4-C) id AA03958; Tue, 4 Apr 89 11:45:54 EDT Message-Id: <8904041546.AA01249@crlwin.DEC.COM> To: danapple@flotsam.mit.edu (Daniel I. Applebaum 4-Apr-89 0939 EST) Cc: paul@athena.mit.edu, jnrees@athena.mit.edu, galatea-bugs@flotsam.mit.edu, hkbirke@flotsam.mit.edu Subject: Re: Correction on G<->clean_temp_areas interaction In-Reply-To: Your message of Tue, 04 Apr 89 10:39:25 -0400. Date: Tue, 04 Apr 89 11:45:58 EDT From: Win Treese Actually, I'm not surprised that the problem exists; I was making a statement about what was supposed to be the case. Anyway, it seems to me that one of three configuration methods should be chosen: 1) galatead runs all the time, and the standard workstation configuration is modified to handle its socket (actually, sockets in the general case) politely. 2) galatead be modified to put its socket somewhere else (e.g., /usr/tmp), which does not get destroyed at deactivation time 3) Modify the standard rc.conf to contain a VIDEO or GALATEA option. Starting galatea in activate and killing in deactivate should be based on this variable. I think I prefer (2), although the rc.conf part of (3) should also be done, if it hasn't been already. - Win From danapple Tue Apr 4 12:46:58 1989 Received: by FLOTSAM.MIT.EDU (5.45/4.8) id AA07412; Tue, 4 Apr 89 12:46:58 EDT From: Daniel I. Applebaum Message-Id: <8904041646.AA07412@FLOTSAM.MIT.EDU> To: Win Treese Cc: paul@athena.mit.edu, jnrees@athena.mit.edu, galatea-bugs@flotsam.mit.edu, hkbirke@flotsam.mit.edu Subject: Re: Correction on G<->clean_temp_areas interaction In-Reply-To: Your message of Tue, 04 Apr 89 11:45:58 EDT. <8904041546.AA01249@crlwin.DEC.COM> Date: Tue, 04 Apr 89 11:46:55 EST Galatea could be modified to put the UNIX socket in /usr/tmp, but that really doesn't provide a good solution, because in the future, some bozo may decide that /usr/tmp should also be cleaned on every machine deactivation. It seems that /etc/athena/deactivate should be set up the same way as clean_temp_areas so that it uses a find(1) and specifically does not remove sockets. (Actually, flotsam ran clean_temp_areas for a long time and never had a problem, so I believe clean_temp_areas works correctly, in this case) The rc.conf idea sounds good to me. Perhaps galatead's start method can simply be toggled on the TOEHOLD variable. That already indicates the toehold status of the workstation. Modifying /etc/rc and /etc/athena/{de,}activate for this variable is fairly simple. Dan. From hkbirke@MOVIES.MIT.EDU Thu Apr 6 20:34:32 1989 Received: by FLOTSAM.MIT.EDU (5.45/4.8) id AA13678; Thu, 6 Apr 89 20:34:32 EDT Received: by MOVIES.MIT.EDU (5.45/4.8) id AA01517; Thu, 6 Apr 89 20:34:42 EDT Date: Thu, 6 Apr 89 20:34:42 EDT From: Hal Birkeland Message-Id: <8904070034.AA01517@MOVIES.MIT.EDU> To: galatea-bugs@FLOTSAM.MIT.EDU forkload: the current forkload seems to "finish" before the VDP50's are all spun up. The first few commands to the players (liek 1 or 2) make be lost because the player has yet to be fully spun up. From danapple Thu Mar 1 11:07:32 1990 Received: by FLOTSAM.MIT.EDU (5.61/4.8) id AA17488; Thu, 1 Mar 90 11:07:32 -0500 From: Daniel I. Applebaum Message-Id: <9003011607.AA17488@FLOTSAM.MIT.EDU> To: crafts@ATHENA.MIT.EDU Cc: galatea-bugs In-Reply-To: Your message of Wed, 28 Feb 90 03:53:12 -0500. <9002280853.AA00721@M11-113-5.MIT.EDU> Date: Thu, 01 Mar 90 11:07:30 EST