| WIKI | FAQ | Tech FAQ | |
 http://csua.com/feed/ | |
| ||||||
| 12/25 |
| 2007/5/16-19 [Academia/Berkeley/CSUA/Motd, Computer/SW/Unix] UID:46659 Activity:nil |
5/16 Gee, I wonder why the motd isn't popular with new CSUAers...
\_ There are too many reasons but for starter, the level of entry
is too high. Unless you breath UNIX and you use vi/emacs
like you use MS Word, then it's just too troublesome. Secondly,
there is simply very little useful information on motd these days.
There used to be quality tech Q&As and foodP/socialP posts but with
the amount of trashy political posts that only a few participate in
these days, motd has become more of a toxic dump for people who
want to vent than a useful place for people to get information.
\_ Who uses MS Word?
\_ Not enough talk about sex???
\_ The MOTD isn't popular with new CSUAers because it's a very silly
way to implement a message board. They've been using blogs and
online forums for years, why would they want to edit a text file
and worry about 80 columns and indenting and all that? Same with
wall. Even if everyone were blowing sunshine up each other's
butts all day, the undergrads wouldn't use motd and wall. -tom
\_ i didn't start reading the motd until after I graduated, when
i started doing sysadmin stuff as part of my job -'97 alum
\_ Let me guess, you breath UNIX and you're bored at the time? |
| 2007/5/16-17 [Computer/SW/Unix] UID:46655 Activity:nil |
5/15 i'm trying to create a webservice where people can upload multiple
doc or pdf files, and then be able to download one big pdf of them
all ... are there open-source / unix utiities i coudl use to
achieve this? thx
\_ I've done this in Java using an OSS library called PDFBox |
| 2007/5/5-7 [Computer/SW/Unix] UID:46533 Activity:low Cat_by:auto |
5/4 Question. Say I have dir1/A and dir2/A. How do I collapse them
together? I'm unable to do "mv dir1/A dir2" or "mv dir1/A/* dir2/A"
\- cd /foo/.../source;tar cf - . |(cd /bar/.../dest;tar xfBp -);rm -rf .
\_ cp dir1/A/* dir2/A; rm -rf dir1/A? What exactly are you trying to
do and why can't you use mv?
\- * globbing is risky with unknown inputs. when you want to
operate on dirs, dont operate on files. of course the
exact details depend on what policy you want to deal with
name collisions. you can use rsync too.
\_ of course. i still want to know what the op is really up to. |
| 2007/4/21-25 [Computer/SW/Unix] UID:46403 Activity:nil |
4/21 Root, can you turn finger back on so we can finger motd@csua?
\- "help me mconst, you're my only hope" |
| 2007/4/13 [Computer/SW/Languages, Computer/SW/Security, Computer/SW/Unix] UID:46294 Activity:nil |
4/13 Can someone w/ root fix this:
$ ls -l /dev/null
crw------- 1 root csua 1, 3 2007-01-25 19:41 /dev/null |
| 2007/4/7-10 [Computer/SW/Unix] UID:46230 Activity:nil |
4/7 Dear VMWare users. I just want to post this because I spent an
entire day trying to find out why my Ubuntu 2.6.18 has a serious
time drift problem on host WinXP Pro. The problem is that even
when you run vmware-tools and sync the time between your VM and
your host, there is still a delay-- vmware-tools syncs every once
in a while, and since my VM is so slow it doesn't catch up and
occassionally has a 2-3 min delay, enough to cause kerberos to
reject me, along with NFS problems. The solution is posted here:
http://kb.vmware.com/selfservice/microsites/search.do?cmd=displayKC&externalId=1420
You'd need to edit /boot/grub/menu.lst and add the changes for
"noapic nolapic". FYI ntp doesn't do jack-- either run ntp
exclusively (not ideal because of time drift), or vmware-tools
to sync your time. Good luck guys!
\_ I've got massive time drift but we don't use Kerberos so I just
smash a new time in manually every few months. It's a dev box
so it doesn't matter that it drifts a full day every month. What
you could do instead of ntp is do it the old fashioned way and put
rdate in cron every 5 minutes against something that keeps real
time. |
| 2007/3/28-31 [Computer/SW/Security, Computer/SW/Unix] UID:46132 Activity:nil |
3/28 What controls the order of files in regards to which file is displayed
as the root file of a webpage? Specifically, I have to have a
index.php in my root directory, but I want my webpage to display
home.php. How can I do this? Thanks.
\_ Just figured it out myself, using .htaccess! -op |
| 12/25 |
| 2007/3/23-27 [Computer/SW/Security, Computer/SW/Unix] UID:46068 Activity:nil |
3/23 hey root can you turn 'PINGS' to soda.csua back on?
thanks
\_ Hey, root, can you disable this h0zer's motd-editing cron-job pls?
\_ and what's up with crippling traceroute? It needs setuid to
function.
> traceroute scotch
traceroute: icmp socket: Operation not permitted |
| 2007/3/19-20 [Computer/SW/Unix] UID:46017 Activity:kinda low |
3/19 Debian Linux NFS client/server configuration for Dummies:
http://www.debianadmin.com/network-file-system-nfs-server-and-client-configuration-in-debian.html
In summary, on the NFS server (assume it is myserver), you do:
% apt-get install nfs-kernel-server nfs-common portmap
% vi /etc/exports
/home 195.12.32.2(rw,no_root_squash) <DEAD>www.first.com(ro<DEAD>
/usr 195.12.32.2/24(ro,insecure)
% exportfs -a
On the NFS client, you do:
% apt-get install nfs-common portmap
% mount myserver:/home /mnt/nfs
For a permanent mount you do:
% vi /etc/fstab
195.12.32.1:/home /home nfs rw,rsize=4096,wsize=4096,hard,intr,async,nodev,nosuid 0 0
195.12.32.2:/usr /usr nfs ro,rsize=8192,hard,intr,nfsvers=3,tcp,noatime,nodev,async 0 0
\- why would you be exporting /usr? 'cause you're a dummy?
\_ maybe for a diskless system? |
| 2007/3/16-20 [Computer/SW/Unix] UID:45996 Activity:nil |
3/16 Hey root, wanna reenable the finger interface to the motd?
\_ Try emailing them instead. The motd is a bad way to
communicate vital messages.
\_ I've sent a few messages regarding more serious things to no avail |
| 2007/3/15 [Computer/SW/Unix] UID:45975 Activity:nil |
3/14 Dear NFS gurus. I have a huge problem that I think may be related
to NFS. I have root on my machine and my /home/joe is mounted on NFS.
According to /etc/mtab it is mounted with "nw,nosuid,rize=8192,
wsize=8192,intr,bg". I am trying to setup suexec on apache2, but
for some reason it just wouldn't work when suexec is enabled. I'm
guessing the "nosuid" in nfs is interfering. I tried taking nosuid
in /etc/mtab and umount, but it keeps saying the device is busy.
What should I do?
\_ Update: I did a umount -l /directory (lazy) and it unmounted.
However I just found out my /home/* is on automount and I'm
unable to do anything in /etc/fstab. I think /etc/mtab is auto
generated, since editing it has absolutely no effect on
mount/unmount. Help! -joe
\_ you're right, mtab is auto generated.
nosuid isn't your problem unless your apache2 setup lives in
/home. You can override automounts with regular mounts.
\_ you should pay mconst $20 to fix and explain it all to you.
you have more "issues" than nfs. --nfs expert |
| 2007/3/11-12 [Computer/SW/Unix] UID:45929 Activity:moderate |
3/10 How does the unix cron implementation deal with daylight savings?
Officially we need to change the time at exactly 2:00am. So let's say
we have a cron job at 1:30am that has been executed, and at 2:00am
we turn it back to 1:00am. Will that particular job get executed
again? Conversely, let's say a cronjob at 2:30am was suppose to be
executed but when we spring forward to 3:00am, will that job be
ignored completely?
\_ This makes me ponder just what timezone the 'cron' daemon is
operating under. The system default? What if a user wanted to
'localize' their own crontab to their personal timezone?
\_ Quick googling shows that different UNIX handle them differently,
and even different distros handle them differently. For example,
IBM's AIX page says that jobs in ambiguous time will get run
twice and jobs in non-existent time will not get run. HP unix
doc says ambiguous time jobs will get run once only, and
non-existent time jobs will run once in the right time before the
DST change. GIYF (Google Is Your Friend). Soda MOTD is more
like your enemy, I guess. |
| 2007/3/9-12 [Computer/SW/Unix, Computer/SW/OS] UID:45916 Activity:nil |
3/9 > wc s2:
wc: s2:51796: Invalid or incomplete multibyte or wide character
\_ Short answer: "setenv LC_ALL en_US.iso88591".
Long answer: soda uses Unicode by default now, so a single character
can now be more than one byte. This means a lot of commands don't
quite work the way they used to -- here, wc is trying to count the
number of (Unicode) characters in your file, and it's complaining
because your file contains invalid Unicode byte sequences. If you
do want to use Unicode in general, but this particular file isn't
Unicode, you can either use "wc -c" for a byte count or set LC_CTYPE
to whatever the file's encoding actually is. Or, if you don't care
about Unicode and you just want all your old commands to work the
way they used to, the setenv command above will disable Unicode
completely and restore the traditional Unix behavior. --mconst |
| 2007/3/5 [Computer/SW/Unix] UID:45880 Activity:nil |
3/5 Root, kindly reinstate finger motd@csua?
\_ Have you tried emailing them? |
| 2007/2/23-26 [Computer/SW/Unix, Computer/SW/Editors] UID:45811 Activity:moderate |
2/23 Pnews - get error
"/bin/cat: /etc/news/organization: No such file or directory"
any ideas?
\_ I've created that file, and Pnews seems to be working now. --mconst
other error:
\_ /usr/bin/Pnews: line 448: /usr/local/bin/jove: No such file or
directory
\_ Ok, I added a symlink /usr/local/bin/jove. Could you
please let me know if that works for you? --mconst
\_ works.. thanks :)
\_ Why would that be necessary? Is that path hardcoded
somewhere or did he have that as his EDITOR or VISUAL?
(if he did it's stupid not to get it from PATH)
\_ It's not hardcoded; I assume the original poster just
had it as their $EDITOR. But there's no harm in having
/usr/local/bin/jove as a symlink, and it'll probably
fix similar problems for other people. --mconst |
| 2007/2/22-26 [Computer/SW/Unix] UID:45797 Activity:nil |
2/22 How come when I create a file in my home dir (e.g. /usr/bin/touch), the
timestamp of the file is always 1-2 min. ahead of the current time?
This doesn't happen when I create a file in /tmp. Thanks.
\_ its using the time of the nfs server, which is evidently out of
time sync with soda.
\_ soda's clock is almost 2 minutes behind right now.
ntp anyone? |
| 2007/2/19-23 [Computer/SW/Unix] UID:45773 Activity:nil |
2/19 Root, please re-enable finger motd@csua.berkeley.edu?
\_ don't forget to enable the motd fist functionality too. |
| 2007/2/18-23 [Computer/SW/Unix] UID:45769 Activity:nil |
2/17 Open source cola:
http://preview.tinyurl.com/ysvhn4 (everythingelse.wordpress.com)
link:preview.tinyurl.com/2c3y8e (upload.wikimedia.org)
\_ Yeah, I know some of the people that worked on opencola (the
software), who, technically, were the same people who worked on
open cola (the soda). They're cool people. -dans
\_ We should kill the uncool people.
\_ Uh huh. -dans |
| 2007/2/12 [Computer/SW/OS/Solaris, Computer/SW/Unix] UID:45719 Activity:nil 75%like:45716 50%like:45722 |
2/12 Trivial remote root exploit in Solaris 10 and 11. AWESOME!
http://isc.sans.org/diary.html?storyid=2220 -dans |
| 2007/2/12 [Computer/SW/Unix] UID:45716 Activity:nil 75%like:45719 |
2/12 Solaris 10/11 trivial remote root hole. AWESOME!
http://isc.sans.org/diary.html?storyid=2220 -dans |
| 2007/2/2-3 [Computer/Networking, Computer/SW/Unix] UID:45642 Activity:very high |
2/2 So is there some serious problem with gigabit? I keep seeing problems
with gigabit devices in 10/100 networks, or vice versa.
\_ do you think gigabit is some exotic, bleeding edge technology
just a few people are using?
\_ All I know is that when I use it in mixed networks, I keep
running into problems.
running into problems. -op
\- there are some cases where you can get unepxected
measurements like a 10mbit thru path that might be
"faster" than a 10mbit-100-10 path for reasons i wont
go into, but if you are getting dramatic failures or
really bad performance, it's probably pilot error or
something out of control somewhere.
