Sun-Spots-Request@Rice.edu (William LeFebvre) (10/20/88)
SUN-SPOTS DIGEST Tuesday, 18 October 1988 Volume 6 : Issue 264 Today's Topics: Re: 4.0/nfs bug followup Re: VME Bus Errors on the Sun4 Re: .cshrc vs .login Mail Notification: "plug-compatile" biff ftpd under SunOS 4.0 (FIX) SUN INR 5.0 Weirdness netstat -r is broken with sub-netted Class B nets (LONG) Looking for 'line drawing' programs Using a HI-res Console with a SUN 3/280 Rack Mount Server? adding SMD disk? 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, 13 Oct 88 18:52:18 EDT From: ekrell@ulysses.att.com Subject: Re: 4.0/nfs bug followup I was told by a Sun instructor in a 4.0 administration seminar that if you were going to mix 4.0 and 3.x machines in a YP domain, you better make sure that the slave and master servers and 4.0 machines. ------------------------------ Date: Thu, 13 Oct 88 13:58:38 EDT From: sky!paavola@swan.ulowell.edu (Steve Paavola) Subject: Re: VME Bus Errors on the Sun4 We were having a number of problems porting support for our array processor products from the Sun3 to the Sun4. The problems all centered around moving floating point data to/from VME bus memory. The symptoms when moving single precision data varied from storing incorrect data to Bus Error messages. This is all to/from a 32d32 device. Sun has produced an ECO #4314 which brings the Sun4 cpu board to rev 13. This took several iterations, but appears to fix all the problems moving single precision data that we know about. However, there remain problems with moving double precision data. The symptom is Bus Error. The problem happens only when moving 8 byte aligned data because the compilers generate ldd/std instructions. Sun has stated that they don't plan to change the hardware to fix this. Instead they will produce a software fix - they won't generate ldd/std instructions from C or Fortran unless they are references to the stack, or the user has specified a switch saying 64 bit accesses are ok. They will also recompile all math libraries in this mode, so they never generate ldd/std instructions. This will be released at some as yet unnanounced date in the future (SunOS v4.1, Fortran 2.0). This means that all Sun4 customers will see a performance hit, whether they have vme memory devices or not. I don't know what the actual performance difference will be, however. ------------------------------ Date: Thu, 13 Oct 88 17:04:56 EDT From: Michael DeCorte <mrd@sun.soe.clarkson.edu> Subject: Re: .cshrc vs .login I disagree with the advice given here that setenv variables go into your .login. This will cause the following alias to work incorrectly. alias rxterm 'rsh \!* -n "xterm -display `hostname | sed s/\.clarkson\.edu//`:0.0" &' This is used to start another Xwindow on a remote host. If I put my setenv's into my .login this would fail. The only thing that I use my .login for is to do a few stty's set my terminal if I am on a dialup line (I do not use tset, it is cursed) and print a few things of interest such as the process load of the machine. Michael DeCorte // (315)268-2292 // P.O. Box 652, Potsdam, NY 13676 Internet mrd@sun.soe.clarkson.edu // Bitnet mrd@clutx.bitnet Clarkson Archiver Server archiver-server@sun.soe.clarkson.edu archive-server%sun.soe.clarkson.edu@omnigate.bitnet dumb1!dumb2!dumb3!smart!sun.soe.clarkson.edu!archive-server ------------------------------ Date: Fri, 9 Sep 88 18:15:17 EDT From: Mark Moraes <moraes@csri.toronto.edu> Subject: Mail Notification: "plug-compatile" biff We had a similar problem on our network of Suns. Our solution was to provide a plug-compatible 'biff' which basically sat around watching the mailbox. People who now use X11 (a rapidly growing community here) just use xbiff instead. [[ The shar file has been placed in the archives under "sun-source" as "biff.shar". It is 23421 bytes long. It can be retrieved via anonymous FTP from the host "titan.rice.edu" or via the archive server. For more information about the archive server, send a mail message containing the word "help" to the address "archive-server@rice.edu". --wnl ]] ------------------------------ Date: Thu, 13 Oct 88 14:42:33 EDT From: dpk@morgan.com (Douglas P. Kingston) Subject: ftpd under SunOS 4.0 (FIX) David writes: > The problem manifests itself in several ways. First, root can login and > transfer