Oh my god, I hate it...
Stefan Froehlich
No, I cannot say that I did not expect it to happen. But still I _hate_ it.
I am some kind of wannabe sysadmin (not that I actually wanted to be though),
I have some experience with Unix and TCP/IP networks and some other things,
but nothing that would qualify me as a guru in any way.
Still, I am perhaps the most unix-aware person here at my department, so I
was given the responsibility for our hp-ux server two months ago, when the
old [1] sysadmin left for Japan [2].
Imagine a machine with five mounted volumes, four of them used for luser
space and the remaining one for all the rest of the system... kernel,
system, applications, spool dir, temp dir. Imagine no user quota on the
machine [3]. Imagine a 95% filled root-file system.
I have already said, that I did expect it, nevertheless I did not have
a real chance to prevent it [4]. Yesterday at noon, two of our lusers
decided to print some nicely colored corel-draw slides through my
printer queues. The result was desastrous: root filesystem full, all
printers stopped working.
As a first action, I deleted some of the bigger annoyances on the disc [5].
Then I cleand the spool dirs, restarted the scheduler. Nothing. No printing
at all. About 1 minute later, the telephone started ringing [6]. After I
got rid of the calling luser, I phoned a friend of mine (who also has some
basic ideas of system administration) and did not stop talking for the next
four hours [7]. Well, to put it simple: we did not find a solution and
finally decided to go home [8] and to restore from tape next morning.
Today, when I go to my desk, open the terminal and view the printer log,
what do I see? Since this morning, everything is working again. Flawlessly.
And without ANY human intervention at all.
Now this reduces my stresslevel quite a bit, but I am still pissed of. What
is an unclued admin needed for, if the OS repairs itself over the night?
IF I have to do this job, I at least want to DO it and not just WATCH.
Stefan
-----------------------------------------------------------------------------
[1] ok, physically spoken, he was not much older than I am
[2] perhaps the most distant place available from here. I think he
knew why.
[3] this seems to be common practice in academic environment.
[4] my job is to _administrate_ the machine, not to spend money for it.
[5] this was the only rewarding moment that day.
[6] service was cut for perhaps 50 people at 7 printers.
[7] this was partly for a common analysis of the problem, but mostly to
justify the busy phone line. I would not want my boss to come to my
room and see the phone off-hook for no obvious reason.
[8] by this time, we were almost alone in the office, anyway.
Lars Balker Rasmussen
sfro...@et.tuwien.ac.at (Stefan Froehlich) writes:
> Now this reduces my stresslevel quite a bit, but I am still pissed of. What
> is an unclued admin needed for, if the OS repairs itself over the night?
> IF I have to do this job, I at least want to DO it and not just WATCH.
Operating systems are like children. If there is noone around to
appreciate your crying, why bother?
ObASR: Finding the subtle differences between Solaris SPARC and x86...
--
Lars Balker Rasmussen "No."
Michael Driscoll
In article <69i05i$9...@news.tuwien.ac.at>, Stefan Froehlich wrote:
>I am some kind of wannabe sysadmin (not that I actually wanted to be though),
I hereby pronounce you a sysadmin, for better or worse.
ObAProgrammingR[-1]:
Did you ever spend eight hours working in a deeply nested function involving
many fork()'s, naively trying to figure out just where the fsck everything
is going wrong, only to discover that you are accidently doing an fclose()
on the read pointer instead of the write pointer?
*weep*
Go ahead, call me a luser. I already have, after all.
Didn't help that this was an FCGI program taking some rather nonstandard
input from the server[0] so that gdbm/strace and such couldn't be used,
and so I had to do all of the diagnosis via syslog().
Ever fill a 1000 megabyte partition[1] with debugging logs? Ever reached
your filesystem's limit on file size? You will.
[-1] Is there such a group? I remember posts about newgrouping
alt.programmer.recovery but it doesn't seem to exist in
these parts.
[0] File upload a la RFC 1867. Whee.
[1] Normally about 10% allocation on this disk.
--
Michael Driscoll
<fen...@frob.ml.org>
<mdri...@mines.edu>
Stefan Froehlich
Perry Rovers (Perry....@IAE.nl) wrote:
> Pink elephants with an attitude inspired Stefan Froehlich <sfro...@et.tuwien.ac.at> to tell alt.sysadmin.recovery:
> Here's where you're wrong. Joe Random Hacker was pissed because your system
> was messed up so he couldn't use it to leech warez/p0rn, spam and modify
> www.fbi.gov. So he fixed it for you. HTH. HAND.
