Sun-Spots-Request@RICE.EDU (William LeFebvre) (07/15/88)
SUN-SPOTS DIGEST Wednesday, 13 July 1988 Volume 6 : Issue 138 Today's Topics: Re: SUNOS 4.0 experiences & impressions Re: How do I make an IBM readable tape on a SUN Re: A tool to draw structure charts fork/exec of some programs on Sun 3 with SunOS 4.0 fun with "limit" To qualify or not to qualify, that is the hostname Tooltool bug Single User Default rlogin/vi problems XDR Question Mail to programs only works sometimes?? Wang Emulators for Suns? DIGITAL Companion Color Printers on a Sun? 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: Tue, 5 Jul 88 13:16:14 EDT From: dpz@njin.rutgers.edu (David P. Zimmerman) Subject: Re: SUNOS 4.0 experiences & impressions From: clyde@emx.utexas.edu (Head UNIX Hacquer) 1. POWER ...We were told that the rack would need a 110-volt 30 amp circuit...Lo and behold, what appeared was a rack which requires a 220-volt, 30 amp circuit! I suspect someone screwed up in Sun's configuration department. You should have gotten a power supply that only needed 115 volts. We've got a Sun 4/280 with one Hitachi and tape drive which has the reportedly correct power supply (115V, single phase, 60Hz, 24amp). However, another Sun 4/280 with two Hitachis and tape drive here has the 230V supply. The February 3, 1988 price sheet only says that if you have *more* than two drives you need the 230V, perhaps the addition of the tape drive bounces the power requirements into that territory? From all I've heard, though, you shouldn't need the 230V power supply in your case. I wonder if you can bill Sun for headaches.... :-) [[ Probably not, but I bet you could *sue* them for headaches! :-) --wnl ]] ^David ------------------------------ Date: Tue, 5 Jul 88 17:46:06 EDT From: Root Boy Jim <rbj@cmr.icst.nbs.gov> Subject: Re: How do I make an IBM readable tape on a SUN ? ...The files I want to move ? are all c source (hence ascii text). I think the IBM can handle the ? ASCII-EBCDIC conversion, but if I can preconvert on the sun that would ? make me more confident.... Anybody know how to do this? No and yes. Write yourself the simplest, stupidest program you can imagine. For each line of C code, extend it to 80 characters, blank filled. Write a record. Repeat until done. Write an EOF. If you want to be smarter and put more than one file on a tape, precede each file with a record containing its filename, also 80 chars. At the end of all files write another EOF so you have two in a row. If the IBM insists on tape labels, label the tape on the mainframe and skip the first record on UNIX by reading it. All this assumes that you have raw interface to the magtape on the IBM. If that fails, write a tape on the mainframe in standard format with known contents, dissect it on the sun, and replace the contents of the mainframe file with the Sun file. Then reread it on the mainframe. (Root Boy) Jim Cottrell <rbj@icst-cmr.arpa> National Bureau of Standards Flamer's Hotline: (301) 975-5688 ------------------------------ Date: Tue, 5 Jul 88 18:22:08 BST From: Paul Hudson <mcvax!moncam!paul@uunet.uu.net> Subject: Re: A tool to draw structure charts > Does anyone have a public domain tool for the sun that will draw a > structure chart (or function call tree)? You want a pretty picture? or just the info? "cflow" will do the latter. Paul Hudson Snail mail: Monotype ADG Email: ...!ukc!acorn!moncam!paul Science Park, paul@moncam.co.uk Milton Road, "It's no use looking, there's nothing Cambridge, witty here" CB4 4FQ ------------------------------ Date: Tue, 05 Jul 88 14:54:53 PDT From: franz!akbar!layer@ucbarpa.berkeley.edu (Kevin Layer) Subject: fork/exec of some programs on Sun 3 with SunOS 4.0 The following program fails on a