|
|
| Author |
Message |
| 25 new of 211 responses total. |
dpc
|
|
response 150 of 211:
| 
May 23 14:48 UTC 1997 |
So basically we've got 13000 accounts which have seen some activity
in the past 90 days. Wow!
Do we have *any* way of telling how many of these accounts have
been used in the conferencing system?
|
senna
|
|
response 151 of 211:
|
May 23 15:51 UTC 1997 |
UIDs?
That's a nice total. Does anyone know what grex's theoretical maximum login
limit it, based on the 8 characters in a login?
|
valerie
|
|
response 152 of 211:
|
May 23 16:35 UTC 1997 |
This response has been erased.
|
rcurl
|
|
response 153 of 211:
|
May 23 17:31 UTC 1997 |
So, how many accounts have a .cf*? (Can a count be done automatically?).
|
arianna
|
|
response 154 of 211:
|
May 23 17:41 UTC 1997 |
Um, I just tried that !menumore /a/s/t/steve/reap/immortals -- and it told
me "permission denied."
|
albaugh
|
|
response 155 of 211:
|
May 23 19:48 UTC 1997 |
FYI: Respond or pass? !fmt /a/s/t/steve/reap/immortal
fmt: /a/s/t/steve/reap/immortal: Permission denied
|
senna
|
|
response 156 of 211:
|
May 23 21:34 UTC 1997 |
and what, precisely, is UID? a slot for an individual login?
|
remmers
|
|
response 157 of 211:
|
May 24 01:37 UTC 1997 |
UID means "user identification number." It's the number by which
Unix identifies you internally.
You can find out what your UID is by looking at your entry in
the password file. One way to do that is to type
!grep '^senna:' /etc/passwd
(or whatever login id you're interest in, in place of senna).
The UID is the number right after the second colon. Mine is 121.
|
senna
|
|
response 158 of 211:
|
May 24 04:59 UTC 1997 |
can the UID limit be exspanded, or is it hardwired into the system?
|
tpryan
|
|
response 159 of 211:
|
May 25 16:18 UTC 1997 |
Seems you would need to look for recent activity on those
*.cf files, in case users have themselves started up in agora without
realy using it.
or the other conferences.
|
mcnally
|
|
response 160 of 211:
|
May 25 20:02 UTC 1997 |
re #158: the UID limit can't be increased without major trouble
(involving, at the least, major modifications to the kernel and
the system libraries and recompiling every program that checks UID
for any reason (which is an awful lot of programs..)) the problem
is that all of the above are now compiled to treat the UID as an
unsigned 16 bit number and programs that check UIDs only allot 16 bits
of space to store the values. more important than the fact that
it's a lot of work to track down all references and recompile
everything it's not even remotely possible to do without SunOS
source code (which is proprietary.)
If we were to switch to Sun's most recent OS version, Solaris 2.5.1,
we'd have more UID room but nobody'd like the number of other problems
we'd have instead.
|
tpryan
|
|
response 161 of 211:
|
May 26 02:33 UTC 1997 |
Would it be better to 'compress' the UID numbers by giving
everybody new UID numbers or start to re-assign unused UID's??
|
srw
|
|
response 162 of 211:
|
May 26 05:33 UTC 1997 |
Compressing would be a major hassle that would gain nothing.
It would be a hassle because the UID is stored in the file system to describe
who own every file. If you changed a person's UID, wyou would have to change
the owner of every file on the system.
It would gain nothing, because we already re-use old UIDs. We're on our second
pass through the 65000 UIds. We're only reassigning the ones that have
become free.
We don't assign the lowest available UID, because then the used ones would
bunch up at the beginning. We cycle through them because that way the ones
that we reassign are much less likely to have been freed recently.
|
valerie
|
|
response 163 of 211:
|
May 26 14:34 UTC 1997 |
This response has been erased.
|
valerie
|
|
response 164 of 211:
|
May 26 14:35 UTC 1997 |
This response has been erased.
|
senna
|
|
response 165 of 211:
|
May 27 04:27 UTC 1997 |
So, in other words grex will never have more than about 60,000 users, barring
an act of God.
Hmm, always wanted to know what groups I was posting to.
|
valerie
|
|
response 166 of 211:
|
May 27 18:41 UTC 1997 |
This response has been erased.
|
dpc
|
|
response 167 of 211:
|
May 27 21:39 UTC 1997 |
Thanx for the clue that you have to check people's home directories
for *.cf or .cfdir files.
I would *love* to do this, especially if that checking could
filter out everyone who has not had any changes in these files for
the past XX days.
Unfortunately, I am a mere "Babe in Unixland." What are the
commands that would run these counts automatically?
|
senna
|
|
response 168 of 211:
|
May 28 02:17 UTC 1997 |
I'd be really impressed if we grew that big, since we're essentially a local
system and the local market only has double the available number of UIDs
|
remmers
|
|
response 169 of 211:
|
May 28 02:44 UTC 1997 |
If you look at the geographical distribution of Grex users, I'm
not at all sure that you'd find it to be "essentially a local
system". At any given time, the vast majority of people logged
in appear to be non-local. Has anybody gathered any statistics
on where our users come from?
|
rcurl
|
|
response 170 of 211:
|
May 28 05:21 UTC 1997 |
Most of those we meet in conferences are local, not that I want it that way.
|
valerie
|
|
response 171 of 211:
|
May 28 14:47 UTC 1997 |
This response has been erased.
|
drew
|
|
response 172 of 211:
|
May 29 05:23 UTC 1997 |
Jared's POP3 rejection routine is working. (I tested it out with Eudora, which
does *not* ask for another password after the error message comes up.)
|
tsty
|
|
response 173 of 211:
|
May 29 10:38 UTC 1997 |
thankxxx jared.
|
kaplan
|
|
response 174 of 211:
|
May 29 12:55 UTC 1997 |
I tried to
telnet grex.org pop3
Trying 152.160.30.1 ...
telnet: connect: Connection refused
Has the fake POP server been killed?
|