I did tell you that the file system was full, didn't I? And this is
certainly not due to a Warez site... I personally know the people who
produce multiMB simulation results.
OTOH I won't tell the machine name here, be sure for that...
> If you want to DO something and become clued, just newfs it and reinstall
> everything from scratch. Great learning experience.
This is what I'll most likely have to do anyways within the next few months.
The unbelievable has happend and we _are_ granted some money to buy a new HP.
Guess who will have to take care of it?
Bye,
Stefan
Chris Saunderson
sfro...@et.tuwien.ac.at (Stefan Froehlich) writes:
> No, I cannot say that I did not expect it to happen. But still I _hate_ it.
[snip]
> Today, when I go to my desk, open the terminal and view the printer log,
> what do I see? Since this morning, everything is working again. Flawlessly.
> And without ANY human intervention at all.
> Now this reduces my stresslevel quite a bit, but I am still pissed of. What
> is an unclued admin needed for, if the OS repairs itself over the night?
> IF I have to do this job, I at least want to DO it and not just WATCH.
You know, I'm sure I'm not the only one who _wants_ to believe you.
"No, really, the machine put itself back together!"
"No. Way."
"Truly!"
"There is a god."
Of course, just to rain on your parade - that bit is working. Guaranteed,
there'll be someone on the phone at some point during the time you're
adminning this box complaining that something isn't working and that
it worked before _caused_ by your need to go to the tapes.
Not that I'm bitter, of course.
Saundo
--
Chris "Saundo" Saunderson Chris.Sa...@adelaide.maptek.com.au
Technical Support Engineer Telephone: (08) 8379 7333
Maptek Pty Ltd Fax: (08) 8379 7377
Powered by Linux, Jolt and the Orb.
Peter Gutmann
fen...@ulfheim.frob.ml.org (Michael Driscoll) writes:
>In article <69i05i$9...@news.tuwien.ac.at>, Stefan Froehlich wrote:
>>I am some kind of wannabe sysadmin (not that I actually wanted to be though),
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>I hereby pronounce you a sysadmin, for better or worse.
You may kiss {the workstation | your life goodbye | the PHB's butt}.
Peter.
Steve Conley
Michael Driscoll <fen...@nospam.frob.ml.org> wrote:
>Ever fill a 1000 megabyte partition[1] with debugging logs? Ever reached
>your filesystem's limit on file size? You will.
"And the company that will bring it to you..."
--
Steve Conley sco...@summit.bor.state.oh.us
http://www.bor.state.oh.us/people/conley.html
ObGov'tEmployee: I don't represent the Ohio Board of Regents.
Ben Cantrick
In article <69l7cj$pjb$1...@charm.magnus.acs.ohio-state.edu>,
Steve Conley <sco...@summit.bor.ohio.gov> wrote:
>Michael Driscoll <fen...@nospam.frob.ml.org> wrote:
>>Ever fill a 1000 megabyte partition[1] with debugging logs? Ever reached
>>your filesystem's limit on file size? You will.
>
>"And the company that will bring it to you..."
....Microsoft, who thinks that /etc is a fine place for a root-owned
RADIUS process to dump its copious authentication logs.
Have you ever had a frantic call from a luser who screams "my root
partition is 110% full, and this is all your fault?"
You WILL.
-Ben
P.S. http://www.dim.com/~mackys/youwill.html
--
Ben Cantrick (mac...@dim.com) | Yes, the BGC dubs still suck.
BGC Nukem: http://www.dim.com/~mackys/bgcnukem.html
The Spamdogs: http://www.dim.com/~mackys/spamdogs
"I don't think so," said Rene Descartes. And then he vanished.
Chris Saunderson
pgu...@cs.auckland.ac.nz (Peter Gutmann) writes:
> >>I am some kind of wannabe sysadmin (not that I actually wanted to be though),
> > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> >I hereby pronounce you a sysadmin, for better or worse.
>
> You may kiss {the workstation | your life goodbye | the PHB's butt}.
T | N > K
Saundo
ObASR:
"I'm looking for a data directory called <blah>"
"So, did you look under /maptek/vcdata/blah?"
"Yeah, it was empty".
"No it's not."
"Yes it is".
"No it's not. See?" <ls -lart /maptke/vcdata/blah>
"Oh."
