roy@phri.UUCP (Roy Smith) (08/15/87)
Here's a strange one. From my MtXinu 4.3BSD/NFS Vax-11/750, I do
"rlogin alanine &", where alanine is a Sun-3/50 running SunOS-3.0.
Eventually the rlogin completes and gets blocked for tty output (I use
tostop mode). When I do a "fg", I get a burst of noise in addition to the
regular output.
Well, it's even stranger than that. The noise is repeatable. It's
always "ss^@^X^@P^@^@^@^@". At least that's what I see on my screen
(CIT-101, a vt-100ish clone). It also happens when I try other 3.0
machines (various Sun-3 models) but not with 3.2 machines. Never happens
if I don't run the rlogin in the background, and doesn't happen if I
suspend an active rlogin and restart it. Anybody know what's going on?
--
Roy Smith, {allegra,cmcl2,philabs}!phri!roy
System Administrator, Public Health Research Institute
455 First Avenue, New York, NY 10016chris@mimsy.UUCP (Chris Torek) (08/17/87)
In article <2850@phri.UUCP> roy@phri.UUCP (Roy Smith) writes: >... I get a burst of noise in addition to the regular output. >Well, it's even stranger than that. The noise is repeatable. It's >always "ss^@^X^@P^@^@^@^@". This one is easy. It is really <DEL><DEL> s s <NUL><CAN><NUL> P <NUL><NUL><NUL><NUL>, and this is window_size = 24 rows 80 cols, 0 ypixels 0 xpixels. Your Vax is trying to tell your Sun about its `window size', and your Sun is not listening. Why it breaks I cannot say; we run locally hacked versions of rlogin and kernels on our Suns. -- In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 7690) Domain: chris@mimsy.umd.edu Path: seismo!mimsy!chris
ram%shukra@Sun.COM (Renu Raman, Sun Microsystems) (08/17/87)
Another strange but similar problem is rcp/rlogin etc work
fine from a Sun (3.0) to a Vax (4.3) but doing the other way around
dumps core on the Sun. Don't know why and didn't have the time to
figure out why before I moved in here.
---------------------
Renu Raman ARPA:ram@sun.com
Sun Microsystems UUCP:{ucbvax,seismo,hplabs}!sun!ram
M/S 5-40, 2500 Garcia Avenue,
Mt. View, CA 94043mouse@mcgill-vision.UUCP (der Mouse) (08/28/87)
In article <7997@mimsy.UUCP>, chris@mimsy.UUCP (Chris Torek) writes: > In article <2850@phri.UUCP> roy@phri.UUCP (Roy Smith) writes: >> [When running "rlogin sun &", I get noise that is always >> "ss^@^X^@P^@^@^@^@".] > This one is easy. It is really > <DEL><DEL> s s <NUL><CAN><NUL> P <NUL><NUL><NUL><NUL>, Shouldn't those two <DEL>s be <META-NUL>s? (Forgive the perversion of the ASCII names, please....) > Your Vax is trying to tell your Sun about its `window size', and your > Sun is not listening. Indeed it is trying. Why it breaks is not clear. What is really hard to understand is why this junk is visible only when the rlogin is run in the background, because normally rlogin doesn't try to tell the rlogind about the window size unless the rlogind volunteers the information that it is interested. der Mouse (mouse@mcgill-vision.uucp)
merlin@hqda-ai.UUCP (David S. Hayes) (09/11/87)
1. The Sun-4 does have ECC (error-correcting core) memory. So, for that matter, do the Sun-3/2xx systems. Mine 3/280 has 8 megs, and has been running for almost 3 months, without even correctable error. 2. There is a system with the file system in memory, sort of. In the Xerox lisp machines, there is no regular file system. There are only "objects", which are memory-resident. The system saves "worlds" on the disk (in addition to VM paging space). A world is merely a copy of the virtual memory space. 3. 300 megabytes is not sufficient as a design limitation. Most people won't need that much, of course, but some need more. I have 14 lisp machines here. Most have at least 400 megabytes of paging space, and several have over 1000 megs! Even so, some applications have to be split over several machines via remote procedure calls. -- David S. Hayes, The Merlin of Avalon PhoneNet: (202) 694-6900 UUCP: *!uunet!cos!hqda-ai!merlin ARPA: merlin%hqda-ai@mimsy.ue, a.
scc@cl.cam.ac.uk (Stephen Crawley) (09/24/87)
In article <424@hqda-ai.UUCP> merlin@hqda-ai.UUCP (David S. Hayes) writes: ... >2. There is a system with the file system in memory, sort of. In >the Xerox lisp machines, there is no regular file system. There >are only "objects", which are memory-resident. The system saves >"worlds" on the disk (in addition to VM paging space). A world is >merely a copy of the virtual memory space. Sorry Dave, but you are only partly right. A Xerox Lisp machine is normally configured to have a conventional file system on the local disc (your LispFiles volume ...). In addition, files can be stored on a remote XNS fileserver. Though a "world" (actually a "sysout") may be saved to disc and used like a file system, users generally use a real filesystem as well. The problem with using a sysout as your only filesystem is that when Lisp falls over or someone trips over the power cord, you may be left with a badly corrupted sysout, and no way to scavenge it. This is not my idea of what a file system ought to be like. [What has this to do with rlogin?] -- Steve