Sun3 with SunOS 4.0 but not SunOS 3.2. It is not sensitive to the system on which the program is compiled and linked (ie, not a dynamic loading problem). #include <sys/time.h> #include <sys/resource.h> main() { struct rlimit rlp; getrlimit(RLIMIT_STACK,&rlp); rlp.rlim_cur -= 5*7*1024; setrlimit(RLIMIT_STACK,&rlp); system("ls"); } gives this error when run: % cc bug.c 1.1u 1.6s 0:10 28% 0+152k 28+21io 73pf+0w % ./a.out ld.so: map heap error (22) for /dev/zero Just another reason to beware of SunOS 4.0, especially if you are interested in using Common Lisp. Allegro CL uses `setrlimit' to set a soft stack limit, thus allowing graceful handling of stack overflow. D. Kevin Layer Franz Inc. layer%franz.uucp@Berkeley.EDU 1995 University Avenue ucbvax!franz!layer Berkeley, CA 94704 (415) 548-3600, FAX: (415) 548-8253 ------------------------------ Date: Tue, 5 Jul 88 18:36:51 EDT From: Scott Schwartz <schwartz@cs.swarthmore.edu> Subject: fun with "limit" csh fans, running 4.0, try this: limit stacksize unlimited now type ls. I get: crt0: /usr/lib/ld.so mapping failure Use limit to put your stacksize back to 2048k. If you are feeling adventurous, try setting it to 2049k. Then when you try to ls you will get ld.so: map heap error (22) for /dev/zero Maybe I am too uptight, but somehow I think that something is not right. Shouldn't specifying "unlimited" give you the largest legal value that actually works? Shouldn't setrlimit return an error if it can't tolerate page-misaligned stack sizes? (maybe it does, I didn't check, but then neither did csh!) Maybe it should truncate to valid offsets. Similarly for ld.so: maybe the loader should be more clever about what it tries to do. Anyway, I don't know enough about this newfangled 4.0 stuff to offer a real solution. Am I correct in assuming there is a problem? -- Scott Schwartz [[ This reminds me of my second-favorite way to log out: "limit cpu 0" sometimes followed by any command. My favorite, of course, is "echo !#:h". --wnl ]] ------------------------------ Date: Tue, 5 Jul 88 13:44:05 CDT From: Jim Knutson <knutson%sw.MCC.COM@mcc.com> Subject: To qualify or not to qualify, that is the hostname People seem to be having problems with using domain qualified hosts names in various places, so I thought I would give an example of what can happen if you don't. We use Sun Unify/Simplify here for "distributed" data bases (data base access over the network). Accessing the database is accomplished in part by NFS and part by daemons fired up by inetd. When we upgraded from SunOS 3.4 to SunOS 3.5, we switched over to using domain qualified hostnames. Unfortunately, Sun Unify/Simplify broke. We could access data remotely, but not update it, reconfigure, or reload. So much for distributed databases. We called the problem into Sun but after a few weeks we were convinced that they weren't going to find an answer. On a wild guess, I changed the mount entry in /etc/fstab to use the qualified hostname instead of an alias. Lo and behold, things started working again. The moral of the story is, there is a lot of broken software out there that depends on exact matches or some kind of relation between what a host sets its hostname to, what the official hostname in the host tables is, and what hostname you ask for. Jim Knutson knutson@mcc.com im4u!milano!knutson [[ Sad but true. Anyone out there tried to get remote lpr queueing to work completely? Did you get "lprm" to work? Know why it's so hard? "gethostname" has to return the *primary* host name and not a nic-name. The name for the queued file is generated by the sending end using (in part) "gethostname", but for an lprm the receiving end gets the originating host name with a "gethostbyaddr" and tries