Same one who interrupted lunch to tell me this. So that HE could go
on leave. Ah-hah.
Stefan Froehlich
Chris Saunderson (sau...@schroedinger.adelaide.maptek.com.au) wrote:
> sfro...@et.tuwien.ac.at (Stefan Froehlich) writes:
>
> > And without ANY human intervention at all.
> > Now this reduces my stresslevel quite a bit, but I am still pissed of. What
> > is an unclued admin needed for, if the OS repairs itself over the night?
> > IF I have to do this job, I at least want to DO it and not just WATCH.
>
> You know, I'm sure I'm not the only one who _wants_ to believe you.
Oh, oh, to disappoint you, I have to admit that I made a mistake. _Big_
mistake. It has all been my fault, I only did not notice it up to today.
To make the story short, the machine is named hal, and last tuesday was
January, 12th. What can you expect from an HP9000, if you forget its
birthday?
Bye,
Stefan
Tom Perrine
In article <69jafv$c4n$2...@Garfield.IAE.nl>,
Perry Rovers <Perry....@IAE.nl> wrote:
>Here's where you're wrong. Joe Random Hacker was pissed because your system
>was messed up so he couldn't use it to leech warez/p0rn, spam and modify
>www.fbi.gov. So he fixed it for you. HTH. HAND.
>
Well, actually, There Was This Site about 2 years ago...
Seems that while an outside consultant/security-BOFH was helping them
recover from an incident, he discovered that the intruders had gotten
pissed off because one of the machines was rebooting on them so often,
so they installed about 30 patches[1], a sendmail upgrade[2],
backed-up, re-partitioned the disks, and reloaded all the data[3].
They're probably still sad that they actually found the intruders,
because they haven't had an OS upgrade since[4].
--tep
[1] including security patches, plus extra stuff[5].
[2] including fixing all their broken aliases, and cleaning the dead
accounts from the passwd file[6]
[3] resulting in about a 15% performance increase
[4] still running SunOS4.1.3
[5] SSH :-)
[6] correctly
--
Tom E. Perrine (t...@SDSC.EDU) | San Diego Supercomputer Center
http://www.sdsc.edu/~tep/ | Voice: +1.619.534.5000
I miss my 36-bit friends: Multics, TOPS-10, and TOPS-20.
Kai Henningsen
Perry....@IAE.nl (Perry Rovers) wrote on 15.01.98 in <69m2cq$q9r$1...@Garfield.IAE.nl>:
> [-2] not to mention the luser who messed with his mail causing the local
> freenet to blow up... which caused their secondary mx (us) to be made
> aware of the fact that indeed.. space is not infinite.
Someone remember the sendsys accident way back when? Some Amiga user, I
believe. Nearly took Germany off the net for a while - his MXes was all
the German backbone back then, and those machines didn't take kindly to a
few thousand simultaneous SMTP connections whenever they tried to reboot.
For some reason, a lot of people hacked their cnews and inn shortly after
not to propagate sendsys control messages ...
"How not to make a good impression."
Kai
--
Internet: k...@khms.westfalen.de
Bang: major_backbone!khms.westfalen.de!kai
http://www.westfalen.de/private/khms/
Craig Falconer
Michael Driscoll (fen...@ulfheim.frob.ml.org) wrote:
: ObAProgrammingR[-1]:
: Did you ever spend eight hours working in a deeply nested function involving
: many fork()'s, naively trying to figure out just where the fsck everything
: is going wrong, only to discover that you are accidently doing an fclose()
: on the read pointer instead of the write pointer?
: *weep*
: Go ahead, call me a luser. I already have, after all.
: Didn't help that this was an FCGI program taking some rather nonstandard
: input from the server[0] so that gdbm/strace and such couldn't be used,
: and so I had to do all of the diagnosis via syslog().
I can top that - how about spending six hours on an assembly program,
wondering why it kept giving either wrong answers or over/underflowing.
Turns out that sll and srl are not the same (shift left logic btw)
--
--
Criggie the Wierd! "Q: How do you make a fruit cordial?"
`'`'`'`'`'`'`'`'`' "A: Or avpr gb zr"
- Unsolicited email costs you $100 NZ per line -
Kai Henningsen
cs...@cantua.canterbury.ac.nz (Craig Falconer) wrote on 17.01.98 in <69pl96$s81$2...@cantuc.canterbury.ac.nz>:
> Michael Driscoll (fen...@ulfheim.frob.ml.org) wrote:
> : ObAProgrammingR[-1]:
> : Did you ever spend eight hours working in a deeply nested function
[...]