Have you tried olc or on-line help yet? Those two sources of information will tell you more than I can about the particular setup in the fishbowl. You could try the consultants office, if open, and see if they're willing (or obligated, at least) to help with problems in the fishbowl. Things you can check: Is there really a disk in the player? Is the videodisk player connected to the workstation? Is the Galatea server running? If the server is running, it may be best to stop it (gshutdown), and then restart it (gstart). There's a good chance that the player is broken, in which case it should be reported to the Athena hardware hotline (hotline@athena). Dan. From danapple Sat Mar 17 12:53:34 1990 Received: by flotsam.MIT.EDU (5.61/4.8) id AA08173; Sat, 17 Mar 90 12:53:34 -0500 From: Daniel I. Applebaum Message-Id: <9003171753.AA08173@flotsam.MIT.EDU> To: dewey@ATHENA.MIT.EDU Cc: galatea-bugs Subject: Re: galatea In-Reply-To: Your message of Sat, 17 Mar 90 12:16:04 -0500. <9003171716.AA02695@E40-008-8.MIT.EDU> Date: Sat, 17 Mar 90 12:53:33 EST Sorry for not answering your first message, of last week. As I'm not sure at what stage you are getting the error, it is hard for me to give you any help. Have you tried starting with 'gstart', then 'gctl'? If you can send me a transcript of the session, that would be more helpful. The help screen is out of date, especially since a complete new version of Galatea is on the 'video' locker now. I don't know when all of the machines in the fishbowl will be updated. Dan. From danapple@flotsam.MIT.EDU Mon Apr 2 14:20:32 1990 Received: from ATHENA.MIT.EDU by flotsam.MIT.EDU (5.61/4.8) id AA29036; Mon, 2 Apr 90 14:20:32 -0400 Received: from FLOTSAM.MIT.EDU by ATHENA.MIT.EDU with SMTP id AA10291; Mon, 2 Apr 90 14:20:30 EDT Received: by flotsam.MIT.EDU (5.61/4.8) id AA29032; Mon, 2 Apr 90 14:20:27 -0400 Date: Mon, 2 Apr 90 14:20:27 -0400 From: Daniel I. Applebaum Message-Id: <9004021820.AA29032@flotsam.MIT.EDU> To: galatea-bugs@ATHENA.MIT.EDU Subject: bug test This is bug test. From @IBM.COM:drake@almaden Fri Apr 13 16:07:28 1990 Received: from ATHENA.MIT.EDU by flotsam.MIT.EDU (5.61/4.8) id AA01863; Fri, 13 Apr 90 16:07:28 -0400 Received: from ibm.com by ATHENA.MIT.EDU with SMTP id AA28992; Fri, 13 Apr 90 16:07:25 EDT Received: from almaden.ibm.com by IBM.COM (IBM VM SMTP R1.2.1MX) with BSMTP id 0633; Fri, 13 Apr 90 13:08:39 PDT Received: by sd2.almaden.ibm.com (AIX 2.1 2/4.03) id AA05663; Fri, 13 Apr 90 13:04:23 PDT Message-Id: <9004132004.AA05663@sd2.almaden.ibm.com> To: galatea-bugs@ATHENA.MIT.EDU Subject: Galatea and scope of declarations Date: Fri, 13 Apr 90 13:04:07 -0800 From: "Sam Drake" Hi, I'm trying to port Galatea to the IBM Risc System/6000 for my own use. It seems that Galatea and the C compiler on this system disagree about the scope of "extern" declarations. There are many cases in Galatea where the software assumes that "extern" declarations have file scope; the variable is declared within the body of the first function in a file, then used without redeclaration in the bodies of other functions in the same file. I (and the C compiler) believe this is illegal; either the declarations should be outside the body of any function or they should be declared in the body of every function that uses them. One example of this is the use of "allavail" in deal.c ... it's declared