\_ "pilot" error? What, you mean I made a mistake when
plugging the wires into the switches? -op
\_ you know there is a lot to configuring switches and
routers. duplex negotiation, potentially how mcast
signaling is handled etc. a colleague of mine found
a piece of networking gear flattended because of a
but in IGRP handling. from the nature of your
description of the problem ["is there a *problem*
with gigabit" without mention is the problem,
with thruput, latency, packet loss, ethernet frame
errors, high cpu load] it seems unlikely anybody
will be able to talk you thru it. your brain has been
probabilistically estimated as: small.
\_ I've tried two different switches on my home network
which was working great with 10/100 devices
(including my router). I added a gigibit device
which failed to simply plug in to my 10/100 router
and work. It worked if I plugged it into a
10/100/1000 switch, but the throughput crashed.
Everything is set to autonegotiate. I've tried this
with two different switches.
Then one of my coworkers got back from a customer
site where the networking wasn't working--using a
10/100 card instead of a 10/100/1000 card solved the
problem. -op
\_ i use gb networking "everyday" for at least
3-4 years. i have some number of problems now
and then but the technology obviously is
solid ... which is not to say there might not
be some porrly designed equipment in the
space, or some particular pieces might be
lemons. we're doing more 10gig work now ...
this is actually kinda interesting. this may
be the first time in a while when the typical
fast computer cannot saturate the link [obviously
hw traffic generators and special hackery can].
note: i dont know anything about 'home quality'
gb networking equipment.
\_ I have never seen any problems. I think it is you.
\_ Oh sure, I'm the only guy with a netgear gs105 switch and two
different gigabit cards. -op
\_ It may be that the cables you're using aren't up to snuff. See
if netstat shows a lot of packet errors.
\_ If so, why do the exact same cables work great when they're
transferring only gigabit traffic? -op
\_ if you want any more help you need to systematically
describe the symptoms and the specifics of your set up.
you can leave out the details and just say "i'm having
problems". come on, dont be a moron.
\_ you realize negotiating duplex setting and link speed
are two different things, right? if you want any more
help you need to systematically descibe the symptoms and
the specifics of your set up. you can leave out the details
and just say "i'm having problems". come on dood, dont be a
moron. |
| 2007/1/31-2/6 [Computer/SW/Unix] UID:45628 Activity:nil |
1/31 Root, could you turn the finger interface to the motd back on? |
| 2007/1/28-2/1 [Computer/SW/Security, Computer/SW/Unix] UID:45607 Activity:moderate |
1/28 Where does inbound mail get spooled now? I had no problem moving
my old spool to /var/mail/$USER, but where is the new mail
spooling? (Yes, I read soda-changes.) Nothing is ending up in
/var/mail/$USER/new. Should it be?
\_ Do you have .procmailrc setup? If so, I needed to add
an additional rule at the end (after setting up the
DEFAULT=/var/mail/loginname):
DEFAULT=/var/mail/$USER):
:0:
$DEFAULT
If I didn't do that, my mail just shows up as a single
"msg.xxx" in the /var/mail/loginname directory.
"msg.xxx" in the /var/mail/$USER directory.
\_ nope, it's on a different server normal people don't have access
to. it's called 'seperation of services'. The next time
someone breaks into http://soda.csua.berkeley.edu, mail will
continue to be delivered since they can't break into
the machine that is handling delivery of email. Were you
in CS? Do you remember how all the instructional machines
didn't store your email on your local machine? Same theory.
\_ So I have to use IMAP or POP now? Is that right? I used to
use UCB mail.
\_ It certainly wasn't stored *locally* on every machine, but
it was available via NFS on instructional machines.
It looks to me like it /should/ be showing up on soda. I presume
'mead-mail' is where it's getting delivered-to on mead anyways.
lrwxrwxrwx 1 root root 22 2007-01-24 00:58 /var/mail -> /mnt/oh/0X0-mead-mail/
lrwxrwxrwx 1 root root 22 2007-01-24 00:58 /var/mail ->
/mnt/oh/0X0-mead-mail/
\_ Absent procmail interference, mail should spool to the Maildir under
/var/mail/$USER. If that's not happening for you, something is wrong.
/var/mail/$USER. If that's not happening for you, something is
wrong. |
| 2007/1/26-30 [Computer/SW/Security, Computer/SW/Unix] UID:45598 Activity:nil |
1/26 Thanks root people!
\_ Many thanks! |
| 2006/12/30-2007/1/16 [Academia/Berkeley/Classes, Computer/SW/Unix] UID:45517 Activity:moderate |
12/30 Of the current EE/CS curriculum, what percentage of it is still
taught using *nix instead of Windows? I'm in the ASIC design
industry and everytime we get new college grads, we have to
spend some time teaching them some pretty basic Unix stuff.
It's, frankly, kind of annoying when we're teaching them that
stuff and not more important things.
\_ For ASIC design in particular, you're screwed. 150 and 152 are both
done solely in Winders. As for the rest of the EECS curriculum,
there's a good percentage that's Java, 188 is moving to Python, EE20
and all the Matlab stuff is on Winders machines, 184 is OS and
lanugage agnostic -- only EE122 have I seen recently as being
quote-unquote fully *NIX-oriented.
The long or the short of it is, I made it through the lower divs
without expanding my *NIX knowledge, and sitting inside a Winders
box. It was the CSUA and some of my pet projects in the dorms
(Linux-box server/NAT) that taught me anything. I think it's
certainly possible for a good percentage of kids these days to get
by without *actually* learning *NIX. Which is a damn shame.
As an aside -- the side of the CSUA soda-hozers don't see, we're
plotting for something tenatively named "Bear-bones" CS -- aka a
crash course in the good life (Unix, shells, emacs/vi, maybe a
little scripting) at the start of this semester --michener
\_ chuckle |
| 2006/12/1-8 [Computer/SW/Languages/Misc, Computer/SW/Unix, Academia/Berkeley/CSUA/Motd] UID:45408 Activity:nil |
11/30 For some reason after soda rebooted the motd defaulted to the Debian
motd. I've fixed the motd stuffs. I really don't know what is going
on with NFS locking. Just as an FYI, mail is local and home dirs are
remote mounted, so any locking issues would be if you have a procmail
or similar script handling your mail. - jvarga
\_ I removed my .forward, deleted a lock file, and it just
reappears. Mail sent to myself still apparently goes to
/dev/null. -asb
\_ mrauser and I are having a horrible time figuring out what is
going on here. root@ is not getting delivered and you say your
mail is being dropped on the floor. Ugh! - jvarga
\_ It seems that most mail that seemed to be going nowhere has
now been delivered. Thanks for fixing everything!! -asb |
| 2006/11/30-12/8 [Computer/SW/Security, Computer/SW/Unix] UID:45402 Activity:nil |
11/29 Pathetic Google engineers:
http://valleywag.com/tech/revisit/man-in-google-lap-pool-217775.php
http://www.valleywag.com/tech/dating/another-chance-to-crash-googles-holiday-party-217736.php |
| 2006/11/30-12/8 [Computer/SW/Editors/Vi, Computer/SW/Unix] UID:45399 Activity:nil |
11/30 What's going on since about 7:45am? Mail is no longer being
received. There is a ~/Mail/backup.lock file that keep being
recreated even after I keep deleting it. And, until right now,
my session would freeze up upon trying to vi a file. Disabling
procmail by removing the .forward file didn't seem to help fix
the lockfile issue, as I thought that was relating to procmail.
\_ What's REALLY wrong is no one (except for jvarga and other
really busy alumni) is stepping in to do the admin work.
Most alumni have grown up and have other things to do,
soda UNIX awareness is decreasing, active csua membership
has been dwindling for many years, and the students
don't have need to learn or admin esoteric UNIX boxes.
Until this trend reverses, admin-less soda will continue
to be hacked, defaced, and eventually die out. No one like
bad news but this is reality we have to face.
\_ whine whine whine, anyway I don't think politburo/root
is going to give root to any random alumni who shows
up at 343 Soda to "help out", are they? If so,
sign me up!
\_ dude, does anyone even use soda for anything important???
when mail goes down for DAYS and we bounce mail, who the
hell is dumb enough to use soda for anything? no one uses
soda anyway nowadays. just kill soda already. jvarga,
you did good, man. -linxu
\_ I still use soda. I use it to share my mp3 and my
private pictures of my geeky self, like you.
BTW I love your site and your pictures and mp3.
Could you please put some warez while you're at it?
http://csua.berkeley.edu/~linxu
\_ you wanna download the mp3s first ? ;)
anyway, what I used to use soda for pre-yet-another
rooting-but-this-time-2-weeks-until-it's-fixed. now
i don't use it, except to whine on motd. - linxu
\_ /var/mail is full!!
\_ Fixed. 2050 inactive accounts now bounce mail rather than sit
and collect spam. If you have any CSUA friends who never got
their account reactivated back in May, let em know that their
account may disappear in a few months. - jvarga
\_ The lockfile issue seems to still exist.
\_ Unfortunately I don't have time to debug this. - jvarga |
| 2006/11/30-12/8 [Computer/SW/Unix] UID:45398 Activity:nil |
11/30 hey dudes, var/mail is full and im getting weird NFS
locking errors when trying to delete a file. |
| 2006/11/28-12/12 [Computer/SW/Languages/Misc, Computer/SW/Unix] UID:45383 Activity:low |
11/28 Now that soda is most likely going to be p0wn3d every six
months, can someone recommend a professionally maintained
unix account where I can do my scripting, web hosting, and
other stuff. One of my friends recommended hostrocket but
what else do you guys recommend?
\_ http://csua.com/?entry=43884
http://csua.com/?entry=43360
http://csua.com/?entry=43868
There are others out there just stfm yourself. At any rate, do NOT,
I repeat, do NOT use DreamHost. They are really NightmareHost.
\_ Just run your own box. If you know me personally, email me, and
I'll see about setting you up on one of mine. -dans
\_ What kind of connection do you get?
\_ Like I said, if you know me personally, email me. -dans |
| 2006/11/21-12/30 [Computer/SW/Security, Computer/SW/Unix] UID:45359 Activity:nil |
11/21 Bad stuff happened. Root is working on restoring all services to
normal. Please email root if something is not installed or is on the
fritz[0rz] (according to michener). - jvarga
\_ How would we be able to tell? Logins are still disabled, according to
someone named "jvarga"... (2006-11-22 07:30)
\_ You have to be THIS tall to enter...
\_ Please email activate@csua.berkeley.edu to get your account
reactivated. |
| 2006/11/17-27 [Academia/Berkeley/CSUA/Motd, Computer/SW/Unix] UID:45351 Activity:nil |
11/16 You know what I think? Kids today don't care about motd. Or wall. Or
nwrite. Or a csua email account. Or UNIX. You know what? They
probably don't even know wall/motd/nwrite exist. |
| 2006/11/12 [Computer/SW/Unix] UID:45333 Activity:nil |
11/12 Samba thinks that Novell should reconsider its deal w/ M$:
http://news.samba.org/announcements/team_to_novell |
| 2006/11/11 [Reference/Religion, Computer/Theory, Computer/SW/Unix] UID:45329 Activity:high |
11/11 Curt looked at his erection, amazed. "I thought it had stopped
earlier... My God, I'm huge!" Tentatively, he touched his engorged tip
and patted it with his finger tips. It bobbed slightly as he stood up.
Picking up a tape measure from the hall desk, Christa quickly came
over to Curt and measured him. She let out a low whistle of
appreciation at his new length as she wrapped the tape measure around
his shaft. She looked up at him... "Curt, you've grown to eleven
inches long and five-and-a-half inches around." She did some quick
mental math. "Jesus, Curt. You're 1.7507 inches thick! Make
sure you keep that thing the hell away from me." |
| 2006/11/3 [Computer/SW/Unix] UID:45151 Activity:nil |
11/3 I can't log into any of the office machines, can someone reboot one
of them to a Unix variant? |
| 2006/11/2-3 [Computer/SW/Unix] UID:45114 Activity:nil |
11/02 I have a bunch of directories with files (and other directories)
in them. I have a mirrored copy that contains many of the same
files, but with different dates. Is there an easy way (short of
writing s/w to check each file against the other) to do something
like : If date on mirror < 10/31, then sync date with other dir?
I don't think rsync can help with this. I just need the dates
changed, not the contents of the files. Changing the contents
could actually be bad.
\_ Why would you want to do this? (Curious, and it might help to come
up with a better solution)
\_ Does one line of perl count as "writing s/w"?
\_ find; grab time stamp; stamp matching file
\_ The catch is that a checksum needs to be done at some point.