> I can top that - how about spending six hours on an assembly program,
[...]
Luser programming mistake dicksize war?
How about taking two fscking years to find a problem? (Well, one after the
first reports.) A flag was set and queried in all the right places. It
just wasn't reset, ever.
The problem here was that this (per-user) flag was set (and should have
been reset) once per year. It controlled sending out reminders about
account renewal.
When it was new, all was fine, of course.
A year later, some people complained they didn't get any reminders. Check
it out - make account, make nearly-run-out, see reminders being sent.
Diagnoze as pilot error (people just didn't notice the mail somehow).
Another year later, another slew of complaints. Look extra hard at source.
Umm, ok, so here it gets set, here it's being queried ... just a moment,
where is it being reset? Pull out grep. Still nothing. Pull out older
versions. Still nothing. Aaaaaarrghh!!! What a *stupid* mistake.
Tanuki the Raccoon-dog
In article <6m8vC...@khms.westfalen.de>, Kai Henningsen
<kaih=6m8vC...@khms.westfalen.de> writes
>cs...@cantua.canterbury.ac.nz (Craig Falconer) wrote on 17.01.98 in
><69pl96$s81
>$2...@cantuc.canterbury.ac.nz>:
>
>> Michael Driscoll (fen...@ulfheim.frob.ml.org) wrote:
>> : ObAProgrammingR[-1]:
>> : Did you ever spend eight hours working in a deeply nested function
>[...]
>
>> I can top that - how about spending six hours on an assembly program,
>[...]
>
>Luser programming mistake dicksize war?
>
>How about taking two fscking years to find a problem? (Well, one after the
>first reports.) A flag was set and queried in all the right places. It
>just wasn't reset, ever.
I once did something that was sorta the inverse of this: testing the
status of a flag, but never setting it! Of course the outcome of the
test is still dependent on a range of externally mediated factors[1]
but not the one you intended.
[1]The luser's grandmother's star-sign, time of the month [if female],
how many cups of coffee the operators have had that day, phase of the
moon, and whether fish is on the menu at the local cafe.
__________________________________________________________________________
!Raised Tails! -:Tanuki:-
http://www.canismajor.demon.co.uk/index.html
"The meek may inherit the earth, but they don't get write permission"
Rackmount
Craig Falconer <cri...@papanui.school.nz> wrote:
: I can top that - how about spending six hours on an assembly program,
: wondering why it kept giving either wrong answers or over/underflowing.
Here's mine:
I spent 4 days trying to get an NTWS client to connect to an office LAN
server... turned out that the cabling (which I had crimped myself), was
crimped backwards on one end... and thusly... I LARTed myself quite
thoroughly by reading a page out of "Networking for Idiots", hitting
myself in the head with the book, then reading another page, etc.
--
QUOTE: <No words of wisdom for today, kiddies.>
Stephen S. Edwards II -- r a k m o u n t @ p r i m e n e t . c o m
Homepage: http://www.primenet.com/~rakmount (under construction)
Pub FTP: ftp://ftp.primenet.com/users/p/rakmount
Mike Whitaker
In article <MpuNxKAT...@canismajor.demon.co.uk>, Tanuki the
Raccoon-dog <Tan...@canis-major.daemon.co.uk> wrote:
> I once did something that was sorta the inverse of this: testing the
> status of a flag, but never setting it! Of course the outcome of the
> test is still dependent on a range of externally mediated factors[1]
> but not the one you intended.
Been there, done that [2]. Worse still, it was not a flag but an error
code, and the random garbage it was pointing to consistently contained a
Highly Significant And Meaningful Value, thus hindering debugging while my
caffeiene starved brain figured out how the function being called was
returning said HS&MV without triggering the longjmp-out-and-panic-loudly
error-trapping macro it was wrapped with.
[1] NMF
[2] NMCode, actually, but the PHB he say feex heem....
--
Mike Whitaker, Email: mi...@altrion.org, WWW: http://www.altrion.org/
Unsolicited commercial email will be dealt with appropriately.
Matthew Crosby
In article <6a3bd4$a...@nntp02.primenet.com>,
Rackmount <HAL9000@USS_Discovery.gov> wrote:
>Craig Falconer <cri...@papanui.school.nz> wrote:
>
>: I can top that - how about spending six hours on an assembly program,
>: wondering why it kept giving either wrong answers or over/underflowing.