to find a queue file that matches exactly. --wnl ]] ------------------------------ Date: Tue, 5 Jul 88 14:43:16 EDT From: Chuck Musciano <chuck@trantor.harris-atd.com> Subject: Tooltool bug A minor bug in tooltool can cause core dumps during initialization. (Imagine what a major bug would do...) The following patch cleans up the problem. Sorry for any inconvenience this may have caused anyone. Chuck Musciano Advanced Technology Department Harris Corporation (407) 727-6131 ARPA: chuck@trantor.harris-atd.com #! /bin/sh # This is a shell archive. Remove anything before this line, then unpack # it by saving it into a file and typing "sh file". To overwrite existing # files, type "sh file -c". You can also feed this as standard input via # unshar, or by typing "sh <file", e.g.. If this archive is complete, you # will see the following message at the end: # "End of shell archive." # Contents: patch # Wrapped by chuck@melmac on Tue Jul 5 14:42:31 1988 PATH=/bin:/usr/bin:/usr/ucb ; export PATH if test -f 'patch' -a "${1}" != "-c" ; then echo shar: Will not clobber existing file \"'patch'\" else echo shar: Extracting \"'patch'\" \(612 characters\) sed "s/^X//" >'patch' <<'END_OF_FILE' X*** tooltool.c Tue Jul 5 14:39:41 1988 X--- tooltool.c.orig Tue Jul 5 14:39:34 1988 X*************** X*** 372,378 **** X p[1] = 'C'; X if (i & S_META) X p[2] = 'M'; X! menu_set(b->menu, MENU_APPEND_ITEM, menu_create_item(MENU_STRING_ITEM, p, b->value[i], 0), 0); X } X } X else X--- 372,378 ---- X p[1] = 'C'; X if (i & S_META) X p[2] = 'M'; X! menu_set(b->menu, MENU_APPEND_ITEM, menu_create_item(MENU_STRING_ITEM, p, b->value[i]), 0); X } X } X else END_OF_FILE if test 612 -ne `wc -c <'patch'`; then echo shar: \"'patch'\" unpacked with wrong size! fi # end of 'patch' fi echo shar: End of shell archive. exit 0 ------------------------------ Date: Tue, 5 Jul 88 15:18:20 PDT From: Richard Fox <rfox@amelia.nas.nasa.gov> Subject: Single User Default One of our suns, when rebooted, always comes up in single user mode. To get it to come up in multiuser mode you then have to hit ctl-d. When the machine first starts to boot it prints: "b vmunix(0,0,0)" but does not specify the -s option. I heard that there is a prom setting that will bring the system up in single user mode even though the -s option is not specified. Can anybody tell me what this flag is so I can change it? And if there is something else going on entirely could somebody let me know how to get the system to come up in multi-user mode without the ctl-d? thanks rich rfox@amelia.nas.nasa.gov (415)694-4425 ------------------------------ Date: Tue, 5 Jul 88 18:10:31 BST From: Paul Hudson <mcvax!moncam!paul@uunet.uu.net> Subject: rlogin/vi problems Sounds like a termcap problem on the Vax to me. Check that the Vax *hasn't* had the multiple delete/insert line entries added for "sun" - I've never made them work. Paul Hudson Snail mail: Monotype ADG Email: ...!ukc!acorn!moncam!paul Science Park, paul@moncam.co.uk Milton Road, Cambridge, CB4 4FQ ------------------------------ Date: Tue, 5 Jul 88 12:12:34 pdt From: Carl Burch <cdb%hpclcdb@hplabs.hp.com> Subject: XDR Question Can anyone give me the general idea behind the XDR (sp?) specification. I believe it stands for External Data Representation, but I can't find any documentation on it. Is it being pushed as a possible standard? Carl Burch cdb%hpda@hplabs.HP.COM ------------------------------ Date: 1 Jul 88 18:41:18 GMT From: asp@cos.com (Andrew S. Partan) Subject: Mail to programs only works sometimes?? We are having strange problems here with mail to programs. We have one alias 'paper' that sends the message to lpr. We have an internal mailing list 'engine' that goes to a number of people and also