Ron Heiby (The Moderator) <unix-request@cbosgd.UUCP> (03/31/85)
Unix Technical Digest Sun, 31 Mar 85 Volume 1 : Issue 36
Today's Topics:
1-second resolution of process accounting times
Cheap unix engine?
Daylight Saving Time??? (3 msgs)
File system limit in 4.2 BSD
Interleaved memory on a VAX-11/780
Interleaved memory performance - a query
need help: RM03 with UNIX 2BSD
----------------------------------------------------------------------
Date: 28 Mar 85 21:11:27 GMT
From: jeff@fluke.UUCP (Jeff Stearns)
Subject: 1-second resolution of process accounting times
(The following applies to 4.2BSD on a VAX....)
Do you ever run lastcomm(1) or sa(8) and wonder about all those processes which
consumed zero seconds of CPU time? For example:
printenv tod tty22 0 secs Thu Mar 28 10:26
ls don tty11 0 secs Thu Mar 28 10:26
sendmail F root __ 0 secs Thu Mar 28 10:25
csh joe ttyp1 2 secs Thu Mar 28 10:25
C'mon now, even the simplest command takes *some* time:
% time date
Thu Mar 28 11:18:35 PST 1985
0.2u 0.3s 0:01 62% 120+24k 1+3io 4pf+0w
So what's the story here? Why do I get such crummy resolution from lastcomm?
Well, it turns out that the accounting file /usr/adm/acct comprises a series of
records, one per process. A record contains various interesting data about
the process, including user and system CPU time. These times are stored in a
cute little 16-bit floating point format with a dynamic range from 0 to 4.58E6
seconds. (Luckily, I don't run that many processes which consume more than 5.3
CPU days.)
But the time is recorded in *seconds*, and is TRUNCATED by the kernel (rather
than rounded) when it is written.
So most times recorded in the accounting file are wildly in error. A check of
yesterday's accounting data shows that 10,000 out of the total 12,000 processes
were recorded as using zero time! I know that DEC says a VAX is fast, but...
This makes the output of sa(8) very untrustworthy.
I would like to see the CPU time data recorded in a form which resolves to
milliseconds at the low end of the scale. It seems to me that the designer
of the current scheme went overboard with 13 bits in the mantissa and cut
himself short on exponent (only 3 bits). How about using a few more bits of
exponent, and recording the time in milliseconds? This would still give us a
couple of decimal digits of precision - and values which are meaningful for
those other 10,000 processes which slipped under the rug.
--
Jeff Stearns (206) 356-5064
John Fluke Mfg. Co.
P.O. Box C9090 Everett WA 98043
{uw-beaver,decvax!microsof,ucbvax!lbl-csam,allegra,ssc-vax}!fluke!jeff
------------------------------
Date: 27 Mar 85 04:07:09 GMT
From: halls@tut.UUCP (Andy Halls)
Subject: Cheap unix engine?
I'm looking for a cheap unix engine, something that I can
get to run System V for a few (2-3) users. What do you
suggest? I'm willing to do a little hardware hacking and to
haunt the used computer stores! To repeat the key word is cheap!
--
Andy Halls, Minicomputer Systems, Inc., 2037 Sixteenth Street
Boulder, CO 80302
(303) 443-3347
uucp: {cires | hao | nbires}boulder!halls
csnet: halls@boulder
------------------------------
Date: 31 Mar 85 02:26:19 GMT
From: leif@erisun.UUCP (Leif Samuelsson)
Subject: Daylight Saving Time???
Well, it seems that all 4.2BSD machines in Europe went on
DST a week too early this year. As I see it there are three
possible solutions:
1) Fix and recompile ctime.c if you have sources.
2) Remove the dst flag from the config file for a week. (Save vmunix!)
3) Reset the date for a week. This is clearly wrong but is probably
the easiest way out.
Any other suggestions?
----
Leif Samuelsson
Ericsson Information Systems AB ..mcvax!enea!erix!erisun!leif
Advanced Workstations Division
S-172 93 SUNDBYBERG 59 19 N / 17 57 E
SWEDEN
------------------------------
Date: 26 Mar 85 17:34:28 GMT
From: avolio@grendel.UUCP (Frederick M. Avolio)
Subject: Daylight Saving Time???
One other solution --- I believe this is good for any 4.2bsd
based U*ix. When you configure the system specify dst followed by a
number. 1 = usa, 2=australia, 3=w. europe, 4=central europe, and 5= e.
europe.
--
Fred Avolio {decvax,seismo}!grendel!avolio 301/731-4100 x4227
------------------------------
Date: 27 Mar 85 17:57:55 GMT
From: gwyn@BRL-VLD.ARPA (VLD/VMB)
Subject: Daylight Saving Time???
Another solution is to do what NBS did with WWV and change over to
using just Universal Time.
