drmurray@brocku.ca (Bill Murray) (11/10/89)
Are university has 5 sun3/50 (diskless) and 2 sun3/60 (servers) all using 4 Megs of memory and running sunOS vers 3.5. We would like to upgrade to vers 4.0 of the operating system, of which we already have some tapes. I have heard rumors that vers 4.0 needs more than 4 Megs, Is this true? If it is true, what is the cheapest and easiest way to upgrade these units? (preferrably something that can be under a maintenance contract, but not absolutely necessary) Should we be scrapping these units and upgrading to something more versitial? One last question, we received a few tapes when version 4.0 of the O/S first came out, what patch tapes have been issued since then or other things should I be aware about with sunOS 4.0? Thanks in advance for any advice you can give in any of these areas. Bill Murray Brock University St. Catharines, Ontario Canada. e-mail drmurray@brocku.ca
perry@tektronix.tek.com (Perry Hutchison) (11/14/89)
In article <2920@brazos.Rice.edu> drmurray@brocku.ca (Bill Murray) writes: > X-Sun-Spots-Digest: Volume 8, Issue 192, message 12 of 14 > Are university has 5 sun3/50 (diskless) and 2 sun3/60 (servers) all using > 4 Megs of memory and running sunOS vers 3.5. > We would like to upgrade to vers 4.0 of the operating system ... > I have heard rumors that vers 4.0 needs more than 4 Megs ... I have heard that 4.0 can be configured (more like stripped) to run in 4 Mb, but it's pretty marginal. This is still at the level of rumor, but perhaps a slightly more detailed rumor. > what is the cheapest and easiest way to upgrade these units? Memory expansion for the 3/60 should be available from Sun as well as from other suppliers. The "official" word from Sun is that the 3/50 cannot be expanded (other than by adding disk and/or tape to the SCSI port). However, memory upgrades of 4 Mb or 8 Mb (for a total of 8 Mb or 12 Mb) are available from third-party suppliers. Ads can be found in Unix World and probably other places. I have had no experience with any such; however I would anticipate the following potential problems: 1. The power requirements and/or heat dissipation of the additional memory might exceed the 3/50's power supply and/or cooling (airflow) capabilities. 2. Since the 3/50 was not designed for memory expansion, these offerings have to be connected in unconventional ways. I have seen a picture in someone's ad which leads me to suspect that they plug into the CPU and/or MMU sockets on the 3/50 board, and the CPU and/or MMU then plug into sockets on the add-on memory board. These problems must not be insurmountable, or the third-party suppliers would not be able to stay in business, but there may be a nuisance factor: if the system is under Sun maintenance you might have to remove the memory expansion before having it serviced. If you are using a third-party maintenance organization, you might want to check with them as to any experience they may have had with this sort of thing, and whether they would be willing to support it.
