|
Grex > Helpers > #48: System Problems Item. Please attempt to rhyme where possible. |  |
|
| Author |
Message |
| 25 new of 231 responses total. |
popcorn
|
|
response 125 of 231:
|
Nov 3 20:48 UTC 1996 |
Hm. Telnetd(?) keeps the previous person's name attached to the pty until
a new person logs in. I wonder if idled is blaming the new person for the
old person's idle time. Weird.
Re 116: Rooby, I think "fack" in this case meant "fact".
Re 119: Rob, I'm not sure what you're saying here.
(Sorry for not rhyming).
|
ajax
|
|
response 126 of 231:
|
Nov 4 18:21 UTC 1996 |
Re 125, 119 was in response to 118: I've gotten regular "terminated:
segmentation" errors in Picospan for a very long time, probably about
once a day. I've reported it before, and at one point logged a few
and mailed them to STeve, at his suggestion, but haven't heard of any
solution. Here are some examples from that log:
-----
Respond or pass? forget
terminated: segmentation
Respond or pass?
-----
Respond or pass?
Ok:j hardware
terminated: segmentation
Ok:
-----
Respond or pass?
Ok:n
No new items in micros
No new items in aaypsi
terminated: segmentation
Ok:
------
You have mail.
Ok:r
terminated: segmentation
Pipe interrupt?
Respond or pass?
-------
Ok:r
terminated: segmentation
Respond or pass?
-----
Respond or pass?
terminated: segmentation
Pipe interrupt?
Respond or pass?
----
|
remmers
|
|
response 127 of 231:
|
Nov 4 19:47 UTC 1996 |
Hmm. I seldom get such messages and suspect it has something to
with how you set things up in your .cfonce file (which I can't
look at right at the moment since I'm not logged into Grex).
From the placement of the messages it looks like what's termina-
ting with segmentation errors is your pager.
|
popcorn
|
|
response 128 of 231:
|
Nov 4 20:26 UTC 1996 |
Ja, that was my thought too.
It's the pager that's making Rob blue.
Here's Rob's pager:
define pager "less -dE -P 'Press Spacebar for more, q to quit'"
And here's the standard pager newuser has been giving to everybody for eons:
define pager "less -dE -P 'Press Spacebar for more, q to quit'"
Oh oh: No difference.
|
davel
|
|
response 129 of 231:
|
Nov 5 11:46 UTC 1996 |
My pager is more (with twit filter piped into it), not less. I know that if
I do an interrupt (^C) instead of a q, it sometimes core-dumps. (But this
probably isn't relevant, since Rob's using less.)
I see that Rob has an 8.5 MB core file, but it doesn't look like anything
related to this problem, at a quick glance ... but someone better at that
kind of thing might look at it.
|
remmers
|
|
response 130 of 231:
|
Nov 5 14:32 UTC 1996 |
The core file is sparse, I'd guess,
Hence not really 8.5MB, but less,
For its true size, use "ls -s".
|
kerouac
|
|
response 131 of 231:
|
Nov 5 23:10 UTC 1996 |
One of the roots on this system wiped out my coop participation
file so that I no longer have any bookmarks there. I have no doubt
this was done out of spite and Iwould like to know which of you did it.
Grex is a great place withlots of great confs, and it just disheartends
me to find that coop is not one of those confs. I try very hard
to contribute there and because I take the time to respondin thebest
way that I can, and am an outsider, I get ostracized. If the
staff wants more regular users to use coop, they should be advised
to accept those people as they areand not prejudge theirmotives.
|
kerouac
|
|
response 132 of 231:
|
Nov 5 23:26 UTC 1996 |
I'd like to apologize for the accusation I just made. I wasnt
at allaware that the system hadjust crashed, and I was reacting
to the fact that it was my coop file that was deleted, the one
conf where I've taken a lot of heat lately. It was a bad
coincidence is all.