files with no problem. A normal user (who can login) gets 'User > <name> access denied.' when they try to ftp login. This is probably because root has a normal shell but the user has some shell that is not /bin/sh or /bin/csh. The 4.0 ftpd is based on 4.3BSD and it expects the file /etc/shells to have a list of the valid normal login shells if the system has alternates. These days, common alternates are ksh and tcsh. See the man page on getusershell(3) for details. By the way, /etc/shells was long overdue. -Doug- [[ Thanks also to Joe Ramey for a similar suggestion. --wnl ]] ------------------------------ Date: Thu, 13 Oct 88 10:41:55 PDT From: geoffb@ale.ind.trw.com (G. Geoffrey Baehr) Subject: SUN INR 5.0 Weirdness Users of SUN INR 5.0 router code under OS 3.5, have you seen this totally bizarre problem ? We have a fairly complicated internal network here, one of the solutions we implemented for connection of remote sites to the internal backbone nets was to use SUN's INR with 9600 baud sync modems attached to 3/50's for use as routers. This provided remote nets with access to the internal nets, Internet etc. However, some TCP connections were being reset due to timeouts with the remote host when using an INR route. Analysis revealed that the INR will not send an IP packet of exactly 108 bytes through from one side to the other. When a IP packet arrives that is 108 bytes in length from the Ethernet side, it is digested by the SUN. The remote router never sees this packet, as first the router never sent it along over the serial line. For TCP connections, this meanes that the sender never has this packet ack'ed, retransmits n times, sees no ack's as the remote has never seen anything to ack and then resets the connection, assuming that the remote host has died. Cute, eh ? Some peripheral information, changing the MTU on these lines to prohibit/force fragmentation, netmasks, trailers etc has no effect. The same problems occur on a subnetted Class B net as on a Class C or Class A. We have also run up an down in packet size, no other sized packets fail to get through. SUN has no idea, says that they are unable to reproduce this weirdness; anyone else seen this ? Thanks in advance geoffrey ------------------------------ Date: Wed, 12 Oct 88 10:33:05 EDT From: ehrlich@shire.cs.psu.edu (Dan Ehrlich) Subject: netstat -r is broken with sub-netted Class B nets (LONG) Machine Type: Sun 4/260S O/S Version: SunOS 4.0 Organization: Computer Science Department, The Pennsylvania State University 333 Whitmore Laboratory, University Park, PA 16802 Phone Number: +1 814 865 9723 Description: Netstat appears to not understand sub-netting correctly. At Penn State we are using the Class B network number 128.118 with sub-nets. If 'netstat -r' is run on a Sun4 running SunOS4.0 all sub-nets of 128.118 are reported to be 128.118. This works correctly on a VAX 11/750 running 4.3 BSD UNIX. (Sigh... will Sun ever get these things right?) The ethernet interface on the Sun4 is configured thusly: SUN4% ifconfig ie0 ie0: flags=63<UP,BROADCAST,NOTRAILERS,RUNNING> inet 128.118.6.4 netmask ffffff00 broadcast 128.118.6.0 Running 'netstat -r' one gets: SUN4% netstat -r Routing tables Destination Gateway Flags Refcnt Use Interface localhost localhost UH 1 26 lo0 psu-edu whitmore UG 0 523 ie0 psu-edu whitmore UG 0 0 ie0 temple whitmore UG 0 0 ie0 default whitmore UG 1 7749 ie0 psu-edu sunws2 UG 0 11193 ie0 psu-edu whitmore UG 0 302 ie0 psu-edu whitmore UG 0 0 ie0 psu-edu whitmore UG 0 0 ie0 u-pitt whitmore UG 0 0 ie0 drexel whitmore UG 0 0 ie0 psu-edu whitmore UG 0 4 ie0 psu-edu whitmore UG 0 0 ie0 psu-edu whitmore UG 2 38725 ie0 cpw-psc whitmore UG 0 0 ie0 cmu-net whitmore UG 0 0 ie0 prpnet whitmore UG 0 0 ie0 psu-edu whitmore UG 0 0 ie0 psu-edu sol4 UG 0 838 ie0 upenn whitmore UG 0 0 ie0 u-pgh-net whitmore UG 0 0 ie0 alcoa-net whitmore UG 0 0 ie0 psu-edu whitmore UG 0 0 ie0 psu-edu whitmore UG 0 6 ie0 psu-edu whitmore UG 0 0 ie0 psu-edu whitmore UG 0 0 ie0 psu-edu whitmore UG 0 0 ie0 psu-edu whitmore UG 0 34 ie0 psu-edu whitmore UG 0 0 ie0 psu-edu whitmore UG 0 0 ie0 lehigh whitmore UG 0 0 ie0 psu-edu whitmore UG 0 4391 ie0 psu-edu whitmore UG 0 0 ie0 psu-edu whitmore UG 0 0 ie0 pscnet whitmore UG 0 0 ie0 psu-edu whitmore UG 0 0 ie0 psu-edu shire U 35 242092 ie0 psu-edu shire U 0 333 ie0 psu-edu