via recnews into a newsgroup. We are also on some outside mailing lists with aliases that send the messages via recnews into newsgroups. Now for the weird part: Sometimes this works and sometimes it does not. The mail always gets to the people on the lists (or postmaster, if postmaster is on the list), but will sometimes die when it gets to the program. Sometimes everything works & sometimes everything does not. It seems to depend on the phase of the moon. Also, I can not always stop error messages from going back to the sender of the message. The error message is always "Unknown user:" with no user given. There is no more info in syslog either. The switch between working & not working comes when we rebuild our alias database (when we change it). We always rebuild both the YP database and the 'normal' alias database (which is only on our master YP server). There does not seem to be any rhyme or reason between it working or not working. Relevant lines in our /usr/lib/aliases file are: # This sometimes works & sometimes does not. paper: "|lpr -J'Paper copies of Mail'" # This sometimes works, but usually does not. isode-list: postmaster,cos-isode@cos.com owner-isode-list: postmaster cos-isode: "|/usr/local/lib/recnews cos.isode '' Submittal" owner-cos-isode: postmaster # This sometimes works, but usually does not. engine: people,people,people,people,cos-engine@cos.com owner-engine: postmaster cos-engine: "|/usr/local/lib/recnews cos.engine '' Submittal" owner-cos-engine: postmaster # This seems to work all of the time. There does not seem to # be any diff between these and the ISODE aliases. telecom-list: postmaster,cos-telecom@cos.com owner-telecom-list: postmaster cos-telecom: "|/usr/local/lib/recnews cos.telecom '' Submittal" owner-cos-telecom: postmaster % ls -l /usr/lib/aliases* -rw-r--r-- 1 root 31908 Jun 30 17:16 /usr/lib/aliases -rw-r--r-- 1 root 4096 Jun 30 17:18 /usr/lib/aliases.dir -rw-r--r-- 1 root 73728 Jun 30 17:18 /usr/lib/aliases.pag % ls -l /etc/yp/cos/mail.aliases* -rw-r--r-- 1 root 4096 Jun 30 17:16 /etc/yp/cos/mail.aliases.dir -rw-r--r-- 1 root 65536 Jun 30 17:16 /etc/yp/cos/mail.aliases.pag # Sample error message: From: MAILER-DAEMON (Mail Delivery Subsystem) Subject: Returned mail: User unknown To: Postmaster ----- Transcript of session follows ----- 550 User unknown ----- Message header follows ----- Received: by cos.com (3.2/smail2.2/04-17-87) id AA20861; Fri, 1 Jul 88 10:08:47 EDT Date: Fri, 1 Jul 88 10:08:47 EDT To: engine We are using Sun OS 3.5 with Yellow Pages. "cos" (aka "cos.com") is our YP Master server. There are 3 slave YP servers. There are a whole pile of clients. The server (and its diskless clients) are the only things running Sun OS 3.5, everything else is still running Sun OS 3.4. We have asked Sun for help on this, but have gotten none yet (it seems that all of their people are on vacation now). If anyone has any clue as to what might be going on, please lets us know. Thanks, --asp (Andrew Partan @ Corporation for Open Systems) -- asp@cos.com or asp%cos.com@uunet.uu.net -- {uunet, sundc, decuac, hqda-ai, hadron}!cos!asp ASN.1 Object Identifier: "{joint-iso-ccitt mhs(6) group(6) 157}" ------------------------------ Date: 5 JUL 88 11:22 From: B37335@ANLEES.BITNET Subject: Wang Emulators for Suns? Are you aware of any Wang Office Systems emulators for the Sun environment? Thanks, Bob Baker (B37335@anlees) ------------------------------ Date: 5 Jul 88 19:44:55 GMT From: lamont!bhaines@philabs.philips.com (bill haines) Subject: DIGITAL Companion Color Printers on a Sun? Does anybody know if DEC's new Companion Color Printers will hookup to a SUN workstation??? ------------------------------ End of SUN-Spots Digest ***********************