.PS 60,72 .LM 9 .RM 70 .NHY .LITERAL #####August Vol. 1 No. 3 1979 .end literal .s 2 .s 2 .s 2 .s 2 .s 2 .s 2 .s 2 .s 2 .s 2 .s 2 .s 2 .s 2 Editor's Comments .s 1 Here is the third issue of the Page Swapper. The Spring Symposium is well behind us and the Fall Symposium is beginning to take form. It is summer and we are all lethargic, at least most of our contributors are lethargic. Once again we have very few outside contributions to this, the third newsletter. I am a novice at this startup activity and therefore not certain as to whether or not this lack of response in to be expected. I do hope that the Page Swapper will provide the vehicle for the exchange of ideas, problems, solutions etc. available no other place. This can only occur if we all take the time to contribute things we have found useful. .s 2 The following paragraphs are abstracted from my trip report on the Spring Symposium and are included as one person's view of how this symposium was conducted. .s 1 .f.j.s 2 I started the symposium by attending the VAX SIG steering committee business meeting on Monday afternoon. We decided that a major goal throughout the symposium should be to foster the interaction between VAX/VMS users and DEC with regards to major problem areas and to get committments from DEC to work on these problem areas. The SIG is also setting up a new activity centered about a new form entitled the System Improvement Request. This form will be filled out by members of the SIG, rated in order of importance by the entire SIG membership, and presented to DEC to guide the development and enhancement of VAX/VMS. .s 1 In the opening VAX SIG session Tuesday, I was impressed at the quantity of VAX machines represented. Of approximately 460 attendees, 40% indicated that they represented a site with a VAX installed. There was no VAX at the Symposium (they had a 2020 instead---the rationale for which totally escaped me) which was a disappointment for about half the attendees. The VMS developers were particularly disappointed since they could not set up and duplicate problems without a machine. .s 1 In the hardware session diagnostics were a big issue. The major point was the difference between version numbers on our floppy disks and those on the field service floppies. It was explained that floppies are updated every two months and it is the field service engineer's responsibility to copy his new floppies onto our old ones. Another problem area was the incompatibility of the floppies and the machine microcode. We need to watch this area. .s 1 I was invited to participate in a System Managers Panel on Wednesday. In the preliminary meeting for this panel Tom Lawrence of MSA, Pittsburgh described his experiences with the STC 6250/125 ips tape drives. He is quite pleased with them and does a full RP06 copy in about 12 minutes. Mark Pilant wrote the device driver for the tape system. They might elect to market this driver to the general public at some point in the future. They use a Systems Concepts SA-11 controller on their UNIBUS. They have found a problem with the UNIBUS timing and speed. There is a mod to the UNIBUS which is approved by DEC field service to change the timing of the NOSACK signal to allow the SA-11 to work properly on the UBA. MSA feels that the 1.5 MB spec that DEC advertises for the UBA should really be quoted at about 1.1 MB. .s 1 There was an overall concensus that a .5 Mbyte VAX was underconfigured for a time sharing shop with more that 4 concurrent users. .s 1 In a VAX SIG session on system performance and tuning I learned that we can make some changes to improve our performance and space utilization. By using a default cluster size of 5 rather than 11 we can recover as much as half of our disk's usable volume. Also by increasing the 'free page cache' (free list and modified list limits), we can improve the paging performance of our system. [Note: I analyzed our disk file sizes statistically and found that a cluster factor of 3 with extensions of 8 optimized our space recovery.] .s 1 The Symposium was notable for its lack of announcements by DEC in the VAX/VMS marketplace. Release 1.5 of VMS was previewed briefly. Release 2.0 was too far in the future to be discussed. There was no new hardware to be presented. Looks like fall Decus will have a whole pile of new stuff if DEC is true to form. .s 1 We are getting a new version of TECO with support