Sun-Spots-Request@RICE.EDU (William LeFebvre) (05/18/88)
SUN-SPOTS DIGEST Tuesday, 17 May 1988 Volume 6 : Issue 89 Today's Topics: Re: YPbind still trying... Re: Sun4/110 memory information Re: When suntools rereads the root menu Re: SLIP on Sun-OS 4.0 - don't hold your breath Re: Cisco terminal servers and SLIP MAXUSERS > 12 in SunOS 3.2 and 3.4 SunOS 3.2 MAXUSER>12 bug fix Official patch #2 for calctool v2.2. Please apply it. another SunOS 3.2 bug Suns lose track of the console I need a caps lock indicator Help: debugging a program that needs mouse input? Wanted: Info on 'IdxTeX', automated index generation for LaTeX reliable and customizable Franz Lisp on Sun3? Send contributions to: sun-spots@rice.edu Send subscription add/delete requests to: sun-spots-request@rice.edu Bitnet readers can subscribe directly with the CMS command: TELL LISTSERV AT RICE SUBSCRIBE SUNSPOTS My Full Name Recent backissues are available via anonymous FTP from "titan.rice.edu". For volume X, issue Y, "get sun-spots/vXnY". They are also accessible through the archive server: mail the request "send sun-spots vXnY" to "archive-server@rice.edu" or mail the word "help" to the same address for more information. ---------------------------------------------------------------------- Date: Thu, 12 May 88 11:11:28 PDT From: jimc@math.ucla.edu Subject: Re: YPbind still trying... Reference: v6n66 Running SunOS v3.4.2, we have the same problem. <randy@ncifcrf.gov> reports a solution in Sunspots v6#75 (grab the 3.5 ypbind, ypserv, etc.). If you don't want to do that, I found that if you do /usr/etc/yp/ypset <server>, and then scurry around killing ypbind processes before they multiply, you often can get the machine back. Restart ypbind after all the lemmings are dead. Note, <server> must be a NUMERIC address, since a YP call is necessary to resolve a site name! On a voluntary reboot of a YP server, or when I have to restart ypserv, I go around to each client and ypset to the other server. I find, for some reason, that the clients generally settle on the slave server after a few days, and even the master server sometimes binds to the slave. Perhaps it's because of a core/CPU hog that (for reasons too complicated to explain) only runs on the YP master. The clients get tired of waiting for ypserv to swap in on the master, and change over to the slave. -- jimc ------------------------------ Date: Thu, 12 May 88 12:37:27 PDT From: Doug Moran <moran@ai.sri.com> Subject: Re: Sun4/110 memory information Reference: v6n60 > The Sun4/110 is user expandable to 32 Meg. is not quite accurate. The Sun-4/110 has slots for 32 SIMMs. IF 1 MB SIMMs were being used, memory would be expandable up to 32 MB. However, because of the shortage of 1 Mb dRAM chips, Sun is using 256 Kb chips in the SIMMs. With 256 KB SIMMs, all 32 slots are used to provide the 8 MB that comes standard on the 4/110. Notice the way the Sun sales lit avoids mentioning this interesting fact. When 1 MB SIMMs for the 4/110 become available (expected to be later this year), you will be able to expand the memory on 4/110, but you will have to replace your some or all of your 256 KB SIMMs with 1 MB SIMMs (depending on how much extra memory you want). I strongly suggest that you talk to your Sun sales rep about how much, if anything, Sun plans to offer as trade-in for people who want memory upgrades (I have heard initial reports, but the above is enough bad news for now). -- Doug Moran, AI Center, SRI INTERNATIONAL ------------------------------ Date: 13 May 88 01:43:30 GMT From: roy%phri@uunet.uu.net (Roy Smith) Subject: Re: When suntools rereads the root menu > [[ ...I know that SunView rereads the root menu file when it changes > and I would hope that it does the same for the submenus as well. --wnl ]] That's not quite right. In suntools.c/wmgr_menufile_changes() you'll find: if (statb.st_mtime > wmgr_stat_array[sa_count].mftime) return(1); which means that a new menu file is only