>
>Here's mine:
>
>I spent 4 days trying to get an NTWS client to connect to an office LAN
>server... turned out that the cabling (which I had crimped myself), was
>crimped backwards on one end... and thusly... I LARTed myself quite
>thoroughly by reading a page out of "Networking for Idiots", hitting
>myself in the head with the book, then reading another page, etc.
Back when I worked for the University, we once had a 10baseT cable
making party, which ended up due to bogus instructions making 120 patch
cables wired with split pairs.
Lots of embarrassment all around.
--
Matthew Crosby cro...@cs.colorado.edu
Disclaimer: It was in another country, and besides, the wench is dead.
Frank Lee
On 21 Jan 1998, Matthew Crosby wrote:
> Back when I worked for the University, we once had a 10baseT cable
> making party, which ended up due to bogus instructions making 120 patch
> cables wired with split pairs.
> Lots of embarrassment all around.
plugs on (checking each others), two others putting the sockets on
(checking each others). All good, error correcting stuff. Apart from the
fact that the socket guys thought the wiring code went:
blue - neutral
brown - eath
green / yellow - live
[2]
Luckily the theatre had RCD circuit breakers...
Frank
[0] Theatres use a _lot_ of mains cable - we used to get 1000' on a monday
and have converted it into 30' cables by Wednesday (i.e. three
lunchtimes).
[1] Well, I did one of the sessions.[3]
[2] N.B - this code _isn't_ the right one. The right one is classified
'useful'...
[3] I know the wiring code, so was overqualified.
--
Frank Lee,
Computing Officer, Sidney Sussex College, Cambridge.
(fr...@sid.cam.ac.uk)
Usenet postings: Remove NOSPAM. from address
Peter Benie
In article <Pine.LNX.3.96.98012...@www.sid.cam.ac.uk>,
Frank Lee <fr...@sid.cam.ac.uk> wrote:
>Confirm-Reading-To: fr...@sid.cam.ac.uk
>Return-Receipt-To: fr...@sid.cam.ac.uk
Do you really want us to mail you after reading this?
>Rather like the mains cabling[0] session I had[1] - two of us putting the
>plugs on (checking each others), two others putting the sockets on
>(checking each others). All good, error correcting stuff. Apart from the
>fact that the socket guys thought the wiring code went:
>blue - neutral
>brown - eath
>green / yellow - live
>
>Luckily the theatre had RCD circuit breakers...
... which won't go off until somebody touches an 'earthed' component.
I do hope that it was the people wiring up the sockets that tested the
efficiency of the RCD.
My younger brother once wired up a plug before he was familiar with
the wiring code. Fortunately he had frayed bits of wire from the live
touching frayed bits of wire from the neutral, triggering the circuit
breaker before he injured himself.
Peter
Chris Richardson
<fr...@sid.cam.ac.uk>, the entity called Frank Lee , wrote:
<Various people discussing wiring cockups>
> <...> All good, error correcting stuff. Apart from the
> fact that the socket guys thought the wiring code went:
> blue - neutral
> brown - earth
> green / yellow - live
While we're on the subject, does anybody know which standards body is
responsible for this stupid colour coding system? I don't know about
anybody else, but for me, yellow and green is the "liveliest" of these
three, followed by the fairly exciting blue. Brown is in last place as
a dead, earthy colour.
In summary, brown as live is not intuitive. Whoever decided that it
was a good colour to associate with live ought to be shot.
Or is this just a cultural thing?
foop
--
foop (fu:p):n. [Pharmacy, King's College London] Departmental PostDoc Geek.
"<class> could actually declare <function> to be pure virtual again. Designs
like this require a note giving permission from your mother embedded in the
comments." -- _C++_The_Core_Language_, Gregory Satir & Doug Brown, ORA.
Nik Clayton
In article <6a7p89$stq$1...@sg4.pcy.kcl.ac.uk>, Chris Richardson wrote:
>In summary, brown as live is not intuitive. Whoever decided that it
>was a good colour to associate with live ought to be shot.
>
>Or is this just a cultural thing?
Dunno. I always remember it as "Touch the brown and you're in the shit".
Works for me.