Obviously, you could add a step to checksum in there, too. I was
just wondering if there's a utility to do this. Essentially,
I want rsync, but to have rsync only change dates and not
contents. If you read the rsync man page, you will see that
rsync has a lot more options than find for deciding when to
do what.
\_ What role does the checksum play in this? Are you trying to
do something more than set the timestamp on /mirror/file to
the timestamp on /primary/file if the timestamp on/mirror/file
is more than N days old, or something more complicated?
\_ If you're really too lazy to write something trivial in perl,
you could use rsync with the -n option to find the relevant
files, and pipe that to xargs/foreach. -tom
\_ I broke down and did this. It was not too bad. Thanks. |
| 2006/11/2-4 [Computer/SW/Unix] UID:45107 Activity:low |
11/02 I need to include instructions on how to create a tcpdump file as part
of a Word Doc testplan . What I would like to do is to be able to cut
and paste from a Unix shell window into a word document without
losing the formatting or having it come out like regular text.
Is there a template or mode I can use for this inside word?
--Thanks Stressed out linux hacker
\_ ??? It's a single command. Why not just type tcpdump and your
options in at that point in the doc? Are you cut'n'pasting tcpdump
output or just the command to run tcpdump?
\__ Sorry; I was just giving a high level summary; there
bunch of other setup to do since this both a playback and
record along with a file transmission. There are about 15 or
so things the user needs to type out and monitor. The real
point of the question is: How can I import something into word
from a unix shell w/o all the tabbing getting screwed up and
looking like crap.
\_ This sounds like a one-off doc. I'd just c'n'p each of
the 15+ lines once per and forget about it. You shouldn't
have funny tabbing or anything from a single line. Do you
have an example of what you're trying to copy in?
\_ Why use a word doc when a plain text file would work better?
\_ Because word looks better.
\_ The tabbing issue comes due the fact that this in outline
form per the corp testplan spec. So I need to have lines
roughly tabbed in 1/2 way. e.g.
%unix ssh root@DUT
%DUT df -f . ( check for free space)
%DUT ifconfig em1 up
[......]
Things need to remain tabbed and all that. I am also supposed
to put commands in one font and user type-ins in italics
( I really really miss the days of plain text docs). I was
hoping that there was something in Word that would already do
this (like an emacs mode).
\_ Sorry but not that I know of. For the tabbing you can
set a marker from the top horizontal bar thingy with the
little arrows on it (someone help me out here) so that
section of text is auto shifted right to where ever.
The italics and etc you'll have to do yourself AFAIK. |
| 2006/10/30-11/1 [Computer/SW/Unix] UID:45041 Activity:high |
10/30 My favorite command in UNIX is "cal". What is your favorite?
\_ Nein, I vill be. A German and a Jerry, danke, und Heil! -John
\_ PARSE ERROR. Are you on drugs again John?
\_ That's just my troll farm being rambunctious. Use new
Extra Strength Troll-Gro for healthy and frisky trolls!
\_ Based on frequency of use, gonna have to go with tcsh
\- "strip tcsh"
\_ unzip; touch; strip; finger; head; latex; mount; gasp; yes;
god; more; comm; spray; umount; sleep; zip; exit; dump
\_ date
\_ bc followed closely by procmail
\_ bc? What's so special about bc? I picked "cal" because it is
useful, and it is Cal!
\_ I'd say I use bc and cal most. Bc is great when you
want to do simple math.
\_ Except that you have to tweak scale to get decimal values.
I prefer nickle or just perl -lne 'print eval', or better
yet, ~mconst/bin/pc --dbushong
\_ You don't have to tweak scale. You put scale= in your
.bcrc file just once. You can also pipe to bc, it has
command line history and editing, variables, all the
basic math functions you'd expect on a decent
calculator and let's you write your own as well. Your
bcfu is weak! When you can pull this bc from my
hand, only then will you will ready.
\_ why can't it graph functions and find intercepts?
\_ why can't vi output quality latex for your
thesis? why can't gcc debug your code? why
can't the government provide services without
waste? Because they're not designed to.
\_ Can you explain why using bc makes sense
instead of programs like mathematica, matlab,
maple, or igor? I mean assuming you're not
pathalogically cheap or are selecting your
math software based only on demonstrating your
"fu"?
\_ If you don't know the difference between a
small but powerful-for-it's-size CLI unix
tool and a full blown math package like
mathematica you wouldn't understand the
reason for using either in the first place.
Why do you use "ls" instead of "echo *"?
Sheesh. Enough trolling. You're way into
lame now.
\_ Oh, no, I get it. Matlab and mathematica
are what those of us who actually need
to do real math use to do useful work.
Bc is what fat sysadmins use to show
eachother how big their dicks are.
\_ Thank you for participating in the
"What is your favorite UNIX command?"
thread. Your math nerd hostility
and insane rant has been duly noted
and disregarded. When you learn the
difference between "unix command" and
"shrink wrapped math package" you may
attract more than disdain the next
time you attempt to communicate. In
the mean time, I suggest classes on
how to behave like a real human being
might (even simulated behavior is
acceptable online) rather than like
the raging flaming asshole you are.
Again, thank you for joining us.
_/
Likewise, asshole
/"\
|\./|
| |
| |
|>~<|
| |
/'\| |/'\..
/~\| | | | \
| =[@]= | | \
| | | | | \
| ~ ~ ~ ~ |` )
| /
\ /
\ /
\ _____ /
|--//''`\--|
| (( +==)) |
|--\_|_//--|
\_ you're pretty touchy for a hostile math nerd,
but you're still welcome to join humanity if
you ever figure it out.
\_ soda% matlab
matlab: Command not found.
\_ /usr/bin/octave
\_ Well I'm just saying, doing that doesn't sound
that hard and would be a cool thing to have
in a quick commandline utility like this.
Newton's method and whatnot. It would have to
have the syntax support though. A simple
ascii-graphics attempt at a graph would be
cool too. Anything like that out there? I
guess it doesn't belong in bc though. gc?
\_ In theory you could get bc to output data
in a format you could feed into something
else to graph, etc. bc supports loops,
logic, functions, etc, so it can probably
be extended to do a lot of things never
imagined by the authors but you'd have to
write those functions yourself or call them
in bc from math library.
\_ My manpage says nothing about .bcrc, and it doesn't
seem to work. What version are you using?
\_ # bc -v
bc 1.06
Copyright 1991-1994, 1997, 1998, 2000 Free Software Foundation, Inc.
\_ That's what I've got and soda's got, and the
.bcrc file doesn't work on either.
\_ Hmm, dunno. I'm using freebsd. Maybe
their version is tweaked a bit. You could
alias bc to echo scale=foo | \bc or
something I guess but it probably isn't
worth it and would probably break other
things.
\_ Those don't have good commandline editing (history etc).
I suppose adding that shouldn't be too hard.
\_ bc -l
\_ nickle has history, and I like its . thingy, but you're
right, my bc fu is weak.
\_ Nickle is overkill. And there are things I use bc
for daily that I'm pretty sure are harder in nickle.
\_ emacs
\- emacs transcends unix
\- find ... | xargs
\_ grep -r something *
\_ I tend to do more grep -r something .
when I'm using -r .... --dbushong
\_ If you like cal(1), try ~dbushong/bin/c
\_ screen
\_ wow, did I actually see someone flicking someone off?
\_ just Hostile Math Nerd upset that someone would dare use a
simple program like bc instead of a monster to do simple math.
\_ You see that Matlab code on your ass? There's identical
Matlab code on my boot.
\_ Nope, just greasy math nerd on the rampage.
\_ awk
\_ perl |
| 2006/10/17 [Science/Electric, Computer/SW/Unix] UID:44842 Activity:nil |
10/17 sed/unix script help.
I am trying to modify ldif file so i can actually import it to
my directory server (SunOne). My current ldif files, each entry
start with the following:
dn: cn=firstN LastN, mail=abc@xyz.org
What I need, is change it to
dc: cn=firstN LastN, dc=foo,dc=bar,dc=com
where dc=foo,dc=bar,dc=com is a constant
abc@xyz.org is different every entry
dc: cn=... is always the first line of an entry
each entry is seperated by a blank line.
How do I use sed to replace that mail=abc@xyz.org with
dc=foo,dc=bar,dc=com
thanks in advance |
| 2006/10/12-13 [Computer/SW/Unix, Computer/SW/Security] UID:44782 Activity:nil |
10/11 Star Wars characters USB thumb drives:
http://tinyurl.com/kjg53 (gizmodo.com) |
| 2006/10/10-12 [Computer/SW/Unix] UID:44753 Activity:low |
10/10 What's a good iCal supporting contact manager in unix?
\_ The Sunbird project's working on it, but isn't there yet. Amusingly,
there's an old program called ical for unix that... doesn't support
iCal.
\_ Google Calendar kicks ass. I love how you can share and overlay
your schedule with someone else. It's a bit slow now but it's
still beta-ish and like all Google products it'll just get
better and better.
\_ The Sunbird project's working on it, but isn't there yet.
Amusingly, there's an old program called ical for unix that...
doesn't support iCal.
\_ The problem with calendar standards today is there aren't any.
Or ok, more specifically, they are incomplete and unspecified in
ways that matter. The reason for this is all the calendar
companies have made their own decisions re: how to handle various
things such as "every other week". Some take it literally and just
do $week+2 while others take it to mean "$week2 and $week4 of
$currentmonth". That's just 1 small example. So when you have a
device or program that supports 1 version of the standard and then
try to sync with a system that has a different idea your calendar
will get all messed up, usually in unfixable ways. --calendar guy
\_ hey, what do you mean you are a "calendar guy?" are you working
on calendar? any recommendation for PERSONAL calendar server
I can run on my laptop?
\_ it's one of the services i currently support at work. i've
attended the ical standards committe talks, on the mailing
list, blah, blah, blah. sorry but i use the same calendar
i support for work for 10k+ people. i don't know much of
anything about the smaller programs.
\_ I like Meeting Maker. What product do you support? |
| 2006/10/5-7 [Computer/SW/Unix] UID:44692 Activity:nil |
10/6 locale question
is en_US equivalent to POSIX and/or en_US.UTF-8?
What does Redhat default to? UTF-8? en_US? POSIX? - danh
\_ no, POSIX is essentially C. en_US.UTF-8 is, UTF-8. |
| 2006/9/27-28 [Computer/SW/Unix] UID:44563 Activity:low |
9/27 I have Verizon and when I got my service they claimed my location
could get the maximum rate of 3000kbps dn/768kbps up. In reality,
I'm getting 2500kbps dn/200kbps up. While I'm happy with the
downlink, the uplink of 200kbps is nowhere close to their claim of
768kbps upload. I called Verizon's leftist tech support 5 times, 4 of
which I had to explain to the tech support the difference between
up/download. The person who did in fact understand the difference
told me he ran diagnostics that showed I was getting a whopping
800kbps upload speed, and could not duplicate my own tests using
FTP, HTTP upload, and Java speed tests like http://www.broadbandreports.com
Has anyone had similar experiences with Verizon internet, the godless
commies??
\_ FTP, HTTP, and Java have overhead which may eat your bandwidth.
Maybe try something which uses UDP packets? Second, if your
download bandwidth is saturated, I've noticed that your upload
bandwidth also suffers. Was your line quiet other than your
test? Last, have you been HAXORED so that you're running
someone's leftish BOT?
\_ also, one thing to get X bandwidth from your provider versus
Y bandwidth from arbitrary sites. |
| 2006/9/18-20 [Computer/SW/Security, Computer/SW/Unix] UID:44435 Activity:nil |
9/18 Any reason that /dev/null is rw by root only?
crw------- 1 root root 1, 3 Sep 13 12:56 /dev/null |
| 2006/9/15-16 [Computer/SW/Unix] UID:44397 Activity:nil |
9/15 Who can explain this (on debian, not soda):
root@mybox etc 15:27:25]# ps -ef |grep grep
root 27419 27390 0 15:30 pts/9 00:00:00 grep grep
[root@mybox etc 15:27:51]#
Note the times. |
| 2006/9/15-19 [Computer/SW/WWW/Browsers, Computer/SW/Unix] UID:44386 Activity:nil |
9/15 Fx 1.5.0.7 is out.
\_ Those of us with Firefox have already been automagically upgraded,
and why would those without care?
\_ unix version doesn't automagically do anything |
| 2006/9/12-13 [Computer/SW/Unix] UID:44364 Activity:nil |
9/12 All bow to the great debian nfs action!
nfs: server 10.0.0.1 not responding, timed out
e1000: eth1: e1000_watchdog_task: NIC Link is Down
e1000: eth1: e1000_watchdog_task: NIC Link is Up 1000 Mbps Full Duplex
e1000: eth1: e1000_watchdog_task: NIC Link is Down
e1000: eth1: e1000_watchdog_task: NIC Link is Up 1000 Mbps Full Duplex
e1000: eth1: e1000_watchdog_task: NIC Link is Down
e1000: eth1: e1000_watchdog_task: NIC Link is Up 1000 Mbps Full Duplex
e1000: eth1: e1000_watchdog_task: NIC Link is Down
e1000: eth1: e1000_watchdog_task: NIC Link is Up 1000 Mbps Full Duplex
Already mailed root.