read if it is *newer* than the old one. On first glance, this seems reasonable, but consider doing: cp .rootmenu .rootmenu.old edit .rootmenu get a menu mv .rootmenu.old .rootmenu get another menu touch .rootmenu get a menu again The second menu you get will be the same as the first one because the news menu file will have a younger creation date than the old one. The first time this happened to me, it took a while to figure out what was going on. I maintain that the test should be if the time changes, not just gets bigger. I once sent in a bug report for this, but I don't know if it got picked up in later versions (we're still running 3.2, waiting for 4.0 RSN). Roy Smith, System Administrator Public Health Research Institute 455 First Avenue, New York, NY 10016 {allegra,philabs,cmcl2,rutgers}!phri!roy -or- phri!roy@uunet.uu.net [[ When I said "changes", I meant "newer". About the only thing you can do to improve the algorithm in suntools is to check to see if the inumber changes, and then check the time stamps if the inumber is the same. But I can even fool that one if I'm clever. --wnl ]] ------------------------------ Date: Fri, 13 May 88 08:49:08 EDT From: ted@braggvax.arpa Subject: Re: SLIP on Sun-OS 4.0 - don't hold your breath >From: rick@seismo.css.gov (Rick Adams) > >It's too bad that Sun insists on whining about how "bad" it is instead of >supplying what the customers want. Well, I suspect there will be a SLIP for 4.0 shortly after it goes out the door. I say this because we have just gotten PCNFS 3.0 and it supports SLIP. Sun has included dialup SLIP for the Sun as part of the server utilities disk, and I can't see them saying that they won't provide it for SunOS 4.0 given it's now officially part of one of their products. Ted Nolan ted@braggvax.arpa ------------------------------ Date: Fri 13 May 88 11:14:36-PDT From: William Westfield <BILLW@mathom.cisco.com> Subject: Re: Cisco terminal servers and SLIP Reference: v6n82 From: rick@seismo.css.gov (Rick Adams) The cisco Terminal server supports SLIP to end nodes only. : This version is compatible with 4.3bsd slip. These two statements are contradictory. If it is truly compatible with 4.3bsd, it should work fine. It's rather surprising. It should only take a few entries in the routing tables to make it work "completely". Nonsense. SLIP is a link level encapsulation scheme, and the cisco SLIP IS compatible with 4.3 BSD SLIP. The problem is that the cisco box supporting SLIP is not a router, and doesn't have any routing tables to "make a few entries in". (The cisco SLIP server can be thought of as a tranceiver extension cord.) We are sorry that the box does not meet Rick's needs, but we do feel that it will be useful for lots of people who only need to supply connectivity to a single host (eg Personal Computers in the generic sense). Bill Westfield cisco Systems. ------------------------------ Date: Thu, 12 May 88 11:14:14 PDT From: weiser.pa@xerox.com Subject: MAXUSERS > 12 in SunOS 3.2 and 3.4 "I've used MAXUSERS = 16 in both 3.2 and 3.4 with no problems. (Also in SunOS 3.5) -- Robert Bruner" The failure only happens if you have enough physical memory. We had no problem with 8MB suns, but did for 16MB. -mark ------------------------------ Date: 11 May 88 19:30:06 GMT From: tekbspa!tss!joe@uunet.uu.net (Joe Angelo) Subject: SunOS 3.2 MAXUSER>12 bug fix This is direct from SUN Tech Support; a bug relating to MAXUSER in SUNOS 3.2 --joe Bug Fix Id: unix/3.2/kernel/maxusers.bzero Problem: In Release 3.2 there is a bug which would occur during boot up of a kernel which had been configured for a larger MAXUSERS value. The value which causes this problem varies with the type of system and configuration. It has failed on a 3/260 with MAXUSERS of 24 and on a 3/160 with MAXUSERS of 12. The kernel with the larger MAXUSERS would fail with a panic: memall message during boot up right after the XY controller was probed out as in below: avail mem = nnnnnn Ethernet address = n:n:n:n:n:n:n:n xyc0 at vme16d16 ee40 vec 0x48 xy0 at xyc0 slave 0 panic: memall intrans|want This problem was caused by a bug in bzero. Release: 3.2 Fixed in: 3.4 Bug Report Id: kernel/other:1002996 [[ Joe was also kind enough to include the patch. However, I think that a patch for a problem like this is best obtained through the local Sun service center. Use the Bug ID given above and tell them you want the patch for that bug. You should be able to get it in a timely fashion. --wnl ]] ------------------------------ Date: 12 May 88 07:56:59 GMT From: munnari!sunaus.oz.au!richb@uunet.uu.net (Rich Burridge) Subject: Official patch #2 for calctool v2.2. Please apply it. This is the second official patch to calctool v2.2. Please apply it. Note that calctool v2.2 was distributed with patch #1 already applied; this is a fix to the first reported bug. As Skip Montanaro (sun!uunet.UU.NET!steinmetz!montnaro) points out: In news.c, init_ws_type() sends the file, "news.ps", to the NeWS server. Since the file reference is not absolute, the program will only execute successfully from the calctool directory. The fix is to define where the news.ps file is to be found, on the NEWSFILE line in the Makefile. If this is not present, it defaults to news.ps in the current directory. Thanks Skip. Rich. Rich Burridge, JANET richb%sunaus.oz@uk.ac.ucl.cs ACSnet richb@sunaus.oz UUCP {uunet,mcvax,ukc}!munnari!sunaus.oz!richb PHONE: +61 2 436 4699 ARPAnet rburridge@Sun.COM Sun Microsystems, Unit 2, 49-53 Hotham Pde, Artarmon, N.S.W. 2164, AUSTRALIA. ------CUT HERE------CUT HERE------CUT HERE------ *** original/Makefile Thu May 12 10:57:14 1988 --- Makefile Thu May 12 17:37:38 1988 *************** *** 14,22 **** BINARIES = sv_calctool ps_calctool BINDIR = /usr/local/bin LIBDIR = /usr/local/lib MANDIR = /usr/man/man$(MANSECT) MANSECT = l ! CFLAGS = -g -DHELPGIVEN -DHELPNAME=\"$(LIBDIR)/calctool.help\" HDRS = calctool.h color.h extern.h IMAGES = calctool.icon help.cursor --- 14,24 ---- BINARIES = sv_calctool ps_calctool BINDIR = /usr/local/bin LIBDIR = /usr/local/lib + HELPNAME = -DHELPGIVEN -DHELPNAME=\"$(LIBDIR)/calctool.help\" + NEWSFILE = -DNEWSGIVEN -DNEWSFILE=\"$(LIBDIR)/news.ps\" MANDIR = /usr/man/man$(MANSECT) MANSECT = l ! CFLAGS = -g $(HELPNAME) $(NEWSFILE) HDRS = calctool.h color.h extern.h IMAGES = calctool.icon help.cursor *** original/calctool.h Thu May 12 10:57:14 1988 --- calctool.h Thu May 12 17:43:04 1988 *************** *** 78,83 **** --- 78,88 ---- #define MAXREGS 10 /* Maximum number of memory registers. */ #define MAXVALID 6 /* Number of valid keys after an error. */ #define MIN(x,y) ((x) < (y) ? (x) : (y)) + + #ifndef NEWSGIVEN + #define NEWSNAME "news.ps" + #endif + #define NOBUTTONS BROWS * BCOLS #define THEIGHT (BROWS*BHEIGHT) + ((BROWS-1) * BGAP) + (2*BBORDER) #define TITEMS NOBUTTONS*2 + EXTRA /* Total definitions. */ *** original/news.c Thu May 12 10:57:15 1988 --- news.c Thu May 12 17:42:11 1988 *************** *** 129,135 **** init_ws_type() { if (ps_open_PostScript() < 0) return -1 ; ! if (send_ps_file("news.ps") == -1) { FCLOSE(PostScript) ; return(-1) ; --- 129,135 ---- init_ws_type() { if (ps_open_PostScript() < 0) return -1 ; ! if (send_ps_file(NEWSFILE) == -1) { FCLOSE(PostScript) ; return(-1) ; *** original/patchlevel.h Thu May 12 10:57:33 1988 --- patchlevel.h Thu May 12 17:46:01 1988 *************** *** 13,16 **** * reported to me then an attempt will be made to fix them. */ ! #define PATCHLEVEL 1 --- 13,16 ---- * reported to me then an attempt will be made to fix them. */ ! #define PATCHLEVEL 2 ------------------------------ Date: Wed, 11 May 88 16:14:24 EDT From: ames!srs!matt@ut-sally.UUCP (Matt Goheen) Subject: another SunOS 3.2 bug The following little program will render useless (until a "kill -9" is performed on the csh controlling the login) any console tty device (on a Sun2 or Sun3). If no one is logged in to the console and you rlogin and run