for terminals other than VT-52's and VT-100's. This version of TECO will also support a much more powerful screen editing mode. [This new version of TECO is more than my most optimistic expectations. Also the TECO SIG meeting was one of the most enjoyable sessions I have ever attended at any conference in my life.] .s 1 I represented LRS in a meeting which Dick Hustvedt and Andy Goldstein chaired for VAX system managers to discuss current backup problems and future backup software. The alternatives boiled down to a choice between 1) blindingly fast offline volume backup with equally fast full volume restore and miserably slow partial volume restore, 2) moderately fast offline backup with moderately fast full volume and partial volume restore and 3) intelligent, on-line, file structured, slow backup and restore. The concensus was that alternative three was the most critical. .s 1 Tryg Fossum of DEC presented a session on user microprogramming on the VAX which indicated some great potential in this area for us (except for the cost). It appears that the IBM-VAX number representation conversions could be readily implemented in user microcode. .s 1 Scott Davis of DEC who is responsible for VAX DECNET reported that the enchancements to DECNET for release 2.0 will be substantial including virtual command terminal support. This will implement most of the functionality we find missing in the present verison. .s 1 In the session on field service experiences, our January down time showed up as significant spikes on every chart presented. DEC is very pleased with remote diagnostics and is pushing it hard. Every user I talked to who has remote diagnostics support is extremely pleased with the responsiveness and the quality of the service. .s 1 To incorporate our devices into the error logging system will not be extremely difficult, but getting out nicely formatted error reports will be either very difficult, very costly, or impossible. To interface with SYE the error log report writer, requires a VMS source license and a BLISS-32 compiler. These two items are well over $20,000. .s 1 DEC's Andy Goldstein presented a very useful session on Ancillary Control Processes (ACP). We no longer need one but if we did we would definitely need more information on how to develop it and interface it to VMS. In this session Andy explained the reason why VMS is stopped by the online rewind of a tape drives. To bowdlerize Andy, the device driver and the controller are in a contest to handle the condition caused by the change of state interrupt. The resultant overload of processing at high IPL stop all background activities. This will be fixed in either VMS 1.5 or 2.0. .s 2 .LIT To: VAX SIG From: John Thompson, VAX SIG Librarian Intermetrics, Suite 415 4733 Bethesda Ave. Bethesda, Md. 20014 Date: 7 May 1979 Subject: Spring 79 DECUS VAX Symposium Tape .END LITERAL .b .p The Spring 79 DECUS VAX Symposium Tape should be available for distribution about the end of May. .br .p The names, regional areas, and telephone numbers of the persons to contact for copies of the tape are: .br .lm 16 .b Roger Vossler, Los Angeles, (213) 535-0312 .br Dennis Perry, New Mexico, (505) 667-7193 .br Don Golden, Texas, (713) 781-8871 .br Rich Greco, Oregon, (503) 244-6161 .br Al Siegel, Ohio, (614) 424-4664 .br Tom Lawrence, Pennsylvania, (412) 683-9533 .br Tom Provost, Massachusetts, (617) 774-2370 .br Bob Havlin, Missouri, (314) 658-7100 .br Art McClinton, Virginia, (703) 827-6358 .br .b To submit items for the FALL 79 TAPE please contact: .br .b John Thompson, Washington DC, (301) 657-3775 .br .lm 8 .b I suggest you call your nearest source and make arrangements with him for getting a copy of his tape. .p Now a word about documentation. With your submissions of programs you must submit some kind of machine readable documentation. Name the file "program-name.HLP" to facilitate identification. Include the following information please. .b .literal Name of program Brief program description Responsible person Date Revision level of program Names of all modules (source, obj, exe, dat, etc.) How to compile the sources (submit *.com files maybe) How to link the program (again maybe *.com files) Instructions for using the program Other appropriate information as you see fit .end literal .br .s 2 .tp 6 .c 72; SPECIAL SYSTEMS SOFTWARE WORKING GROUP FORMED .S 2 At the New Orleans meeting a