\_ the NFS error is a symptom there, not the problem. -tom |
| 2006/9/10-11 [Transportation/Bicycle, Computer/SW/Unix] UID:44337 Activity:nil |
9/10 Word of advice for road bikers-- DO NOT take your road bike [with
slick wheels] to the beach. I took my bike to the beach and there
were many idiots on the bike path that almost hit me. Yesterday,
I was passing this guy to the left, going at about 18-20MPH when
he made a sudden lane change to the left, which made me swerve
further to the left. Despite the fact that the path wasn't sandy,
my rear wheel slipped. I attempted to recover but went down. As
I recovered, I realized that there were lots of fine particles on
the path even though you don't see sand, and they make your slick
700 low-rolling resistant road wheels slippery. There are too many
idiots on the beach bike path, and the path is as slippery as oil.
\_ Bike paths are not appropriate places for people who go 18-20MPH
(i.e., bicyclists). -tom |
| 2006/9/9-11 [Computer/SW/Mail, Computer/SW/Unix] UID:44331 Activity:nil |
9/9 After rotating my /var/log/mail.* files and restarting postfix,
I no longer have any mail log files in /var/log/mail.*! Is there
something else I have to do for postfix? I never had this problem
when I was using sendmail. Please help! -learning unix
\_ man -k syslog |
| 2006/9/8-12 [Computer/SW/Unix, Computer/SW/Security, Computer/SW/WWW/Server] UID:44325 Activity:nil |
9/9 Is there a gzip-like unix command that will encrypt a file?
I'm looking for something that's widely available. Thanks
crypt (not very secure - DES). Or failing that, openssl or gnupg
\_ openssl or gnupg... what are you looking for? Those will work fine..
\_ Thanks for the recommendations. I'm basically experimenting
with a way of using my friend's computer to backup my
personal files and using my computer to backup theirs.
Of course, this means storing files in a way where we can't
see each other's personal files.
\_ I'd recommend checking out http://dar.linux.free.fr
It makes the whole "backing up a bunch of files, encrypting
it, and chunking it into bite-sized pieces" thing much easier
than dump/tar + gzip + openssl. --dbushong
\_ Oh, that is so cool. Thanks. My way was going to
be much more convoluted involving ssh and a bunch
of script writing. This should save some time.
\_ One nice thing about using gpg (dump/tar | gpg) is you
can do public key crypto and not ever have passwords stored
in the script. I believe gpg also can chunk it into X
byte chunks, optionally ascii armored, for emailing as
well. (well, I suppose you could mime-attach it)
\_ openssl bf-cbc -in file.txt -out file.txt.bfcbc # encrypt
openssl bf-cbc -d -in file.txt.bfcbc -out file.txt # decrypt
--dbushong
\_ /usr/bin/{zip,unzip} on soda can take passwords. Don't know if
they're widely available on other *nix's. |
| 2006/8/29-31 [Computer/SW/Unix, Computer/SW/Languages/Misc] UID:44191 Activity:nil |
8/29 Any recommendations for windoze software for backing up files to an
external hard drive? I'd prefer that the backup isn't one huge .bkf
file a directory? Is this possible with the ntbackup.exe that comes
with windows XP.. Also I'd like the software to be a GUI that normal
non-technical people could run. THanks.
\_ Not for windows but if you're running any flavor of Unix, just
add a daily crontab
> rsync -az --rsh=ssh <src dir> <remote dst dir>
\_ Not sure if you're looking for freeware or not. I 'borrowed' a
program called Second Copy from a friend and I think it does a
good job (but it costs $30 to buy). http://www.centered.com
\_ Probably not what you're looking for, but I use the following
batch script (where z: is the backup drive):
xcopy /e /c /d /y /f c:\to_back_up z:\
Recursively copies files with newer modification times, overwriting
without prompting, displays what was being copied, and will recover
from network hiccups if the drives are network shares.
\_ Thanks.. those are just what I was looking for -op |
| 2006/8/29-31 [Computer/SW/Unix] UID:44188 Activity:nil |
8/29 % uptime
13:46:57 up 41 days, 2:41, 100 users, load average: 18.09, 18.04, 19.47
\_ % uptime
11:25:33 up 42 days, 19 min, 90 users, load average: 31.67, 28.68, 20.81
What the? |
| 2006/8/7-11 [Computer/SW/Unix, Computer/SW/Security] UID:43929 Activity:nil |
8/7 hey ax watch this when you get home from work
http://www.youtube.com/watch?v=f83L9iWIx54
\_ Is there a CSUA login?
\_ Use http://www.bugmenot.com to find a login.
\_ That was work safe in my book, but thanks for thinking
of me. Has anyone one noticed any patterns concerning
women who wear Bebe shirts? -ax
\_ I wanna work where you work. I think there's
a pattern for chicks who wear Hollister shirts.
\_ one more for you
link:tinyurl.com/gmaxd
\_ Jesus Christ.
http://i43.photobucket.com/albums/e381/oklahomaok/DSCN0419.jpg
\_ That's not Jesus Christ. --Mel Gibson |
| 2006/8/4-6 [Computer/SW/Unix] UID:43913 Activity:nil |
8/4 Is there a standard unix tool for splitting input into multiple
lines based on some pattern? thx (I can do a Perl one liner or
learn sed; just wondering if there's something else.)
\- that's a rather general statement of the problem. so grep might
work, it might not. an example might be helpful.
\_ No. -proud American
\_ well I was just thinking of "this is a long line..." >
"this\nis\na\nlong\nline\n..." i.e. sed 's/ /\n/g' and vice
versa although looking at that I guess this is a job for sed... |
| 2006/7/19-23 [Computer/SW/Unix] UID:43723 Activity:nil |
7/18 We have lots of Unix IT admin openings in our Check Point/Zone Labs
RWC office. If you are interested send me an email at
sking@@us.checkpoint.com with your resume. Thanks!
-sky |
| 2006/7/16-18 [Computer/SW/Unix] UID:43681 Activity:nil |
7/16 I am in the process of switch to UTF-8 as default encoding
environment for various reason. But I just discover that most of
there are a great number of UNIX utilities that doesn't support
text file that is encoded in UTF-8 (e.g. less, more). Has anyone
here seriously tried to migrate to UTF-8? what kind of problem
have you encountered? thanks kngharv
\_ you sure it's not just your term or ssh client?
\_ that was what I thought at first too. What I did was
viewing a utf-8 encoded document in couple ways: 1. use less
2. use more, and 3. use cat
when I cat the file, everything renders correctly. That is
about the only time the file renders correctly. I suspect
I did something wrong ehre, because not even emacs seems to
be render utf-8 encoded text. In all failed cases, something
like 25% of the Chinese text is being display correctly. The
sheer fact that 1. cat display all characters, and 2. when
failed, some text actually displayed puzzles me.
\_ That sounds like your locale is set wrong. Try running
"setenv LC_CTYPE en_US.UTF-8". --mconst
\_ Did you try setenv LESSCHARSET utf-8 ? |
| 2006/7/15-18 [Computer/SW/Unix] UID:43678 Activity:nil |
7/15 Mi comrades, how to samba mount on MacOS X? Spasibo!
\_ Learn proper grammar first, you fucking immigrant
\_ I guess you work in a company which doesn't have native English
speakers in the QA department who are too lazy to write
understandable bug reports. Those reports drive me crazy.
\_ Tovarich, depends on whether you're doing workgroup or domain
auth. Both should actually just show up under 'network' in
finder. Click on a server->connect and it'll show you the
list of shares on that box. -John
\_ http://www.macdevcenter.com/pub/a/mac/2003/03/18/samba.html
http://www.macdevcenter.com/pub/a/mac/2002/11/19/mac_pc.html
Go to Finder, click Connect to Server, type:
smb:\\your.ip.here
\_ I think it's smb:// (though it may not matter)
\_ In Soviet Russia, Spanish learns you? |
| 2006/7/14-18 [Computer/HW/Memory, Computer/SW/Unix] UID:43671 Activity:nil |
7/14 So as I understand it, Intel introduced Hyperthreading because the
penalty for a cache miss or branch misprediction on Netburst was huge
so to make up for it they could work on a different thread while
waiting for memory latency. Given that the Core2 architecture has
such a wide execution path, couldn't they use HT to try and keep all
those execution units full?
\_ I attended one of Intel's talks on campus and the "huge penalty"
myth is a myth. As the number of pipelines increases, the depth
of the miss and a pipe flush is longer, yes. However, keep in mind
that since each stage in the pipe is shorter the clock cycle is
also faster. Thus, in terms of absolute time (time=number of
pipes that need to be flushed * cycle time), the increase in P4
from the old architecture is only increased by ~20% time, which
is insignificant in computer science speak.
As for keeping execution units full, it's very much application
dependent. Even with the huge instruction reorder mechanisms some
applications still can't utilize them all. |
| 2006/7/12-18 [Computer/SW/Unix] UID:43648 Activity:nil |
7/12 http://www.ktvu.com/video/9502605/index.html They interview a guy named Mike Litoris. \_ Speaking of I-580, anyone know what the HUGE construction project happening all along the north hillside of the Dublin grade is? \_ Who's that? \_ He's a protege of Seymour Butts. \_ Is his middle initial 'C'? \_ No, his middle name is Ockrubbingagainstyermomsc \_ wait, how do you know? didnt mention that in the article \_ Oh well. It would have been "Mike C.Litoris". \_ Do you sta-stutter? \_ Umm.. Mike Litoris == "My Clitoris" say it out loud. \_ ]finger beavis finger: beavis: no such user. Are you using an alias of some kind, Beavis? \_ $finger cornholio \_ Is it from The Simpsons? |
| 2006/7/5-6 [Computer/SW/Unix] UID:43562 Activity:nil |
7/5 How can I remove my info from soda's web finger gateway thing?
\_ touch ~/.nofinger |
| 2006/6/27-29 [Computer/SW/Unix] UID:43509 Activity:nil |
6/27 Socket question: does it matter where in the filesystem you
put the socket file? I.e., /tmp/ vs ~/ ? When could it matter?
Thanks, --PeterM
\_ "the" socket file? You mean screen? Nope.
\_ Not screen. There's a program (not mine) that
works when the socket is located in /tmp and
doesn't otherwise.
\_ Try creating a directory, turning on the sticky bit
and putting it there. That is the only thing I can
think of that is "special" about /tmp. -ausman
\_ Agreed. I've created sockets in other directories
before, when it was owned and used by a same user.
It worked fine for me.
\_ screen defaults to using /tmp because it wants the socket tobe
on a 'local' filesystem. sockets can do odd things on NFS file
systems. /var/tmp is ok... ~/ .. is nfs mounted. -ERic |
| 2006/6/25-28 [Computer/SW/Security, Computer/SW/Unix] UID:43493 Activity:nil 53%like:43401 |
6/25 Hey root, could you please reenable finger motd@csua? Can't be a
security issue since fingerd is enabled ...
\_ Done. For some reason linking it refused to work, so I added it
as a cronjob that happens just as the motd concatenation happens
(every 2 minutes). --michener |
| 2006/6/23-28 [Computer/SW/Security, Computer/SW/Unix] UID:43487 Activity:nil 80%like:43482 80%like:43483 |
6/23 Soda rooted by sendmail bug. Will be going down at 8pm.
\_ Resetting accounts again?
\_ Good thing I stopped using my @csua.berkeley.edu address as my main
non-work e-mail address!
\_ Let's try FreeBSD again!
\_ Let's try Windows!
\_ Er, why is this in motd.public?
\_ soda is run by liburals, always aiding and comforting Terrorists
\_ Maybe 'cause it's a lie?
\_ So, why is it still up?
\_ It used to say 5pm.
\_ The crackers have changed the root passwd! Root is powerless!
\_ I can assure you this has not happened. --michener
\_ They probably exploited something to put in a trojan su.
Did you test this by suing? Now they probably have the
old root password!
\_ Someone should go to the server room and destroy soda with
an sledgehammer before the crackers unleash the skynet on us.
\_ I know of no such issue and have not heard from the rest of root
about it. If this is not a lie, will whoever wrote this email root?
--michener
\_ Did the cracker post this to freak everybody out?