N
--
--+==[ Nik Clayton is Just Another Perl Hacker at Interactive Investor ]==+--
. . . and relax
Peter Gutmann
fo...@sg4.pcy.kcl.ac.uk (Chris Richardson) writes:
><fr...@sid.cam.ac.uk>, the entity called Frank Lee , wrote:
> <Various people discussing wiring cockups>
>> <...> All good, error correcting stuff. Apart from the
>> fact that the socket guys thought the wiring code went:
>> blue - neutral
>> brown - earth
>> green / yellow - live
>While we're on the subject, does anybody know which standards body is
>responsible for this stupid colour coding system? I don't know about
>anybody else, but for me, yellow and green is the "liveliest" of these
>three, followed by the fairly exciting blue. Brown is in last place as
>a dead, earthy colour.
>In summary, brown as live is not intuitive. Whoever decided that it
>was a good colour to associate with live ought to be shot.
>Or is this just a cultural thing?
AFAIK it's a colourblindness-proof thing, the colours are the combination
which are least likely to result in confusion due to various types of
colour-blindness.
Peter.
Frank Lee
On 22 Jan 1998, Peter Benie wrote:
> In article <Pine.LNX.3.96.98012...@www.sid.cam.ac.uk>,
> Frank Lee <fr...@sid.cam.ac.uk> wrote:
> >Confirm-Reading-To: fr...@sid.cam.ac.uk
> >Return-Receipt-To: fr...@sid.cam.ac.uk
yesterday too... Fits in well with the subject, though.
Consider LART applied
Frank
Alistair J. R. Young
On 22 Jan 1998 15:39:53 -0000, in message <6a7p89$stq$1...@sg4.pcy.kcl.ac.uk>,
Chris Richardson <fo...@sg4.pcy.kcl.ac.uk> (== foop)
praised Shub-Internet thus:
> In summary, brown as live is not intuitive. Whoever decided that it
> was a good colour to associate with live ought to be shot.
> Or is this just a cultural thing?
No, probably not.
What makes it particularly idiotic, IMO, is that they changed it to
this from the perfectly intuitive red-black-green scheme [1].
Alistair
[1] Well, yes, maybe not if you were colour blind. But
red-black-green/yellow would have done just as well.
--
Computational Thaumaturge -- Sysimperator, dominus retis deusque machinarum.
e-mail: avata...@arkane.demon.co.uk WWW: http://www.arkane.demon.co.uk/
"The number of sysadmins who will decline to carry the group if properly voted
on and newgrouped by Tale is precisely zero. Well, maybe 1, but Barry's always
been a little strange." -- Lee Daniel Crocker
Kai Harrekilde-Petersen
longword@!spam.super.zippo.com (Lionel Lauer) writes:
> Quoth mi...@altrion.org (Mike Whitaker) :
> >
> >Been there, done that [2]. Worse still, it was not a flag but an error
> >code, and the random garbage it was pointing to consistently contained a
> >Highly Significant And Meaningful Value, thus hindering debugging while my
>
69, of course.
--Kai
Mike Whitaker
In article <34d09c2f...@snews.zippo.com>, Lionel Lauer wrote:
>Quoth mi...@altrion.org (Mike Whitaker) :
>>
>>Been there, done that [2]. Worse still, it was not a flag but an error
>>code, and the random garbage it was pointing to consistently contained a
>>Highly Significant And Meaningful Value, thus hindering debugging while my
>
>What, 42?
22. Happened to mean 'invalid entity tag'.
The subset of people for which this is actually useful information
is not *terribly* large.
Mike Whitaker
In article <slrn6cf21...@carrig.strand.iii.co.uk>, Nik Clayton wrote:
>In article <6a7p89$stq$1...@sg4.pcy.kcl.ac.uk>, Chris Richardson wrote:
>>was a good colour to associate with live ought to be shot.
>>
>>Or is this just a cultural thing?
>
>
A magazine in the UK a long time ago ran a short set of letters on
mnemonics for this. The two that stick in my mind were one which
is convoluted and warped (like my mind, I guess) and the one I actually
used. Both have the dubious advantage of not requiring you to know that the
live pin is the one with the fuse attached, but knowing left from right[2]
prevents a seftLARTing.
The warped one ran:
Yellow/Green has as first letters YG, which are the initial letters of
YGgdrasil[1], the tree in Norse myth that linked heaven and EARTH.
Brown shoes are not the RIGHT wear to attend a funeral in.
When you subtract yellow from green, blue is LEFT.
I don't use it, but it was wierd enough to stick in what passes for a mind.
The one I used up to the point where it just got second nature:
bLue Left, bRown Right, green-and-yellow middle.