/.
|
e4808mc
|
|
response 133 of 231:
|
Nov 6 01:13 UTC 1996 |
I too am an outsider and a newcomer to Grex, and I will say publically what
I have said in private email to several people: kerouac's responses and his
ill mannered tone are a real turnoff to new users.
|
arthurp
|
|
response 134 of 231:
|
Nov 6 02:01 UTC 1996 |
I know most of the staff very well, and they would never do anything like
that. It greatly disheartens me that anyone would make such an accusation.
Especially with nothing more than internal emotions to back it up. I was
greatly offended on behalf of everyone who donates so much time to keep grex
alive.
|
jor
|
|
response 135 of 231:
|
Nov 6 02:11 UTC 1996 |
I have lost participation files, it can really frost your shorts.
|
arthurp
|
|
response 136 of 231:
|
Nov 6 02:30 UTC 1996 |
I've lost them too, but I know better than to respond with groundless
accusations.
|
janc
|
|
response 137 of 231:
|
Nov 6 05:27 UTC 1996 |
I thank Kerouac for his thoughtful apology. It's always tempting to blame
random bad luck on enemy action. We all make that mistake sometimes.
|
popcorn
|
|
response 138 of 231:
|
Nov 6 09:39 UTC 1996 |
Core file tip:
Type "file core" to find out what's in it.
(Hm. "tip" and "it" *almost* rhyme, if you pronounce "it" as "ipp".)
|
popcorn
|
|
response 139 of 231:
|
Nov 6 09:39 UTC 1996 |
(that is, "file core" will give you back a response like "core dump file from
lynx")
|
davel
|
|
response 140 of 231:
|
Nov 6 11:03 UTC 1996 |
I've had my coop participation file get trashed two or three times lately.
At least once it was apparently because I got interrupted & the modem
kicked me off while I wasn't looking.
Anyway, I finally wrote a little nawk program to rebuild my file. (You can
use FIXSEEN, but then whenever someone responds to an item it shows up as
brandnew. There are of course other ways to solve this problem. *Nothing*
except backing up your participation file regularly will take care of knowing
which items you'd forgotten ...)
|
popcorn
|
|
response 141 of 231:
|
Nov 6 19:42 UTC 1996 |
Odd... I did a "fixseen" on agora after I lost my participation file, and then
a "read since 10/25". Items that didn't have any responses in my "read
since", but which later got responses, are showing up correctly. That is,
I'm seeing only the new responses, not the entire item.
|
kerouac
|
|
response 142 of 231:
|
Nov 6 19:55 UTC 1996 |
This was actually the first time I can recall ever having a
participation file eaten. Guess I've been luckier than most.
|
scg
|
|
response 143 of 231:
|
Nov 6 19:58 UTC 1996 |
re 141:
It's supposed to work that way, isn't it?
|
tsty
|
|
response 144 of 231:
|
Nov 6 20:25 UTC 1996 |
question, perhaps a problem:
from the menu, !<alias> fails in many cases while exiting
the menu and then <alias> works in the intended way.
i was skeptical about that until i was at the source keyboard and
tested both mechanisms.
thoughts?
|
popcorn
|
|
response 145 of 231:
|
Nov 6 20:36 UTC 1996 |
Aliases shouldn't work at the menu, period.
|
scg
|
|
response 146 of 231:
|
Nov 6 20:44 UTC 1996 |
I don't think !<alias> works anywhere, generally.
|
tsty
|
|
response 147 of 231:
|
Nov 7 09:42 UTC 1996 |
okfine ... seemed reasonable to me, helped-person will just get
out of the menu first.
as for the participation files.. if you hvaen't had a couple munched,
you haven't grexed right .... <g>
|
draven
|
|
response 148 of 231:
|
Nov 13 15:45 UTC 1996 |
#144 (tsty):
The aliases in question are defined in the .cshrc. Only csh and tcsh
load that file. The menu system is an sh shell script and sh doesn't
understand aliases defined in the .cshrc and Grex's version of sh doesn't
support aliases at all.
So, when you're in menu, the !<command> must follow sh rules, which
means no aliases. When you leave menu, the system drops back to your
login shell (csh) which loads and accepts all of your aliases.
|
drew
|
|
response 149 of 231:
|
Nov 14 12:59 UTC 1996 |
Mail from ford.com does not seem to be making it over.
|