\_ The only non-anonymous evidence I see is on wall log, where
Paolo posted a snippet showing brg and sly speculating on
whether soda had been hacked
\_ which had nothing to do with sendmail at all.
\_ Lying about a rooting is l4m3.
\_ My account has been hacked! Last login from China! |
| 2006/6/23 [Computer/SW/Unix, Computer/SW/Security, Computer/SW/Mail] UID:43483 Activity:kinda low 80%like:43482 80%like:43487 |
6/23 Soda rooted by sendmail bug. Will be going down at 3pm.
\_ Resetting accounts again?
\_ Good thing I stopped using my @csua.berkeley.edu address as my main
non-work e-mail address!
\_ Let's try FreeBSD again!
\_ Let's try Windows!
\_ Er, why is this in motd.public?
\_ soda is run by liburals, always aiding and comforting Terrorists
\_ Maybe 'cause it's a lie?
\_ So, why is it still up?
\_ It used to say 5pm.
\_ The crackers have changed the root passwd! Root is powerless!
\_ I can assure you this has not happened. --michener
\_ I know of no such issue and have not heard from the rest of root
about it. If this is not a lie, will whoever wrote this email root?
--michener |
| 2006/6/22-26 [Computer/SW/Languages/Misc, Computer/SW/Unix] UID:43464 Activity:nil |
6/22 How do I make rsync exclude any directory containing
a file named NOBACKUP ?
\_ see --exclude in the man page. and really you should have been
able to find this yourself.
\_ I did read the man page. I really don't think it's that simple.
I can easily exclude a DIRECTORY named 'NOBACKUP'.
I can exclude a file named 'NOBACKUP'. I don't see
an easy way to exclude all files in a directory
that contains a file named 'NOBACKUP'
\_ That's a bit more complex. You will need to write a find script\
to generate a list of files, and exclude from taht list all
\_ That's a bit more complex. You will need to write a find scripti
to generate a list of files, and exclude from taht list all
files that meet your criteria. Then have rsync use rsync's
--files-from= option. That or make a exclusions list
(Again with a find script) and use the --exclude-from option.
\_ I want this feature too, maybe we should add it. |
| 2006/6/15-19 [Computer/SW/Unix] UID:43401 Activity:nil 53%like:43493 |
6/15 Hey root, can you reenable finger motd@csua.berkeley.edu?
\_ Also can you please re-enable kais motd's search and intellidiff? |
| 2006/6/6-9 [Computer/SW/Unix] UID:43293 Activity:nil |
6/6/6 Daemon is here.
\_ ^Daemon^Penguin. We're running Linux now! |
| 2006/5/29-31 [Computer/SW/Unix, Computer/SW/Languages/Python] UID:43216 Activity:nil |
5/29 What's the Python library for downloading a file via HTTP? I'm
thinking of something along the lines of http.get(url). (simple)
\_ >>import urllib
>>page= urllib.urlopen('<DEAD>www.python.com').read<DEAD>
-scottyg |
| 2006/5/23-28 [Computer/SW/Languages/Perl, Computer/SW/Unix] UID:43168 Activity:nil |
5/23 I'm trying to learn sh, bash, and csh programming. What's a good
site that has sample syntax of the following ***psuedocode*** for
each of the shells?
foreach $file in (<dir/*>) do;
echo I have file $file
done;
switch ($var) {
case 0: echo hello; break; // this is a comment
default: echo default; break;
}
if ($z>=5 || $str eq '12345) {print $hello}
\_ not really. I'll start w/ sh: --dbushong
for file in dir/*; do
echo I have file $file
done
case "$var" in
0)
echo hello
;;
*)
echo default
;;
esac
if [ $z -ge 5 -o $str = 12345 ]; then
echo $hello
fi
\_ bash is basically a superset of sh. Whenever I need to brush up
I usually just google for shell scripting, and frequently
(re)discover this site:
http://vertigo.hsrl.rutgers.edu/ug/shell_help.html
Incidentally, csh is a great interactive shell, but there are some
compelling arguments why you may want to avoid it for general
scripting purposes (of course, one nerd's compelling arguments are
another nerds religious claptrap so ymmv):
http://www.faqs.org/faqs/unix-faq/shell/csh-whynot
Also, I find the O'Reilly UNIX Power Tools book to be a very useful
reference to have around when shell scripting. -dans
\_ I gave up on csh for all but the simplest shell scripting when
I noticed it didn't handle nested if/then cases sanely, and went
through coding hell trying to workaround it. If you're
going anything more than 2-3 lines, use a sh based shell, or
a "real" script language like perl. -ERic
\_ I'll second that. I still use tcsh from the cmd line just because
I find it more pleasant than bash, but meh. I'll tend to use
perl unless the script is going to end up being more than about
25% program executions. --dbushong
\_ My rule is 10 lines. If I can't write it in 10 lines of some
shell script language in a few minutes I'll use perl instead. |
| 2006/5/11 [Computer/SW/Unix] UID:43018 Activity:nil |
5/11 I've answered lots of Unix Qs, so now my turn: Say I have the
output of "find ... -print" in a file, one per line. Is there a
STARDARD TOOL to change this one-per-line format to a NULL
delimited format I can feed to xargs -0 to operate on ... so I
dont have to worry about spaces and other weird characters legal
in file names. Obviously I am trying to avoid re-running with -print0
and the one-per-line fmt is convenient for other tools like "grep".
--psb |
| 2006/5/9 [Computer/SW/WWW/Server, Computer/SW/Unix, Computer/SW/Languages/Misc] UID:42993 Activity:nil |
5/9 Running httpd as nobody isn't that secure. If one asshole decides to
do a DoS (fork script) as nobody, there's no way to track down the
perpetrator. This is why "suexec" is highly recommended, plus
users don't need to chmod a+rx script.cgi.
\_ Uh, it's totally trivial to track down the perpetrator with or
without suexec. httpd should run as something other than nobody,
but that's only because nobody is over-used, and whether httpd
runs as nobody is orthogonal to the question of whether suexec
should be on. -tom
\_ Ok fine. I gave a bad example, but we both agree that nobody
is good. |
| 2006/5/8-9 [Computer/SW/Security, Computer/SW/Unix] UID:42979 Activity:nil |
5/8 A friend of mine still hasn't gotten his account reactivated even
though he sent photo id. Is this still being worked on?
\_ soda root == students
may 12 == start of finals.
I'd say a little patience is in order.
\_ Granted, but 3 weeks is a long time. Could you update the
website so they have a clue. 4/17 was in a galaxy far, far away. |
| 2006/5/4-7 [Computer/SW/Unix] UID:42934 Activity:nil |
5/4 in light of the recent compromise, how do we tell if our account
has been broken into? Is there a command like "last" instead tells
when I previously logged in?
\_ if the system's been compromized, you can't trust the logs to
really tell you. We do have 'last' here, for what little its
actually worth.
\_ last gets rotated. 'lastlog | grep `whoami`' will show you the
last login. (should) |
| 2006/5/4-7 [Computer/SW/Security, Computer/SW/Unix] UID:42931 Activity:nil |
5/4 Ok I need to make a hosting choice soon because my current co-op
colo is falling apart like http://autobahn.org in the old days. I can
go with http://dreamhost.com, http://textdrive.com, or http://johncompanies.com [from
which dans heard good things about]. I really like johncompanies'
virtual machines because you get root, but it is a whopping
$47/month!!! http://dreamhost.com is dirt cheap, but you share
resources and it's probably just as secure as soda (which is
not very). I haven't heard anything about http://textdrive.com.
What do you guys use and recommend? Thanks.
\_ i have a bunch of stuff hosted with dreamhost and have been
very happy with them. i was referred by another sodan. alot
of guys at gamespy use dreamhost as well and love it. -shac
\_ How much quota, IP/hosts, do you get, and how much do you pay?
\_ i pay annually. so i paid $120 for 1 year starting at 20GB
disk, 1TB monthly transfer. each month they increase both
of these numbers for you so im probably at like 22GB and
whatever monthly transfer. the longer you are a customer
the larger your quotas are. see their pricing comparison.
http://www.dreamhost.com/shared/comparison.html
-shac
\_ which coop is dieing? how much do you use now?
\_ i've heard good things about simpli.biz
\_ JohnCompanies kicks ass. You send mail and.. John mails you back
in like 2 minutes. None of the trouble-ticket-queue bullshit. On
the flipside, yeah, they're (relatively) pricey and you have to do
all the admin themselves, and if John dies in a carcrash, I'm not
\- that is the scenario i call
BUS TERMINATED. --psb
sure how much human failover they have. --dbushong
\_ HA that's pretty funny. But how do you know johncompany is
run by 1 man, and if other companies aren't in the same
situation?
\_ I don't, but if you go with <insert random huge company>
it's unlikely.
\_ Oh, if you are doing anything art or community oriented, consider
Laughing Squid. -dans |
| 2006/5/1-4 [Computer/SW/Unix] UID:42879 Activity:nil |
5/1 Has more always been less? Now I have to hit q when I get to the end
of the motd, as opposed to just returning me to the shell. Is this
new or did I just not notice this change? Now I have to do LESS=-EF
\_ on FreeBSD, more was basically less with certain arguments assumed.
on most linux distribs, it's just less, period. (more or less)
\_ What are you talking about? If I use /bin/more it returns me to
the shell at the end.
\_ /bin/more is a separate binary from /usr/bin/more, which is a
symlink to /usr/bin/less If (like most people), you have /usr/bin
before /bin in your path, you'll get less.
\_ Ah, I had /bin before /usr/bin. What's the reasoning behind
having /usr/bin before /bin? Or maybe, that symlink should
be removed. It makes no sense, because less doesn't behave
identically to more. Maybe if less had a feature to behave
like more when invoked as that. (Like bash/sh)
\_ except bash doesn't
really act like sh still
\_ Usually the crappiest, simplest, oldest, safest versions of
things are in /bin, the system standard ones in /usr/bin,
the sexy new weird local ones in /usr/local/bin, and your
own custom build in ~/bin. So most people order their path
~/bin /usr/local/bin /usr/bin /bin
\_ Well, ok but normally, stuff in /bin wouldn't be also in
/usr/bin. But mainly yeah more just shouldn't be linked
to less like that. |
| 2006/5/1-4 [Computer/SW/Security, Uncategorized/Profanity, Computer/SW/Unix] UID:42873 Activity:nil |
5/1 Fuck it.
RSS wall feed isn't down, so much as only serving out my little diatribe on
the life and death of the hack.
I got shit to do. Later.
--michener
\_ Bring it back. This is *your* organization. It should run *your*
code. If bitchy alums, myself included, have problem with it, fuck
em! If bitchy alums have a problem with google seeing it, tell
them to get off their asses and write robots.txt file to fix the
problem. -dans
\_ root staff seems much more inclined to ignore than do anything.
It's rather discouraging for alums to even bother trying to
help when work making old desired software run is ignored, and
for requests that people "mail root if you want to do this," are
met with much silence.
\_ yes, you can "fuck em" but if the existence of a feed
makes people revolt, and not use wall logging, and therefore
make the feed pretty much useless as well, what's the fucking
point?
\_ Lowers the alumni noise floor of wall thereby making it a
useful channel for undergraduate signal? Sounds like a win
to me. -dans
\_ that way all 6 undergrads can talk. If the alumni noise
is a concern for them, I'm sure they can figure out
how to make a second wallall channel
\_ If, as you indicate, there are only 6 active undergrads
in the organization, then perhaps the CSUA has run its
course and should be shut down. Of course, you're
wrong, so the point is moot. -dans
\_ Do you actually work on getting stupider every
day? -tom
\_ For you tom? Anything. :-*
It's cute how you define stupid as "Any view
that doesn't support what you believe." Here in
reality, i.e. that place outside of the bubble
you live in, we call that closeminded and
juvenile. meh. Bored now. -dans
\_ Coming from you, this is hilarious.
\- but is it ironic?
\_ Let me add some useful info to this debate. For the record, I
think it's a cool idea, and don't care of Google indexes wall, in
which case you could just reenable it as is. But, the bitchy
alumni can read these:
http://csua.org/u/fok - How do I request that Google not crawl
parts or all of my site?
http://csua.org/u/foj - How can I remove content from Google's
index?
\_ Interesting and useful. But, really. I seriously have code to
write for classes and such and shit to do. So I'll revisit at
a later date. Until then, bad alums, no cookie! ;) --michener |
| 2006/4/29-5/2 [Computer/SW/Unix] UID:42863 Activity:nil |
4/29 Freepers rejoice! Your master is back:
jblack:x:13096:100:Justin Black:/home/apollo/jblack:/bin/tcsh
\_ Who's that? |
| 2006/4/27-29 [Computer/Blog, Computer/SW/Unix] UID:42853 Activity:low |
4/27 This is a pretty hilarious (apparently genuine) rant by a
sociology prof. Spot the hypocrisy!
http://drmandrake.blog-city.com/i_hate_my_students.htm
\_ Haha, I love that guy.