in the body of one function, but used in another. Moving the declaration (actually ALL the "extern" declarations) outside of the scope of all functions seems to resolve the problem. No big deal, but I thought I'd mention it! Cheers, ...Sam From danapple@flotsam.MIT.EDU Sat Apr 14 12:39:54 1990 Received: from ATHENA.MIT.EDU by flotsam.MIT.EDU (5.61/4.8) id AA02549; Sat, 14 Apr 90 12:39:54 -0400 Received: from FLOTSAM.MIT.EDU by ATHENA.MIT.EDU with SMTP id AA15937; Sat, 14 Apr 90 12:39:52 EDT Received: by flotsam.MIT.EDU (5.61/4.8) id AA02477; Sat, 14 Apr 90 12:37:18 -0400 From: Daniel I. Applebaum Message-Id: <9004141637.AA02477@flotsam.MIT.EDU> To: "Sam Drake" Cc: galatea-bugs@ATHENA.MIT.EDU Subject: Re: Galatea and scope of declarations In-Reply-To: Your message of Fri, 13 Apr 90 13:04:07 -0800. <9004132004.AA05663@sd2.almaden.ibm.com> Date: Sat, 14 Apr 90 12:37:15 EDT Thanks for mentioning it. I'll go through the server and try to catch all those that I can find. If you happen to have a list of errors when you first tried to compile, that would be helpful, since my compiler doesn't seem to catch those problems. Are you using an ANSI C compiler? I'm trying lint(1) right now, maybe that'll help me. Take it easy, Dan. From roy@willow.ccsf.caltech.edu Fri Jun 1 15:06:30 1990 Received: from willow.ccsf.caltech.edu by flotsam.MIT.EDU (5.61/4.8) id AA29115; Fri, 1 Jun 90 15:06:30 -0400 Received: by willow.ccsf.caltech.edu (4.0/1.2) id AA11648; Fri, 1 Jun 90 12:04:41 PDT Date: Fri, 1 Jun 90 12:04:41 PDT From: roy@willow.ccsf.caltech.edu (Dr. Roy) Message-Id: <9006011904.AA11648@willow.ccsf.caltech.edu> To: galatea-bugs@flotsam.mit.edu Subject: Thanx ! Dear Dr/Mr/Ms Galatea, Thank you for producing and distributing galatea. Its the flexible and useful link from our Sun net to our Panasonic optical disk that has allowed us to make videos of scientific visualizations while thinking about science and not about the video process. There seems to be a slight problem which I'd like some help with. Every time I eject the optical disk, either from the gctl tool or manually, the galatea server disappears. When I try to start it again, I get a message about broken sockets and how UNIX hasn't figured out that galatea has gone. So the only way to restart galatea is to reboot the server, the result being that changing the optical disk takes 15 minutes and everyone using the server remotely gets rather pissed. I'm a bit new to the position of video daemon, and galatea was installed by the last daemon, so I don't really know what I'm doing with it. Thank you Roy Williams Caltech Concurrent Computation Project From danapple Fri Jun 1 15:45:22 1990 Received: by flotsam.MIT.EDU (5.61/4.8) id AA29152; Fri, 1 Jun 90 15:45:22 -0400 From: Daniel I. Applebaum Message-Id: <9006011945.AA29152@flotsam.MIT.EDU> To: roy@willow.ccsf.caltech.edu (Dr. Roy) Cc: galatea-bugs@flotsam.mit.edu Subject: Re: Thanx ! In-Reply-To: Your message of Fri, 01 Jun 90 12:04:41 -0700. <9006011904.AA11648@willow.ccsf.caltech.edu> Date: Fri, 01 Jun 90 15:45:20 EDT I remember that at one point there was a bug with Galatea which would cause the problem you described. You may wish to grab a new copy of Galatea from flotsam.mit.edu Use anonymous ftp. Also, what version are you using now? I'm not sure if Rod ever grabbed the 2.4 code, which is the latest released copy. Dan.