[1] almost certainly mispelt, but I'm sure someone will correct me
[2] I note that IEC line sockets have little earth and live symbols by the
cable grips: but then, they're designed for people with clues.
--
Mike Whitaker: Sysadmin, Entropic Cambridge Research Labs
Alister
m...@malay.entropic.co.uk (Mike Whitaker) wrote:
>Nik Clayton wrote:
[Electrical plugs]
>>Dunno. I always remember it as "Touch the brown and you're in the shit".
>A magazine in the UK a long time ago ran a short set of letters on
>mnemonics for this.
Imagine the hours of fun (in your 'copious free time') of figuring out
what original sequences people remember with various mnemonics and so
on. Like the (pretty easy)
'All presentable session Trombonists never die physically'
Alister
--
Probabilities are that you are more likely to be killed by a mass
extinction caused by an asteroid/comet strike than an airplane crash.
(Y2k excepted), Sleep well, and don't worry there are Millions^W
Thousands^W Hundreds^W about a dozen people looking for them.
Lionel Lauer
Quoth fo...@sg4.pcy.kcl.ac.uk (Chris Richardson) :
><fr...@sid.cam.ac.uk>, the entity called Frank Lee , wrote:
>
> <Various people discussing wiring cockups>
>
>> <...> All good, error correcting stuff. Apart from the
>> fact that the socket guys thought the wiring code went:
>> blue - neutral
>> brown - earth
>> green / yellow - live
>
>While we're on the subject, does anybody know which standards body is
>responsible for this stupid colour coding system? I don't know about
I don't remember exactly which standards body made the final call, but
it was one of the international ones. ISTR it being ISO, but I wouldn't
swear to it.
>anybody else, but for me, yellow and green is the "liveliest" of these
>three, followed by the fairly exciting blue. Brown is in last place as
>a dead, earthy colour.
This very topic pops up in the trade press over here every now & then,
always with much lively debate.
The original system was:
Red = active,
Black = neutral,
Green = earth.
There were a number of other colouring schemes around, which created
enough ambiguity to cause problems.
A major reason behind the choice of colours was the danger of people
(such as myself) with some degree of red/green colour blindness (the
most common variety, posessed by a large percentage of the male
population), confusing red & green in poorly lit environments, such as
in ceilings, or under buildings etc.
>In summary, brown as live is not intuitive. Whoever decided that it
>was a good colour to associate with live ought to be shot.
You could try a mnemonic such as: "Active = brown, because it's the one
that'll make you shit yourself if you do the wrong thing with it"...
Lionel.
--
W Lionel Lauer - longword@*fnord*.super.zippo.com McQ
. | ,. w , "Some people are alive only because
\|/ \|/ it is illegal to kill them." Perna condita delenda est
---^----^---------------------------------------------------------------
Lionel Lauer
Quoth baratz@/dev/null-cybercom.net (Josh Baratz) :
>Still, It's not as bad as the US scheme.
>Black == hot
>White == neutral
>Green == ground.
Bloody hell! - I've always wondered why most PSUs use that weird colour
scheme. It's confusing as all hell when you're used to the international
system.
>Really fun when you've been working of DC electronics all day then go
>to wire stuff.[1]
I'll bet. :(
Tim Pailthorpe
Peter Gutmann wrote:
>
> AFAIK it's a colourblindness-proof thing, the colours are the combination
> which are least likely to result in confusion due to various types of
> colour-blindness.
Let the f**kers fry, genetic failures that they are ;)
p.s. I know a couple (a bloke and a lass) who are both colour blind. They
are both into DIY. They can paint a house in completely f**ked up colours
but you cann ot argue with them when you say that it doesn't match.
Show them a green and red room and they claim it is 2 tone grey.
Francesco Benvenuto
On Thu, 29 Jan 1998 17:28:27 +0000,
in alt.sysadmin.recovery,
Tim Pailthorpe <Tim@no_spam.hirst.pass.theplanet.co.uk> wrote:
plink, plink... is it raining? Or is my sarcasm detector broken again?
> X-Mailer: Mozilla 4.04 [en] (Win95; I)
No, I think it's raining.
There are legitimate reasons for posting from Win95 with Mozilla, but
those usually involve being strained on a desert island. Or trolling.
>p.s. I know a couple (a bloke and a lass) who are both colour blind. They
>are both into DIY. They can paint a house in completely f**ked up colours
>but you cann ot argue with them when you say that it doesn't match.