it, it will hang the console getty until someone does a "kill -9" on the getty running on the console. I have NO IDEA how or why. It will print the message "prog_name: not a SUN-3 keyboard" and then lock up. Note that this is ONLY for consoles that are ttys, it works fine for any "normal" console (i.e. one with a monitor and keyboard). #include <sys/types.h> #include <sundev/kbd.h> #include <sundev/kbio.h> #include <sys/file.h> #include <stdio.h> extern int errno; main(argc, argv) register int argc; register char **argv; { register int fd; int type; if (argc > 2) { (void) fprintf(stderr, "%s: too many args\n", *argv); exit(1); } else if ((argc==2) && strcmp("on", argv[1]) && strcmp("off", argv[1])) { (void) fprintf(stderr, "%s: bad arg (\"%s\")\n", *argv, argv[1]); exit(1); } if ((fd = open("/dev/kbd", O_RDONLY)) == -1) { (void) fprintf(stderr, "%s: can't open kbd\n", *argv); exit(2); } if (ioctl(fd, KIOCTYPE, &type) == -1) { (void) fprintf(stderr, "%s: can't get kbd type (errno: %d)\n", *argv, errno); exit(3); } if (type == KB_SUN3) { (void) fprintf(stderr, "%s: a SUN-3 keyboard\n", *argv); exit(0); } else { (void) fprintf(stderr, "%s: not a SUN-3 keyboard\n", *argv); exit(4); } } uucp: {rutgers,ames}!rochester!srs!matt Matt Goheen maybe-nets: matt@srs.uucp OR matt%srs.uucp@harvard.harvard.edu ------------------------------ Date: 13 May 88 17:14:31 GMT From: fed!m1rcd00@uunet.uu.net (Bob Drzyzgula) Subject: Suns lose track of the console Does anyone know why Suns lose track of thier console? It seems that the problem has started to occur since we put up SunOS 3.5: After a few days of running fine, our Sun 3/160 server and its Wyse 30 console just stop talking to each other. The Sun will not respond to any terminal input, *even break*! Echos to /dev/console will not appear. The only solution is to rlogin to the system and reboot. Any ideas? Robert Drzyzgula Senior Information Systems Analyst /rcd Federal Reserve Board / m1rcd00 20th & C Streets, NW uunet!fed!< postmaster Washington, DC 20551 \ usenet ------------------------------ Date: Wed, 11 May 88 10:39:14 PDT From: gatech!tektronix!beatles.MEN.TEK.COM!jerryn@ut-sally.UUCP (Jerry Nelson) Subject: I need a caps lock indicator Has anyone developed an icon or something that will indicate whether or not the caps lock is set on a 3/60? The standard setup ( ie. no indication what so ever ) can be a big nuisance at times. ------------------------------ Date: 13 May 88 15:05:28 GMT From: johnksun@citron.uucp (John Kai-Yai Sun,) Subject: Help: debugging a program that needs mouse input? Dear Netnews readers, Would any kind soul out here tell me how to use dbx with graphics tool and debugging a piece of code that involved mouse input? Thanks. .....!gatech!hubcap!scarle!citron!johnksun ------------------------------ Date: Fri, 13 May 88 14:07:30 CDT From: rgh@shell.UUCP (Richard G. Hash) Subject: Wanted: Info on 'IdxTeX', automated index generation for LaTeX In TUGboat, Volume 8 (July 1987), No. 2, the program IdxTeX is described as an automated index generation for LaTeX from .idx files. It has been distributed to the TeX Users Group and to the DECUS library, since it was written for VAX/VMS. Has anyone out there got this running on a Sun? (I've got a 3/50), and if so, can you mail me info directly. Thanks, Richard G. Hash Shell Oil Company, Bellaire Research Center (713) 663-2311 ...!{sun,rice,psuvax,ut-sally,ihnp4,soma}!shell!rgh rgh@shell.com ------------------------------ Date: Fri, 13 May 88 08:33:40 EDT From: Anne.DeNiel@a.gp.cs.cmu.edu Subject: reliable and customizable Franz Lisp on Sun3? Does anyone know of a reliable, customizable (and cheap) version of Franz Lisp for Sun3 machines (running Sun Unix)? The customization I'm looking for is the possibility of enlarging the size of the namestack. Thank you for any pointers you can offer! Anne.DeNiel@a.gp.cs.cmu.edu ------------------------------ End of SUN-Spots Digest ***********************