\_ He took down the blog. Anyone have a mirror?
Nevermind, Google cache FTW: http://tinyurl.com/pfg8z
\_ he has many grammar errors. -tom
\_ so? it's a blog.
\_ He didn't have "many"... I saw a couple typos and yeah,
it's a blog.
\_ He had many. The worst was "low and behold." And
if you're on a grammar rant, you'd better be accurate
yourself. -tom
\_ Well I think that may also be a typo or at any rate
not strictly grammar. The guy is still a jackass
though.
\_ Tom's "he has many grammar errors" was satire.
That said, grammar is "right use and rules of
language". Mistaken phraseology can definitely
fall under "bad grammar".
\_ I don't think it can be considered a typo if it
was intentional.
\_ If you are typing quickly it's possible to make
mistakes like that even if you are aware it
is wrong. (since it sounds the same.) That said
maybe it was intentional. I still wouldn't call
that grammar. It's kind of like the toe/tow the
line thing where people don't really know
what it means. I guess a diction issue.
\_ Toe/tow the line isn't a typo. I think
it's closer to 'ignorance' or simply
'wrong'. If your point is that we can't
be sure based on limited data -- then, sure.
But folding ignorance and spelling
errors under the general heading 'typo' in
*this* context invites this sort of
pointless semantic 'dialogue'.
\_ He upbraids his students for confusing
"than" and "then," which is a very similar
error. -tom
\_ He is not upbraiding them as he never
voices that to them. He mocks them to
other people. Tom, you have used the
word incorrectly! You should be more
careful on the MotD!
\_ You're officially an imbecile. |
| 2006/4/21-24 [Recreation/Activities, Computer/SW/Unix, Computer/SW/Security] UID:42798 Activity:low |
4/22 ok, so maybe a dumb question, but a coworker just asked me and I'm
not sure the answer: so is it possible to view the standard
output of a process running on your system? I do have root. thx
\_ truss/strace, with the option to print the entire syscalls
\_ ok, let me rephrase: there is a process running on my system.
I did not start it. I am root. I have just the process id
(from ps) ... is there some way I can see std out/err? thx
\_ you can't see what has -already- gone out to stdout/stderr
if you look at the write() calls for stdout/stderr (by fd)
you can see what it is putting out -now-. truss -p pid
\_ try /proc/<pid>/fd \
\_______________\_ this was all helpful. thanks. |
| 2006/4/18-20 [Computer/SW/Unix] UID:42784 Activity:nil |
4/15 Bummer. Reboot due to what looked like NFS woes.
Regardless, props to the current undergrads, they're doing some great
work.
\_ The kids are alright.
\_ Seconded. Everything looks good so far. Keep up the good work! |
| 2006/4/18-5/21 [Computer/SW/Unix] UID:42778 Activity:nil |
4/18 Soda and Keg had a spot of trouble talking to each other and as a
result, NFS mounts had a slight problem, which was fixed for the
moment with a restart. Investigation into the problem is proceeding
apace. We apologize for the inconvenience. |
| 2006/4/18-23 [Computer/SW/Unix] UID:42776 Activity:nil |
4/18 Soda just rebooted. What happened?
\_ For some reason the CPU load was growing out of control. I don't
know the culprit. -!root
\_ Do you know the process id? Was it wait time or cpu or user?
\_ It was NFS hosage to keg. Unknown why (keg seemed to be fine)
\_ That seems likely, but was not definitively confirmed. BSD's
NFS implementation and Linux's NFS implementation have a
history of not playing nice. Though I've always heard of BSD
NFS clients not talking to Linux NFS servers, and, in this
case it would be a BSD NFS server (keg) hosing a Linux NFS
client (soda). -dans
\_ pave keg
\_ Let's not be so hasty. Perhaps we should take steps to
make sure that the problem is, indeed, BSD NFS <->
Linux NFS interaction before doing a lot of work to
pave keg. -dans |
| 2006/4/18-20 [Computer/SW/Security, Computer/SW/Unix] UID:42775 Activity:moderate |
4/18 Some thoughts about securing a machine. Feel free to add your
expert opinions. --ricky
* Securing a machine that allows interactive logins by users
is _very_ hard.
* Reduce suid binary to absolute bare minimum.
* Perform automatic _remote_ checksums from a machine that is
separate and is not accessible by regular users. Usually,
NFS is recommended for this. Basically, have a remote
machine regularly check critical files on the machine and
alert root if anything changed.
\_ This existed a while ago, called Tripwire. Started as a
a research project and grew to a startup. Many people tried
it but gave it. The concept is easy, but in practice, it takes
damn too much time. All of the above suggestions are good,
but in the end, if the cost of manageability is high, no
one will care. Lastly root and politburo aren't paid to do any
of the above stuff and most people have better use of their
time so... why cares. Would YOU like to volunteer
\_ Why are suggestions being taken as a demand that they
do it. If alumni (or "members") do all this stuff, aren't
they just "fucking the undergrads" ?
\_ No. If they storm into the machine room or the office
and insist that it be done there way and be done right
this minute, then they are fucking the undergrads.
Historically, asking nicely and accepting a polite `No.'
is not one of the strong suits of the alumni. Though
anecdotal, it's also worth noting that the amount a
given alumnus bitches appears to be inversely
proportional to the amount of meaningful contributions
(time, money, hardware, etc.) he makes to the
organization. -dans
\_ so you contribute absolutely nothing, eh? -tom
\_ Ah let me clarify that. The amount a given
alumnus bitches at the current undergrads appears
to be inversely proportional to the amount of
meaningful contributions he makes to the
organization. If the alumni bitch at each other,
it has no bearing on the CSUA or its future.
-dans
doing these things ricky? You should attend politburo.
\_ Agreed, I tried to set up a modern version of tripwire on
hosts I administered in my last job, and it's nigh unusable.
It smacks of overengineering, and has too many features
apparently added by marketing folks trying to sell to the
enterprise software market. Furthermore, if you want to be
really secure, running _remote_ checksums isn't good enough
since the credentials for soda are likely the same as the
credentials for other CSUA hosts. Thus, checksumming soda's
binaries from screwdriver takes a non-trivial amount of work
for a trivial gain. Also, what happens when people trojan
libraries not binaries? Should we checksum those to? Which
libs? -dans
\_ ideally you checksum everything, and flag what is 'volatile'
and likely to change from day to day.
\_ ideally, yes, but that's a really time consuming,
tedious, manual process. Unless you have some '1337 tool
to do that for us. If so, please post a url. -dans
\_ I have used aide, a tripwire-like tool that checksums files
in two ways. It works pretty well, and isn't that difficult to
use. I found it annoying if I didn't check/update signatures
before doing package upgrades, which meant I couldn't tell
whether the changes were intentional from the update or if
someone had done something to the binaries the same day.
While there are certain more-secure "ideal" ways to set things
up (binary on immutable media, running on a separate system,
database on immutable media, etc.) A simple "on this system"
"aide running out of /usr/sbin" "database stored locally" while
not great from a security standpoint, as long as one doesn't
rely on the lack of warnings and messages to mean you are
secure, is still a useful tool.
* Educate users about ssh. For example, unless the user is
extremely certain that their private keys are safe (resides
in encrypted partition, etc.) having empty passphrase is a
bad idea. Assuming above is met, using passphrase protected
key pair and setting up authorized_keys is safer than using
passwords.
\_ Education works the best, when people are willing to
be educated. Do you think people like to be educated?
\_ It's also vital to keep up with patches to OS and utilities.
\- ssh wont solve the problem. the problem is a combination of
clueless users and users who dont care about security [and
are willing to login from machines with kbd sniffers]
combined with the close to inevitability of local account ->
local exploit -> root. i think sloda should adopt the
position: 1. soda will be broken into and should not be
trusted ... meaning it should not be used as an outbound
stepping stone ... no rsh, rlogin, ssh, telnet. i suppose
you can leave ftp on and i guess scp. 2. do what you can
about prevention [applying patches etc but also invest some
in rapid detection. tripwire is a piece of crap but there are
other tools to do this with ... i maintain checksums on about
50 things [in some cases OSes, in other cases various data
trees] and while i dont look at all the data everyday, with
disk being cheap i can store enough snapshots i can at least
go back and tell a story if there is a problem found at some
point. even a half asses checksumming system will get you
pretty far ... and would certainly pickup a trojaned daemon
or client. we have some not-very-portable hacks to address the
case of trojaned libs [these check low level information in
inodes and compare them to higher level queries and look
for inconsistencies ... like say in the link count] but these
are probably not worth the effort ... they were crafted for
very specific rootkits. |
| 2006/4/17 [Computer/SW/Unix] UID:42761 Activity:nil |
4/17 > grep safesorry /etc/passwd|wc
2454 5541 185708
> grep sorry /etc/passwd | wc
2842 6399 213409
> grep -v sorry /etc/passwd | wc
128 305 8611
128 users got their accounts so far. I guess safesorry is one
way to solve resource problems.
\_ I guess I wasn't the only one who wanted to see how many folks were
back. |
| 2006/4/16 [Computer/SW/Unix] UID:42754 Activity:nil |
4/16 1 day uptime! I can't believe i'm marking this moment, but there it is. |
| 2006/4/15-21 [Computer/SW/Unix] UID:42739 Activity:low |
4/15 Another idea. I don't know if it's practical. But is it possible
for each user to provide a yahoo or other address that mail can
be forwarded to during extended downtimes like this?
\_ Certainly, you can add a .forward file. Now if you want something
that well work during downtime... thats not the answer. We are
seriosly considering moving mail to another box so when soda
gets rooted we can spend less time unable to deliver mail.
\_ I think that's best left up to the users. I got no mail thru the
downtime, but i Expected that, and did not care.
\_ So, as a user, what can I do, if anything, to set my
incoming mail to forward to another address _when soda is
down_?
\_ As a user, absolutely nothing. The CSUA doesn't have users,
it has members. As a member, you could write some code that
\_ as a user of the systems, geez, as in, versus a root
staffer who can do things with their mail.
\_ Yes, as a member, you can write code which you can hand
to root and say, check out this neat program I wrote
that lets our members do this cool thing they couldn't
do before. -dans
makes what you are requesting possible. Alternatively, you
could offer the current undergrads assistance with migrating
mail to a host other than soda, though after the amount of
time they just spent on soda, I don't think they're too keen
on taking on another big project right now. -dans
\_ how would this be a big project? The same way sendmail
is setup now, set it up on box B. change MX records.
NFS mount mail from box B. Setup imap/pop on box B
instead of on soda.
\_ The fact that your solution involves NFS mounting mail
with the existing sendmail setup shows that you have no
idea what you're doing. That, or, in the two years
since I last looked at this problem, file locking over
NFS has changed so that it works, and sendmail uses
this shiny new magic (i.e. working) NFS file locking.
-dans
\_ Does this mean sims, eecs, ocf, etc, plus all
the other places that export their mailspools,
all of them have "no idea what [they're] doing" ?
\_ dans and tom: a case of convergent evolution
\_ If the following reasoning is flawed, please point
out the error, and I will admit that my previous
statement was wrong. Of course, all of the
reasoning below goes out the window if NFS now
features a working locking mechanism, a
qualification I made above:
1. Multiple processes cannot safely write to a
classical unix mbox file without a functional
locking mechanism.
2. Modern MTAs, a set I begrudingly include sendmail
in, use multiple processes or threads to speed
mail delivery. As a consequence of this, modern
MTAs cannot safely deliver mail to an mbox file
hosted on a remote NFS server.
3. When a mail client, e.g. pine or mutt, works
directly with an mbox file (vs. through a POP or
IMAP server), it will need to write changes to
the file. If the file is hosted on a remote NFS
server, these writes cannot be performed safely.
Ergo: Working with mail stored in unix mbox file
format via a remote NFS server carries the risk of
corrupting mailboxes and losing mail.
-dans
\_ Sendmail doesn't have to use mbox. However,
this is the same concern a number of us raised
when soda went to nfs mounted homedirs. It
was summarily ignored.
\_ procmail is nfs-safe -- or claims to be
anyways.
\_ That's nice. It's also an additional detail
that adds to the work required to move from
the existing mail setup to your shiny
hypothetical new mail setup. Are you
volunteering your time and services? -dans
\_ possible, but not currently, some sendmail hackery would
do this quite nicely.
\_ Does anyone know if mail is being queued during downtime?
\_ mail is queued for 5 days or until scotch's disks fill.