norm@ames.arc.nasa.gov (Norm Finn) (11/15/89)
Our experience with 4.0.1/4.0.3 and 3/50s and Clearpoint memories: 1) Out-of-the-box 4.0.x is horribly slow on a 4M 3/50. Slow as in almost unuseable if you're used to 3.x, and obviously slow if you've never seen a Sun in your life. 2) If you follow all the suggestions in Sun's documentation to strip the system to its bare bones, you can get a 3/50 that is definitely slower on the average than 3.x, but still useable. It is especially annoying when you're running makes and textedits at the same time. Call it a half-fast system :-). 3) If you get a 4M from Clearpoint to upgrade to 8M total, and strip the OS down, you have a machine that is almost always faster under 4.0.x than it was under 3.x. We haven't tried a non-stripped 4.0.x on 8M. We've had half a dozen upgraded 3/50s running for several months, now, and have experienced no failures. You can rip the extra memory out if a 3/50 breaks and you have a Sun service contract, and put it back when the FE leaves. The Clearpoint documentation is very good. Norm Finn domain: norm@ultra.com Ultra Network Technologies Internet: ultra!norm@ames.arc.nasa.gov 101 Daggett Dr. uucp: ...ames!ultra!norm San Jose, CA 95134 (408) 922-0100
madd@world.std.com (jim frost) (11/16/89)
sequent!ccssrv!perry@tektronix.tek.com (Perry Hutchison) writes: >The "official" word from Sun is that the 3/50 cannot be expanded (other >than by adding disk and/or tape to the SCSI port). However, memory >upgrades of 4 Mb or 8 Mb (for a total of 8 Mb or 12 Mb) are available from >third-party suppliers. Ads can be found in Unix World and probably other >places. >I have had no experience with any such; however I would anticipate the >following potential problems: >1. The power requirements and/or heat dissipation of the additional memory > might exceed the 3/50's power supply and/or cooling (airflow) capabilities. >2. Since the 3/50 was not designed for memory expansion, these offerings > have to be connected in unconventional ways. Helios makes such an expansion that is easy to install and works great. No problems with cooling, power, etc. You remove several chips and snap the board into place in their sockets, then put one of the chips you removed in a socket on the board. They even supply the tools to do all of this. Takes about five to ten minutes and can be done by anyone who can handle removing and replacing a chip. It's simple and it works. Highly recommended. jim frost software tool & die madd@std.com
cortesi@ames.arc.gov (David Cortesi) (11/16/89)
A few weeks ago I asked about information on adding RAM to a 3/80; others
have asked related questions. I now have more information which may be of
interest to others.
The 3/60, 3/80 and SparcStation all use the same RAM format, namely
1 megabyte x 9 bit, 100ns, low profile SIMMs
The ones in my 3/80 were made by Texas Instruments; however they are an
industry standard part. A T.I. representative verified to me that there
was nothing special or Sun-specific about the part number.
SIMMs are a commodity item whose price changes daily. The general trend
is downward; a recent posting in this forum claimed that due to a new
factory about to come on-stream prices should plunge in 1990. (Then
again, new factories are sometimes delayed coming up...)
4MB upgrade kits (4 SIMMs, a grounding strap, and instructions) are
available from these vendors at least (alpha order)
Clearpoint 1-800-253-2778
Helios 1-408-432-0292
Parity 1-408-378-1000
They charge a bit more than the street price for SIMMs (about $500 for
4MB, as of last week). In exchange you get advice & support from a vendor
experienced with Sun hardware. If you feel adventurous, you can save
another 25% by ordering from Chip Merchant (1-619-268-4774) who last week
had 9x1Meg, 80ns (faster) SIMMs for $89. But they know nothing about Sun
hardware; I didn't ask about warranty, return policy or service.
Flat-topped 3/50s can be upgraded by the addition of a piggy-back board.
The above vendors all carry these; they currently charge around $950 for a
board that adds 4MB (total of 8MB). In all cases they assert there are no
heating problems; that installation is very simple; that if a problem
develops in the board they will fix it. Regarding Sun service they say
that "at worst" you'd have to remove the board before Sun would work on a
problem; one claimed that if a Sun service person diagnosed a problem in
the board, they would pay for the Sun service call.
Nothing above should be taken as an endorsement of any company, and
especially in regard to warranties your mantra should be "caveat emptor."
Neither I nor Informix have any connection with the aforementioned outfits
- not even as customers as my req. is still working its way through the
purchasing department.