>Show them a green and red room and they claim it is 2 tone grey.
It really is 2 tone grey, to them.
Arguing over perceptions is like arguing over postulates, usually
pointless, fun at times, but boring most of the time.
One could indeed argue that the human colou*r perception is flawed in
general, but why?
Of course, if that were _your_ house, it'd be your right to complain.
Loudly.
--
fB
Tony Finch
neuhaus@-nOsPaM-assenmacher.com (Stephan Neuhaus) wrote:
>imho+...@fis.unico.it (Francesco Benvenuto) wrote:
>> One could indeed argue that the human colou*r perception is flawed in
>> general, but why?
>
>ITIYM "colou?r" HTH. HAND.
></PEDANTIC>
BUGS
If there was a single standard for the English language it
would not be necessary to support redundant spellings.
Tony.
--
F.A.N.Finch ::0:0::: <d...@dotat.at>
System Administrator <fa...@demon.net>
Demon Internet [speaking in a private capacity]
Tanuki the Raccoon-dog
In article <6b79kl$ndf$2...@info.noc.demon.net>, Tony Finch <d...@dotat.at>
writes
>neuhaus@-nOsPaM-assenmacher.com (Stephan Neuhaus) wrote:
>>imho+...@fis.unico.it (Francesco Benvenuto) wrote:
>>> One could indeed argue that the human colou*r perception is flawed in
>>> general, but why?
>>
>><PEDANTIC>
>>ITIYM "colou?r" HTH. HAND.
>></PEDANTIC>
>
>BUGS
> If there was a single standard for the English language it
> would not be necessary to support redundant spellings.
Surely, there *is* a "single standard for the English language".
And it's not the widely-spoken ANSI variant...
__________________________________________________________________________
!Raised Tails! -:Tanuki:-
http://www.canismajor.demon.co.uk/index.html
"Licenced speaker of the Queen's English"
Alistair J. R. Young
On 3 Feb 1998 14:29:41 GMT, in message <6b79kl$ndf$2...@info.noc.demon.net>,
Tony Finch <d...@dotat.at> (== dot)
praised Shub-Internet thus:
> neuhaus@-nOsPaM-assenmacher.com (Stephan Neuhaus) wrote:
>> imho+...@fis.unico.it (Francesco Benvenuto) wrote:
>>> One could indeed argue that the human colou*r perception is flawed in
>>> general, but why?
>>
>> <PEDANTIC>
>> ITIYM "colou?r" HTH. HAND.
>> </PEDANTIC>
> BUGS
> If there was a single standard for the English language it
> would not be necessary to support redundant spellings.
ObQueen'sEnglish: There *is* a single standard for the English
language, it's just that most of the implementations suck.
or, if you like
British English : American English :: BSD : System V ?
Swap those last two over to suit your own prejudices, of course.
Alistair
--
Computational Thaumaturge -- Sysimperator, dominus retis deusque machinarum.
e-mail: avata...@arkane.demon.co.uk WWW: http://www.arkane.demon.co.uk/
"Usenet is defined as all sites receiving newsgroup net.announce."
-- Bill and Karen Shannon, Usenet Logical Map, July 24, 1984
Joel Herda
Oh yow!
In article <y1KB.3$CE2....@typhoon.texas.net>, afc...@texas.net (Al
Castanoli) wrote:
>In article <OBP10kyI...@fis.unico.it>,
>Francesco Benvenuto <frances...@fis.unico.it> wrote:
>
>[...]
>
>>There are legitimate reasons for posting from Win95 with Mozilla, but
>>those usually involve being strained on a desert island. Or trolling.
Oh yow, indeed!
Does this mean we can now ship the Win95 lusers to desert islands to be fed
through strainers? Are they extra-large custom-built strainers, or do we
have to compact them down to fit through tea-strainers?
joel
--
Joel Herda sysadmin-biker-skum 1983 Suzuki GS1100GL
jjo...@tiac.net DoD#2053 1995 Neon Sport Coupe DOHC
remove the leading j from my address to have email get through
Vadim Vygonets
> BUGS
> If there was a single standard for the English language it
> would not be necessary to support redundant spellings.
Hint: POSIX.
Vadik.
--
Vadim Vygonets| If you think C++ is not overly complicated, just
Unix sysadmin | what is a protected abstract virtual base pure
spam filtered | virtual private destructor, and when was the last
with MAPS RBL | time you needed one? -- Tom Cargil, C++ Journal.