\_ #f. Normally mail is queued for 5 days (this is a
sendmail default) or until scotch's disks fill, but the
max queue time was increased to prevent mail loss. -dans
\_ How about next time soda is taken offline for a significant amount
of time that everyone gets suitable warning first?
\_ While we're at it, why don't we put a message on the website
asking hackers to kindly inform us one week in advance before
breaking into soda? -dans
\_ What does that have to do with anything? If soda was already
compromised for who-knows-how-long, what's an extra day? And
ssh access could have been stopped while giving people some
time to download their email locally.
\_ As soon as the extent of the compromise was realized we
immediately brought the machine down. I apologize, but I
imagine that when we get hacked the next time we will have
to bring the machine down quickly again.
\_ You're not familiar with UCB campus network policies, are
you. The root staff did what campus policies required and
what any responsibly system administrator would do. There
are more important aspects to running soda than providing
you with continuous, uninterrupted access to your mail.
Now that you know this, perhaps you will set up your
.forward to save a copy locally and automatically forward
a copy to a secondary mail address. Come on, knowing is
half the battle! -dans |
| 2006/4/11 [Computer/SW/Unix] UID:42722 Activity:nil |
4/11 telnet: connect to address ::1: Connection refused
telnet: connect to address 127.0.0.1: Connection refused
telnet: Unable to connect to remote host
Trying ::1...
Trying 127.0.0.1...
(from soda)
fingerd also down. |
| 2006/4/6-7 [Computer/SW/Security, Computer/SW/Unix] UID:42706 Activity:low |
4/6 /var/mail is full. I'd mail root, but...
\_ soda: [~] % du -h /var/mail/kislyuk
16G /var/mail/kislyuk
\_ Last login Sun Dec 4 18:44 (PST) on ttyB5 from ....
New mail received Thu Apr 6 09:12 2006 (PDT)
Unread since Sat Dec 3 12:47 2005 (PST)
\_ Isn't there a 25M quota on /var/mail? How did it get to 16G? |
| 2006/4/6-7 [Computer/SW/Security, Computer/SW/Unix] UID:42704 Activity:nil |
4/5 Problem: sshd acting weird. Platform: Linux 2.6.x. Symptoms: Ssh
\_ I thought Soda ran FreeBSD
connection got stuck all of a sudden. Cannot ssh into the machine.
Ping ok, and apache2 apparently working. Console log-in takes +5 min &
nothing weird in /var/log/*.log. Restarted sshd a few times, no luck.
Restarted the machine, everything's normal. Two hours later, sshd
is weird again. Same symptoms. What are some possible culprits?
\_ NIS or NFS?
\_ Hmm... any chance you have a bad disk? sshd's virtual memory is
writing to bad blocks, which causes it to run very slow? Or the
blocks where your auth.log or something else that gets written to
on login? -dans
\_ NFS mounted home dir on remote file server. DNS lookup failure
on that NFS mount, or DNS reverse lookup failure on remote host
but the console login delay implies NFS failure. Or it could be
something entirely different. :-) But I'd check those two first. |
| 2006/4/4 [Computer/SW/Unix] UID:42659 Activity:high |
4/4 In Unix how does one figure out the amount of space a directory (all
it's files and subdirectories, recursively) takes up?
\_ man du
\_ man du gook
\_ mandu gook
\_ mashitt da.
\_ Thanks |
| 2006/3/25-27 [Computer/SW/Languages/C_Cplusplus, Computer/Theory, Computer/SW/Unix] UID:42432 Activity:nil |
3/25 The Ultimate Casio Watch:
http://www.watchreport.com/2006/03/review_of_the_m.html
\- i used to have an older incarnation of the Pathfinder [till it
fell off at Mughal Sarai Junction] and a couple of comments:
1. it is good they have changed this to a strap rather than a
bracelet design. bracelet not good for these kinds of watches. i
nearly lost mine after a b'let failure on a catemaranin 9ft
seas, which is not where you want to try and fix something with
<1mm springs. 2. i find the themometer of limited use, since
while on your wrist you are really measuing the "temperature
near you wrist". 3. if you really need a compass, you need to
take a real orienteering compass ... these watch compasses are
sort of just opportunistic things. 4. some of the featuers like
altitude alarm, altitude recording, multiple alarms i dont find
that useful [you cant upload the telementry data to a computer
can you ... GPS mre useful in this area] and these things tend
to make accessing the frequently used function slower and more
complicated (like say converting between ft <-> m ... which was
my main complaint with my old watch 5. often you can get an
older model for vastly cheaper by just sacrificing some function
you may not really need [you can get some older p'finder models
for around $85 vs $300+ for this one, which is significant if
you think of this as a piece of gear rather than your
watch]. thanks for posting the link. i am happy to give advice
on outdoor gear except for aid climbing and snow and ice. --psb
\- i used to have an older incarnation of the Pathfinder
[till it fell off at Mughal Sarai Junction] and a couple of
comments: 1. it is good they have changed this to a strap
and not a bracelet design. bracelet not good for these kinds
of watches ... nearly lost mine after a b'let failure on a
catemaran ... which is not where you want to try and fix
something with <1mm springs. 2. i find the themometer of
limited use, since while on your wrist you are really measuing
the "temperature near you wrist". 3. if you really need a
compass, you need to take a real orienteering compass ... these
watch compasses are sort of just opportunistic things. 4. some
of the featuers like altitude alarm, altitude recording, multiple
alarms i dont find that useful [you cant upload the telementry
data to a computer can you ... GPS mre useful in this area] and
these things tend to make accessing the frequently used function
slower and more complicated (like say converting between ft <-> m
... which was my main complaint with my old watch 5. often you can
get an older model for vastly cheaper by just sacrificing some
function you may not really need [you can get some older
p'finder models for around $85 vs $300+ for this one, which
is significant if you think of this as a piece of gear
rather than your watch]. thanks for posting the link. i am
happy to give advice on outdoor gear except for aid climbing
and snow and ice.
\_ I wanted to get this watch b/c it was the first casio I've
seen that has world atomic timekeeping, is solar powered,
water resistant and it has that "indiglo"-like lighting.
The compass was an added bonus, but I usually carry my eTrex
GPS so its not an essential for me. |
| 2006/3/24-27 [Computer/SW/RevisionControl, Computer/SW/Unix] UID:42408 Activity:moderate |
3/24 I want to hire an experienced Release Engineer to build our
Perforce system. We're converting from CVS. -- jsjacob@lyris.com
http://www.craigslist.org/eby/sof/140795729.html
\_ hey jsjacob, you should come across the hall and say hi to all
the sodans at wavemarket sometime.
\_ You could just BTK our build engineer.
\_ I rather BBFS your hot blonde HR hello nurse.
\_ BTK?
\_ Blind, Torture, Kill
\_ Bind
\_ Sure, right after I use the patio table again. Hi. -- jsjacob
\_ quick question. Why did you decide on Perforce, instead of (say)
Subversion?
\_ Not jsjacob, but perforce is a lot slicker the subversion
\_ Perforce has cool tools *right now* and I have a budget.
I think subversion will be equivalent in a few years but
I can't wait. -- jsjacob
\_ Okay, just out of curiosity, what are the features you liked?
\_ Views can be incredibly powerful if you use them well. For
instance, are you only working on part of the code base
and don't want to compile the rest? Open up your view
which uses the daily builds to populate most of your build
tree. Want to start working on another part of the code?
Just tweak your view (or use a different part of the view)
so you have to build the rest of the code. SVN's lack of
view support makes me sad.
\_ Views can be incredibly powerful if you use them well.
For instance, are you only working on part of the code
base and don't want to compile the rest? Open up your
view which uses the daily builds to populate most of
your build tree. Want to start working on another part
of the code? Just tweak your view (or use a different
part of the view) so you have to build the rest of the
code. SVN's lack of view support makes me sad.
\_ Proxy, graphical merge, permissions admin. -- jsjacob
\_ I don't know what "proxy" is for perforce, but graphical
merge and permissions admin exist for SVN.
\_ I know, but not at the same level. I know SVN has
similar tools and in a few years I'm sure I'll be
considering moving to SVN to drop the license fees.
This is proxy:
\_ I know, but not at the same level. This is proxy:
http://www.perforce.com/perforce/products/p4p.html |
| 2006/3/23-25 [Computer/SW/Unix] UID:42401 Activity:nil |
3/23 FYI, I extolled the virtues of Emacs TRAMP mode on the motd a little
while back. It still rocks out, but takes a little tweaking to work
effectively on soda. I was unable to get it working with tcsh, so I
changed my default shell to bash. I didn't put much effort in to
this. If you're married to tcsh, it's probably doable. More
importantly, you'll want to set '(tramp-chunksize 150) in your .emacs.
-dans |
| 2006/3/2 [Computer/SW/Security, Computer/SW/Unix] UID:42066 Activity:nil |
3/2 very bizarre pass login behavior on soda ... I am able to login
using an old password, and variations of that password ... anyone
ever heard of this behavior before? I emailed root ... is anyone
checking that now???? |
| 2006/2/20-23 [Reference/RealEstate, Computer/SW/Unix] UID:41930 Activity:nil |
2/20 Has anyone bought a place in the Bay Area and gotten a one year home
warranty. If so, who did you use? Email, website, ph # ?
\_ We did. American Home Warranty, I think. Totally worthless (not
that we paid for it (directly)) Took forever to respond, didn't
cover the stuff hat actually broke, didn't replace stuff, just
(sort of) repaired it.
\_ I got a warranty from First American
(http://homewarranty.firstam.com from our real estate agent who
wrote that into the bid. Took advantage of it for the central
heating system, drainage problems, and dryer. Response times
were reasonable. I can't remember exactly, but the deductible
was $40-$45 per claim.
\_ In my case, the seller did pay for 1-year home warranty through
American Home Warranty (it was around $450/yr). I negotiated w/
the seller's agent in order to get this, so it's not by default.
I didn't renew it after it expired since *knock on woods* my
appliances are still relatively new and the past owner was very
anal-retentive about keeping them pristine. (we got lucky). YMMV,
if you feel your appliances may be at their last leg when you
inherit them, might want to look into renewing it. (or do a math
comparison vs. buying a new one)
\_ You can get the home warranty from the seller/seller's agent
written into the bid, but if the market is competitive, you
may lose the property to someone else who has offered about
the same amount with no/fewer contingencies. The housing
market seems to have cooled down a bit now, though. |
| 2006/2/17-20 [Computer/SW/OS/Windows, Computer/SW/Unix] UID:41912 Activity:moderate |
2/17 What does you use for bug tracking? Does it suck? We use Bugzilla now,
but we'd like something that's more customizable and that provides
better reports, of a sort we can send to customers. -gm
\_ Siebel 2000
\_ I use Siebel too, it sucks
\_ We use bugzilla. Being open source, it is very customizable. We
pull data out of it to generate some pretty good statistics
and reports. --dim
\_ The problem with Bugzilla is that you can either customize it
or be able to upgrade easily; you can't have both, as far as
I can tell. As to reports, in my experience you have to write a
bunch of code to get customizable reports out of it. That's not
something we can point our sales people at to get what they
need. -gm
\_ Bug tracking? We don't make bugs.
\_ We just switched from Bugzilla to Jira. It is nice. -ausman
\_ We use Devtrack. It's pretty good at reports and summaries, so I
see why our PM loves it. But speaking as a developer, it sucks.
It is based on the idea of the admin setting up a very strict
work flow and you must fill out all the required form fields and
forward it to the next owner before they can then fill out their
fields and yadda yadda. I can see that for a PM with undisciplined
developers and QA staff, this rigidity is quite nice. But if you
just want to enter a quick issue like "Was not checking for a null
pointer in FooClass.barFunc(). Fixed." Devtrack makes it a major
pain in the ass. |
| 2006/2/16-17 [Computer/SW/Unix] UID:41884 Activity:nil 77%like:41882 |
2/15 I'm using rsync -ravH to make an identical backup. When
comparing the two resulting filesystems, I notice that some
directory files have different sizes. /usr/doc/. is 8192 and
/backup/usr/doc/. is 4096 Any idea why this would be? I would
like the two rsync'd filesystems to be identical sizes.
\_ -r is included in -a for rsync.
\_ Read up on filesystems.
\_ Directories grow as there are more entries at that immediate level
of the structure, then don't always give back that allocated space.
Try making a directory, then creating 10,000 files in it, then
deleting them all. Note the directory size. --dbushong
\_ Thanks! -op
\_ Also, different OSes/FSes do initial allocations differently.
If both these directories are empty, that would explain the
discrepancy.
\_ In this case they are both linux and ext3.
\_ Same linux versions? flavors?