Working Group was formed to coordinate information on special (non-DEC) operating systems and/or user interfaces to operating systems. This includes, but is not limited to, the Bell Labs UNIX operating system and the Software Tools Subsystem which is a very "friendly" user interface to VMS. SSSWG hopes to sponsor tutorials and in depth sessions at the San Diego meeting. Anyone wishing to present a paper in these areas, or having software they wish to distribute, should contact one of the co-charimen: .lm 25 .s 1 .literal Henry Camp Joe Sventek Box 48088 Bldg 50 Room 3224 Atlanta, GA 30362 Lawrence Berkley Labs 404-894-3658 Berkley, CA 94720 415-486-5181 .end literal .tp 10 .s 2 .lm 9 .literal TO: VAX SIG FROM: Ross W. Miller, President ONLINE DATA PROCESSING, INC. N. 637 Hamilton, Spokane, WA 99202, 509-484-3400 DATE: 31-May-79 SUBJECT: FILE SHARING UNDER COMPATIBILITY MODE PROGRAMS .end literal .s 1 We have found quite a few people in desperate need of the facility for file sharing of compatibility mode programs, i.e., Basic Plus II programs doing business applications, therefore, we are writing to let you know that we have developed such a product. This product allows a near perfect RSTS environment on the VAX computer which includes the ability to share files. .s 1 If you are interested, please contact me at the above address and/or phone. .s 1 In addition, for those of you who are having trouble managing your system because of an inordinate number of files building up on your VAX system due to version levels being kept around, the following command file may prove very useful. .s 1 .nf.nj ! ****** FILEPURG.COM - PURGES ENTIRE SYSTEM ********** $ INQUIRE DEL_FILE "FILE TO DELETE FROM ALL ACCOUNTS?" ! ASK FOR FILES TO DELETE $ DEL_YES = 'F$LENGTH(DEL_FILE) ! SET DELETE FLAG $ SAVE_DIR := 'F$DIRECTORY() ! STORE CURRENT DEFAULT DIR $ ON CONTROL_Y THEN GOTO INTERRUPT ! CLEAN UP IF INTERRUPT $ DIR [0,0]/OUTPUT=DIR.KEY ! WRITE MFD TO FILE $ OPEN/READ IN DIR.KEY ! REOPEN FILE FOR READING $ START: ! START OF PURGE LOOP $ READ/END_OF_FILE=FINISHED IN DIR_RECORD ! GET RECORD (GOTO FINISHED ON EOF) $ DOT = 'F$LOCATE(".",DIR_RECORD) ! LOOK FOR DOT IN RECORD $ PURGE_DIR := 'F$EXTRACT(DOT,4,DIR_RECORD) ! GET FILE EXTEN $ IF PURGE_DIR .NES. "DIR" THEN GOTO START ! DIR FILE NAME? $ DOT=DOT-2 ! SET FOR LENGTH OF DIR NAME $ PURGE_DIR := 'F$EXTRACT (2,DOT,DIR_RECORD) ! GET DIR NAME $ PURGE_DIR := ['PURGE_DIR'] ! ADD BRACKETS $ SET DEF 'PURGE_DIR' ! SET DEFAULT $ PURGE/L ! PURGE DIRECTORY $ IF DEL_YES .GT. 0 THEN DELETE 'PURGE_DIR' 'DEL_FILE'/L $ GOTO START $ FINISHED: ! DONE $ CLOSE IN ! CLOSE FILE $ INTERRUPT: $ SET DEF 'SAVE_DIR" ! RESET TO ORIGINAL DEFAULT $ DELETE DIR.KEY;*/L ! DELETE TEMPORARY FILE $ EXIT .F.J.S 2 [Editor's note: I want to echo Ross's comments about needs for commercial software for the VAX. I have received several inquiries as to where commercial software is being developed. Those of you who have an interest in commercial software for the VAX can address letters to the SIG membership and get them into these very pages. So long as there are no advertising claims and prices in the letters, letters from vendors describing packages will be published also.] .s 2 .tp 8 .c 72; COBOL SIG INFORMATION .s 1 The COBOL SIG discovered in New Orleans that the VAX users thought we are a COBOL-11 Special Interest Group. WE ARE NOT! The COBOL SIG has two goals: 1) to share knowledge about COBOL among users and 2) to provide a strong common voice to Digital on improvements to COBOL products. These goals apply as much to COBOL-74 as they do to COBOL-11. The COBOL SIG is structured to accomplish these goals through a high level of user interaction. We are building a COBOL library tape, preparing prioritized wish lists for Digital, disiminating information on known COBOL bugs, and generally trying to help each other to get the most from our COBOL systems. If you would like to become a member of the COBOL Special Interest Group, simply notify the DECUS office (the address is on the backpage of the newsletter). For more information, contact: .s 1 .nf.nj ################Frank Anderson ################COBOL SIG Chairman ################Grand Canyon College ################3300 West Camelback Road ################Phoenix, AZ 85017 .s 2 .F.J VAX/VMS SPR's of interest: .s 1 [Editor's note- In this column we wish to present SPR's that are