------------------------------
Date: 29 Mar 85 21:05:45 GMT
From: dhb@rayssd.UUCP
Subject: File system limit in 4.2 BSD
Has anyone ever successfully gotten more than 15 file systems on a 4.2 BSD
system? After many long delays, we are finally going to convert from 4.1
to 4.2, and we need to be able to mount more than 15 file systems. I tried
making the same changes that I made in 4.1 (increase the size of mdev in the
cmap stucture, increase NMOUNT and NSWAPX in param.h, fix mount/umount) but
it doesn't seem to work. I even talked to Mike Karrels in Dallas and he
indicated that that was all I had to do. The problem we are experiencing
is that random processes dump core at random times. This can be very
annoying if the shell core dumps, and it can be disastrous if "init" core
dumps. The behaviour seems to indicate some kind of swapping error. At
first I didn't even associate this problem with the changes to the coremap
structure but in a final act of desperation I backed off the change and
now the system runs fine. We have been trying to track what we thought
was a weird swapping error for three months (tues and wed eve.) and have
now been running smoothly WITHOUT the coremap changes for over two weeks.
We now feel that all our other changes are done and the system is ready to
release to the users. The only problem is that one of our machines
currently has eighteen mounted file systems and another one has
twenty-three! To compound the problem, we are also expecting delivery
of six new disk drives (400M Eagles). Before anyone says "Why dont you
just make a few bigger file systems?", there are internal political
reasons why we need to portion out the disk space in relatively small
(30 - 60 Meg) chunks.
Sorry for rambling on so much but if anyone has ever gotten more than
15 file systems to work, PLEASE let me know how you did it.
--
Dave Brierley
Raytheon Co.; Portsmouth RI; (401)-847-8000 x4073
...!decvax!brunix!rayssd!dhb
...!allegra!rayssd!dhb
...!linus!rayssd!dhb
------------------------------
Date: 26 Mar 85 18:30:06 GMT
From: mckay@burdvax.UUCP
Subject: Interleaved memory performance - a query
We here at burdvax are about to install an interleaved memory system
on a VAX 11/780 running Berkeley 4.2 and have had a couple of lively
discussions about the performance value of interleaved memory. We
have a somewhat largish configuration: a) 1 Unibus with 2 CDC9766's
off of an Emulex SC21V controller, an 8 line DZ connected to a
Rascal-Vadic modem rack, 64 tty ports on an Emulex CS21 emulating 4
DH's, an Interlan 10MB Ethernet controller, 2 Versatecs, a Printronix
interfaced thru an LP11 and a Cipher 800/1600bpi tape drive off of a
TU11 controller; b) an RM03 disk on a Massbus; c) 4 CDC9766's off of
an Emulex SC780 (a second Massbus).
I was wondering whether there is any experimental data or rational
consensus out in netland about this performance issue. Also, does
anyone run 4.2 without interleaved memory? Finally, do some of the
Berkeley Unix subsystems, eg "Fast file system", expect to be running
on systems with interleaved memory? What I am thinking of here is
the disk block sizes, eg 8K blocks (1/2 track) vs 4KB blocks (1/4
track), seem fairly large for a system with a single memory
controller.
Any info, including references on this topic would be appreciated.
Please begin discussion by mailing to me and I will summarize.
Thanks in advance.
Don McKay
...sdcrdcf!burdvax!mckay
...psuvax1!burdvax!mckay
Research & Development, System Development Corporation, Paoli, PA
------------------------------
Date: 27 Mar 85 21:55:41 GMT
From: aps@decvax.UUCP (Armando P. Stettner)
Subject: Interleaved memory on a VAX-11/780
Hi.
Interleaved memory on a 780 (or 785) will give you a little
better "overall" access time. The operating system does
not need to know whether or not the memory is interleaved.
The start-up command files on the console floppy will have to
adjust the memory controllers (those for booting the system,
which include the XXXBOO.CMD and those for comming up after
powerfailures ...).
I know of no real performance studies nor any reason not
to run interleaved if you have the right number or type
of memory controllers.
aps.
------------------------------
Date: 25 Mar 85 01:31:50 GMT
From: jwp@uwmacc.UUCP (jeffrey w percival)
Subject: need help: RM03 with UNIX 2BSD
I have UNIX 2.8BSD living on a Fujitsu Eagle with its own controller.
In addition, I have an RM03 drive on a separate controller. My goal
is to use the RM03 as a removable-medium device. My problem is that
when I spin down the RM, put in a new pack, then spin it up again,
I can't even "dd" a block to /dev/null. It says "read error". A reboot
clears the problem. Can anybody tell me how to make the RM useable
short of a reboot? Thanks!
--
Jeff Percival ...!uwvax!uwmacc!jwp
------------------------------
End of Unix Technical Digest
******************************
--
Ronald W. Heiby / ihnp4!{wnuxa!heiby|wnuxb!netnews}
AT&T Information Systems, Inc.
Lisle, IL (CU-D21)