whitmore UG 0 0 ie0 psu-edu whitmore UG 0 0 ie0 psu-edu whitmore UG 0 0 ie0 The ethernet interface on the VAX 11/750 is configured thusly: 750% ifconfig de0 inet de0: flags=63<UP,BROADCAST,NOTRAILERS,RUNNING> inet 128.118.6.2 netmask ffffff00 broadcast 128.118.6.0 Running 'netstat -r' on the VAX 11/750 one gets: 750% netstat -r Routing tables Destination Gateway Flags Refs Use Interface localhost localhost UH 0 36 lo0 default Whitmore.psu.edu UG 5 61843 de0 otc-backbone Whitmore.psu.edu UG 0 80 de0 cac-ether1 Whitmore.psu.edu UG 0 1181 de0 cmu-net Whitmore.psu.edu UG 0 132 de0 willard-ether1 Whitmore.psu.edu UG 0 0 de0 alcoa-net Whitmore.psu.edu UG 0 0 de0 tb-ether1 Whitmore.psu.edu UG 0 69 de0 ee-ether1 Whitmore.psu.edu UG 1 10406 de0 cs-main psuvax1 U 8 89431 de0 cs-sys sunws2 UG 0 0 de0 cs-terminal Whitmore.psu.edu UG 0 0 de0 psu-cmpsc gondor UG 0 3209 de0 cs-work morelock UG 0 0 de0 cpw-psc Whitmore.psu.edu UG 0 0 de0 u-pgh-net Whitmore.psu.edu UG 0 205 de0 biotech-ether1 Whitmore.psu.edu UG 0 0 de0 boucke-ether1 Whitmore.psu.edu UG 0 19 de0 mcalistr-ether1 Whitmore.psu.edu UG 0 132 de0 drexel Whitmore.psu.edu UG 0 0 de0 otc-rs232 Whitmore.psu.edu UG 0 56 de0 hendersn-ether1 Whitmore.psu.edu UG 0 3 de0 upenn Whitmore.psu.edu UG 0 1 de0 osmond-ether1 Whitmore.psu.edu UG 0 0 de0 walker-ether1 Whitmore.psu.edu UG 0 58 de0 davey-ether1 Whitmore.psu.edu UG 0 191 de0 cac-ether2 Whitmore.psu.edu UG 0 0 de0 temple Whitmore.psu.edu UG 0 0 de0 ie-ether1 Whitmore.psu.edu UG 0 0 de0 areo-ether1 Whitmore.psu.edu UG 0 0 de0 findlay-ether1 Whitmore.psu.edu UG 0 11 de0 mystery-ether1 Whitmore.psu.edu UG 0 16 de0 cac-ether3 Whitmore.psu.edu UG 0 1 de0 otc-testnet Whitmore.psu.edu UG 0 0 de0 willard-ether2 Whitmore.psu.edu UG 0 0 de0 dieke-ether1 Whitmore.psu.edu UG 0 2 de0 mystery-ether2 Whitmore.psu.edu UG 0 0 de0 u-pitt Whitmore.psu.edu UG 0 0 de0 lehigh Whitmore.psu.edu UG 0 0 de0 pscnet Whitmore.psu.edu UG 0 0 de0 prpnet Whitmore.psu.edu UG 0 2 de0 Repeat-By: Run 'netstat -r' on a machine connected to a sub-netted class B network. Making sure the netmask is set correctly. ------------------------------ Date: Thu, 13 Oct 88 15:32:07 edt From: osu-cis!chemabs!lwv27@tut.cis.ohio-state.edu (Larry W. Virden) Subject: Looking for 'line drawing' programs We are looking for commercial sources for drawing programs. The users are familar with Macs so the best way they have for describing what they are looking for is 'MacDraw-like' - they want to generate technical drawings, slide presentations, etc. The environment is Sun 386i model 150 , SunOS 4.0 or newer. Thank you for your responses. Larry W. Virden UUCP: osu-cis!chemabs!larryV 674 Falls Place, BITNET: lwv27@cas.bitnet Reynoldsburg, OH 43068-1614 CIS: [75046,606] ------------------------------ Date: Thu, 13 Oct 88 13:38 EST From: STELLABO@CSHLAB.BITNET Subject: Using a HI-res Console with a SUN 3/280 Rack Mount Server? REQUEST FOR HELP: We have a rack mounted Sun 3/280 with a High Res (1600x1280) monitor. We We have a rack mounted Sun 3/280 with a High Res (1600x1280) display. We would like to locate the display, keyboard, and mouse about 30 feet away from the CPU box, to get it out of the Noisy computer room. When we tried to use cable extenders we lost the video signal after appox 6 ft. Does anyone know of a viedo signal repeater that will work in this configuration. Thanks for your help. Fred J. Stellabotte stellabo@cshlab.BITNET Cold Spring Harbor Lab (516)367-8420 ------------------------------ Date: Thu, 13 Oct 88 16:05:26 EDT From: betty symonds <symonds@chagall.cad.bnl.gov> Subject: adding SMD disk? This is a first time sending a request for information, so please bear with me. We have a Sun 3/160 C-4 with a 71 mb SCSI drive. We would like to add an SMD drive to avail ourselves of the larger disks. Can anyone tell me what we need in terms of hardware for connecting one and what one has to do to get the XYLOGICS controller to be known to the system. Also, where would we obtain a formatter tape for such a disk. This is information needed should we choose to purchase from other than our Sun dealer. We are currently running 3.2 on this system. Any information will be greatly appreciated. Thanks. Betty Symonds symonds@chagall.cad.bnl.gov ------------------------------ End of SUN-Spots Digest ***********************