useful, informative, interesting, aggravating... The purpose is to share common problems and solutions. This column is also designed to fill in the gaps left by the absence of a VAX/VMS Software Dispatch. Another purpose is to disseminate DEC's SPR responses which solve problems for many users. The final purpose is to get real solutions to SPR's which DEC has closed without solving. Send me copies of SPR's (and responses if approporiate) and I'll try to put the essence into this column. Please be aware that the folks who respond to SPR's are willing to review their responses when we, the users, feel that they are inadequate. The mechanism to activate this review is a second SPR. This SPR should reference the number of the original SPR, redefine the problem, indicate that the solution presented in the initial response is inadequate, and give details as to why this solution is inadequate.] .s 1 ------------------------------------------------ Under VMS 1.01 and VMS 1.05, turning a DMC-11 on and then off again causes the user buffered I/O quota to be incremented. .s 1 RESPONSE .s 1 The XMDRIVER preallocates receive buffers when the device is started and deallocates them when the device is shutdown. A bug in the shutdown code causes more bytes to be returned to the user than were originally charged when the device was started. This will be fixed in release 1.6. .s 1 ------------------------------------------------ The DELETE/AFTER command does not work for ISAM files (in this case created by Datatrieve). You get a message telling you that the file-structure is unsupported. ( Which it is in native mode RMS, but still you can have those files, and definitely have if you use Datatrieve). .s 1 RESPONSE .s 1 This will be corrected in the next autopatch kit which has ISAM support in RMS-32. .s 1 --------------------------------------------------- When opening an existing file for direct access in FORTRAN, and not specifying the record size, the actual record size is not always correctly determined. .s 1 RESPONSE .s 1 Thanks for finding the problem. There is an inconsistency in the way that the Run Time Library computes the record size. This will be fixed in VMS Release 2.0. As a temporary workaround, always specify the record size when opening a file. .s 1 --------------------------------------------------- Magtape problem .s 1 If a batch job is stopped (i. e. DELETE/ENTRY=??? SYS$BATCH) while it is waiting fir a second volume to be mounted, the process will not disappear until OPCOM gets its reply from the operator, AND THE MAGTAPE IS ONLINE. The latter is the problem if the reason for stopping the job is a malfunctioning magtape unit. .s 1 --------------------------------------------------- You cannot create a directory with protection specified as "NO ACCESS" for some class of user, e. g. WORLD. .s 1 RESPONSE .s 1 The CREATE/DIRECTORY command can be used to create a directory with the "NO ACCESS" attribute for a given class of user but not as documented in the VAX/VMS Command Language User's Guide. What is indicated in the User's Guide via the reference to the /PROTECTION qualifier in the SET PROTECTION command would logically be expected to work and the software will be changed to allow this syntax in a future release. Until then the desired effect can be had by completely omitting the class of user from the /PROTECTION qualifier as: .s 1 $ CREATE /DIRECTORY [JUNK.JUNK]/PROT=(S:RWED,G:RWED,O:RWED) .S 1 To give the "NO ACCESS" attribute to the world. .s 1 It is important to note that if the /PROTECTION qualifier is used, every class which you desire to have access must be included or it will be defaulted to "NO ACCESS". .s 1 --------------------------------------------------- Problem with setting terminal spooled: .s 1 $ SET TTC6:/SPOOLED=(TTC6) .s 0 gives error device not in system because SYS$DISK translates to UDEV: (process) and UDEV: to DBA1: (SYSTEM). .s 1 $ SET TTC6:/SPOOLED=(TTC6:,DBA0:) .s 0 works. .s 1 --------------------------------------------------- .tp 10 .s 2 Unsolicited Random Input .s 1 [Editor's note: In this section we will present interesting trivia and helpful facts about VAX/VMS. The sources will be unnamed at present to protect the innocent.] .s 2 The substance of the following paper was presented at the Fall 78 DECUS symposium in San Francisco. It is reprinted here with the permission of the author and Digital Equipment Corporation.