/////// / David Cortesi {pyramid|uunet}!infmx!cortesi
////// // ///////////////////////////////////
//// / /// Informix Software // //
/// / //// 4100 Bohannon Drive // (aphorism goes here) //
// / ///// Menlo Park, CA 94025 // //
/ //////// (415) 926-6300 ///////////////////////////////////guy@uunet.uu.net (Guy Harris) (11/19/89)
>I have heard that 4.0 can be configured (more like stripped) to run in 4 >Mb, but it's pretty marginal. This is still at the level of rumor, but >perhaps a slightly more detailed rumor. As someone who, on a regular basis, uses a diskless 3/50 with 4MB of memory, running SunOS 4.0.3, I can contribute at least one non-rumor data point: I run with a configured kernel (a MUST for 4.x on small machines - but then you're wasting memory if you run a GENERIC kernel for ANY release of SunOS, including 3.x, 2.x, or 1.x, or probably for any other system with BSD-flavored configuration mechanisms, unless you have one of every device, file system, etc. that the OS supports); the configuration file is vanilla DL50. I have a sparse desktop (a "cmdtool" to capture console output, two or three 54x80 shelltools, a shelltool running a mail-watcher script) and no fancy root window backgrounds. (I'm running the 4.0.3 versions of the Sunview stuff, not the 3.x versions.) I find the performance acceptable, although it could certainly be better. I run compiles on my machine; it does have some impact on the other "shelltool" (usually logged into a 4/280 here), and isn't blazingly fast, but is "good enough". I sometimes fire up Berkmail on my machine while the compile is running; it takes a while, but then again: auspex% ls -l /usr/spool/mail/guy -rw------- 1 guy 2567314 Nov 18 11:55 /usr/spool/mail/guy (I cut and pasted that; the number is NOT a typo, it really is that big). It does fire up much faster on the 4/280, but that has 32MB of memory, a much faster CPU, and a direct connection to the disks on which "/usr/spool/mail" resides (no, we're not using the 4/280 as our primary server, it just has "/usr/spool/mail" on it; the primary server is, as you might expect, an Auspex NS5000, so we do use our own product). This doesn't include all the assorted tuning suggestions Sun has put out. I still run "routed", but then we don't have enough machines that the route table fills up memory; if you have a big network, turning "routed" off on non-multi-homed hosts may help. I still run "sendmail", just in case somebody sends mail to me on my home machine rather than on "auspex", although outgoing mail from here now has internal host names stripped out in the hopes that mail delivery doesn't involve the user's machine. Note: if you're running only X11 (not X11/NeWS), and not Sunview, you may want to consider removing the "pseudo-device" lines for "dtop" and "win", as the MIT server, as I remember, gets keyboard and mouse input directly from "/dev/kbd" and "/dev/mouse" and doesn't require all the Sunview support kernel goo (which, I seem to remember from a while ago, takes about 70KB on a Sun-3 - it includes pixrect code for cursor tracking). None of this, of course, says "memory upgrades don't do any good"; I suspect things would run better with 8MB. It does, however, say that, at least in some situations, 4MB is *not* unusable. Your mileage may, of course, differ. (I don't run big LISP jobs, for example. :-) I also don't run GNU Emacs; I run MicroEMACS at present, which "size" tells me is smaller than "vi".) The statements above held true before I started running 4.0.3, and was running 4.0.1 (and, I think, back when I was running 4.0.)
pat@decwrl.dec.com (Pat Lashley) (12/20/89)
In article <3272@brazos.Rice.edu> auspex!guy@uunet.uu.net (Guy Harris) writes: >X-Sun-Spots-Digest: Volume 8, Issue 206, message 7 of 15 > >I find the performance acceptable, although it could certainly be better. >I run compiles on my machine; it does have some impact on the other >"shelltool" (usually logged into a 4/280 here), ... I tend to be a workstation `power user' - multiple shelltools, GNU emacs with ispell and emacs-server processes, continuous mailtool, perfmeters, etc.; I also tend to use the GNU/FSF (memory is cheap, and getting cheaper...) versions of various utilities. While first testing X/NeWS on a color 386i (Beta-1 was significantly slower than the final release :-), I discovered a very simple (almost too obvious to see) method for preventing compiles from degrading the interactive performance of other windows. It should work for (almost) any windowing system. Our old friend nice(1). Where `make foo' would bring even mouse tracking to a near standstill; `nice make foo' hardly affected interactive performance at all. -Pat