\_ Indeed: same computer. The issue seems to be
the directory size that dbushong explained. I wonder
what is the best way to verify that these directory
file sizes are the only differences between the FS -op
\_ try another tool. unison? |
| 2006/2/15 [Computer/SW/Security, Computer/SW/Unix] UID:41851 Activity:low |
2/15 Can one of you root guys please explain how tom has been
eearily and correctly identifying anonymous motd posters? Is he
abusing root or abusing his connection to root?
\_ I don't think tom's been abusing root. Unfortunately, it's not
too hard to identify most motd posters even without root. If you
have ideas for how to make it harder, please let us know. --root
\_ why should it be hard to identify who is posting to the MOTD?
Do we really think the MOTD is a better place for not having
the basic auditing capability that every other forum on the
net has? -tom
\_ I bet tom doesn't need technical means to know identify most
posters.
\_ Several of us have various scripted ways of figuring out who other
posters are but only a childish schmuck would descend to that
level in a cheap attempt to 'score' points on the motd. And his
predictive abilities are hardly "eery". He has a terrible track
record of identifying people especially considering how often he
names names.
\_ ooh, big bad tom naming names... Grow up, reiffin.
\_ nice try but wrong. |
| 2006/1/31-2/2 [Computer/SW/Unix] UID:41609 Activity:nil |
1/27 Is there some common unix tool to chop off the head of a file up to
some marker? Other than perl etc.
\_ sed or awk
\_ what the pp said. or tail if you know the line number, and
grep can tell you the line number. actually, there's an
ungainly grep way to do it.
\- hasnt this come up before?
sed -n '/<regexp>/,$p'
but that will operate on lines (as will grep).
--partha "sed" banerjee
\_ Ok I just wrote a little perl script. Basically this tool
takes two files and diffs them, but you can tell it
a different "differ". I need it to ignore these files'
variable length headers. So my diff script processes them into
temp files and calls diff. Is there some clever way to
get that done in a commandline substitute for diff?
\_ look up "named pipes" if you want to avoid temp
files... other than that i don't think there's
much you can do
\_ tail +/marker -- for a reasonably modern version of tail.
\_ It looks like "tail +" takes an offset rather than a
match. What version of tail are you talking about? |
| 2006/1/25-27 [Academia/Berkeley/CSUA, Computer/SW/Security, Computer/SW/Unix] UID:41509 Activity:nil |
1/25 to root: just curious... what might be the causes of recent
Soda unstability? are you guys doing something that may crash soda?
are you guys trying to fix something?
\_ The root of the problem is that the root used to be run by
experienced late 20/early 30 something folks, and when the
root was handed down by the new gen-Y 20 year old kids, they
don't know how to run the system. In fact they prefer soda
running on Windown XP.
\_ http://csua.org/u/erg
Rest assured, when Soda recovers from its Jan. 24th funk, it
will be much happier and stable. -mrauser
\_ I prefer the more straight-forward approach of the VP bat.
- jvarga |
| 2006/1/21 [Computer/Theory, Computer/SW/Unix] UID:41467 Activity:nil |
1/21 Is there a good way to write a log file in a multi-threaded system
that doesn't require locks? It seems like there should be some
sort of atomic fsync() or something. |
| 2006/1/19-21 [Computer/SW/Unix] UID:41443 Activity:nil |
1/19 Anyone know of a good neighborhood website? Like a sort of discussion/
photos/party planner/for sale site etc. specific to a fairly small
area (few blocks?). I'd love something like that. And before anyone
says "just walk next door you geek", this is impractical for all sorts
of time-logistical reasons.
\_ In other words, you're just a lazy, fat geek, AKA The Comic Book
Guy.
\_ Yahoo Groups was practically designed for this. I built one to
accommodate our apartment block. Getting people to actually use it
is another matter....
\_ http://www.meettheneighbors.org
the guy who runs it is a dweeb ... but this is exactly what
you want |
| 2006/1/16-18 [Computer/SW/Unix/WindowManager, Computer/SW/Unix] UID:41385 Activity:nil |
1/16 Since X11 forwarding is gone ; my emacs key definitions no longer
work. I was wondering if anyone could point me to a setup that would
at least define basic functionality for emacs inside an x-term. Thanks
\_ Maybe you could mail root and ask them to turn X11 Forwarding
back on.
\_ Or maybe not.
\_ Just delete/rename your ~/.emacs and you'll get the default key
binding which works the same in a text terminal and in an xterm.
\_ Possibly your default xterm is lame (like on Solaris) and doesn't
do Meta, which breaks all kinds of things for emacs. In that
case, get a better xterm.
\_ I am using freebsd 5.4/kde 3.5 and konsole/xterm.
\_ Meta works fine in my xterm on Solaris. --- !OP
\_ Solaris 8. Maybe it's fixed in later versions. |
| 2006/1/14-17 [Computer/SW/Unix] UID:41375 Activity:low |
1/13 Does anyone know a free server side HTTP parser? (In C/C++)
\_ http://www.google.com/search?q=HTTP+parser
\_ Yeah, none of those seem to do what I need.
\_ speaking of which, does anyone know of commandline unix util that
will strip all html from a text file?
\_ lynx
\_ sed
\_ outsource it.
\_ Try one of these (html2text):
Perl: http://www.greenend.org.uk/rjk/2000/10/html2text.html
C++: http://www.mbayer.de/html2text
\_ http://www.zazzybob.com/bin/striptags.sed.html
\_ What do you mean by HTTP parser? Are you looking for a request
parser or something else? Maybe libcurl will work for you.
\_ Yes, a request parser. Looking at libcurl now, it may work
for me, but I haven't yet found exactly the right code. |
| 2006/1/5-9 [Computer/SW/Unix] UID:41259 Activity:nil 50%like:41255 |
1/5 Any idea when SED monitors might hit the market?
\_ From the first google hit article about SED monitors:
"UPdate 10/19/05:The first monitors of this type might be released
by Toshiba in 2007. Canon speaks of computer monitors end of 2006,
early 2007." --dbushong |
| 2006/1/5 [Computer/SW/Unix] UID:41255 Activity:nil 50%like:41259 |
1/5 How long until SED monitors will be available? |
| 2006/1/3-5 [Computer/SW/Unix] UID:41211 Activity:moderate |
1/3 what's the best terminal program (I just want to ssh from my laptop
to soda) for macosx?
\_ don't like terminal?
\_ What's wrong with /Applications/Utilities/Terminal.app?
\_ iTerm looks/works well for me. -scottyg
\_ second.
\_ Terminal doesn't have mouse support.
\_ What do you mean by "mouse support"? If you want cursor
positioning, you can turn it on via the window settings.
\_ Terminal.app does not have xterm mouse support. iTerm does not.
Glterm does, but a new version has not been released in years.
\_ Terminal.app does not have xterm mouse support. iTerm does
not. Glterm does, but a new version has not been released
in years.
\_ you don't need xterm mouse support to ssh. -scottyg
\_ Who are you to tell me what I need or not? xterm mouse
support doesn't work in Terminal.app. it DOES work in
xterm when running X11 on the mac. it works in Glterm.
I do not think you know what xterm mouse support is.
\_ Who are you to tell me what I need or not? xterm
mouse support doesn't work in Terminal.app. it
DOES work in xterm when running X11 on the mac.
it works in Glterm. I do not think you know what
xterm mouse support is.
\_ You need to limit your posts to 80 columns.
\_ Fixed. -formatd
\_ I'll bite, what is xterm mouse support? Is
this the selecting text copies and middle
button pastes "feature"?
\_ for instance run a text browser on Gmail.
you can click on things with your mouse if
you have proper xterm support.
\_ I still can't figure out what you mean.
When I click on stuff in lynx it just
copies it (I'm using an xterm in S10)
or pastes it or brings up a menu.
\_ Uh, assuming you're op, you said "...just want to
ssh..." So scotty was correct, and you're a twit.
\_ Sorry, I thought you were not the OP and that you were\
adding additional requirements. I also use a program
called xterm when I need X11 support, but I don't
generally like the look or feel of it. -scottyg
\_ Sorry, I thought you were not the OP and that you
were adding additional requirements. I also use a
program called xterm when I need X11 support, but
I don't generally like the look or feel of it.
-scottyg |
| 2005/12/24-28 [Computer/SW/Unix] UID:41136 Activity:kinda low |
12/24 What sorts of things are more appropriate to put into .bashrc instead
of .bash_profile?
\_ Check the man pages for the difference between login vs non-login
and interactive vs non-interactive shells. Would you need to check
your mail or start up irc or a news reader on a non-interactive
shell? Would you want irc to startup in every new shell or just
the initial login?
\_ I know what difference is, but it's not clear to me which
environment variables are best suited for where. PATH in
.bash_profile? aliases in .bashrc? etc.
\_ Would you need a $PATH in a non interactive shell? Would
you like access to anything more than the default paths?
Possibly. Would you like aliases in a non-interactive
shell? Some, possibly. This will vary by environment.
No one can tell you what you need in yours.
\_ Well, of course it will vary. I just would like to know
what the conventions are and people typically do. As for
interactive vs. non-interactive, that's not really an
issue anyway since I (and presumably most other people?)
gate lots of things on $PS1.
\_ Its probably better to check if $- includes 'i'
rather than $PS1.
\_ Seriously, people do what they need to do. That is
the convention. This is unix. There is no one right
way to do most things. |
| 2005/12/22-24 [Computer/SW/Unix, Computer/SW/OS/Windows] UID:41124 Activity:nil |
12/22 On XP, I have folders using Chinese characters. I also have
file names using Chinese characters. When I do a samba mount
from my Linux to XP, I can do everything if the file name
isn't using Chinese characters. How do I get around this
restriction? Thanks.
\_ you need to be specific on your XP's file system. FAT uses
legacy encoding. NTFS uses some sort of Unicode encoding
(I think it's UTF-16). I vaguely remember newer version of
samba handles unicode encoded filename rather well. |
| 2005/12/20-22 [Computer/SW/Unix, Computer/SW/OS/OsX] UID:41097 Activity:nil |
12/20 I'm trying to determine the cause of an curious behavior: When
I login to a Linux box from my Mac (rxvt in X11), there is
some lag on the text output during the printing of multiple
lines of output. When I login to a FreeBSD box from my Mac,
there is no lag. Curiously, when I login to either the FreeBSD
box or the Linux box from my Linux laptop, there is no lag
either. Any ideas what differences between the network
settings could cause this. Could it be buffered I/O or minimum
packet size? It happens with telnet and ssh.
\_ Are your TERM settings the same on both?
\_ yep TERM=xterm
\_ Have you tried this on multiple Linux boxes or just the one?
\_ yep, tried it on about 4 debian boxes.
\_ perhaps you might try some other telnet programs on your mac? |
| 2005/12/18-20 [Computer/SW/Unix] UID:41067 Activity:nil |
12/18 Bash 3.1 is out:
http://cnswww.cns.cwru.edu/~chet/bash/README
ftp://ftp.cwru.edu/pub/bash/bash-3.1.tar.gz |
| 2005/12/7-9 [Computer/SW/Mail, Computer/SW/Unix] UID:40899 Activity:nil |
12/7 So what are the correct SMTP settings for soda now? I've been
using port 465 + SSL + password authentication and it's stopped
working since the reboot.
\_ I just mailed myself using port 25 + SSL + password.
\_ % telnet http://csua.berkeley.edu 25
Trying 128.32.112.233...
telnet: connect to address 128.32.112.233: Connection refused
telnet: Unable to connect to remote host
\_ I just did that too and got a connection.
\_ many isp's block port 25.
\_ if you figure this out, please let me know, because I can't do
it neither. I can only email to myself. |
| 2005/12/7-9 [Computer/SW/Unix] UID:40898 Activity:low |
12/7 After I start a background job from tcsh in an xterm, I try to exit
the shell. But it stays around until the background job exits. Is
there a way to tell the shell not to wait for the background job to
finish? Thx.
\_ Yes.
\_ How?
\_ nohup?
\_ you might need to redirect to /dev/null also.
\_ It still doesn't work. What I do is:
1. ssh into a Linux host. My shell is tcsh.
2. Type "nohup sleep 10 >& /dev/null &"
3. Type "logout"
The session doesn't close until "sleep" terminates. I also tried
/usr/bin/nohup and /bin/sleep, but got the same result.
/usr/bin/nohup and /bin/sleep, but got the same result. --- OP
\_ maybe your tcsh has some customization that makes it wait.
check the startup files. the default tcsh behavior is to
disown jobs started with & upon exit. either that, or your
tcsh is broken.
\_ More info: I just found that this problem only happens if
I ssh into the host. It doesn't if I rsh/rlogin/telnet
into the host. --- OP |
| 2005/12/4-6 [Computer/SW/Security, Computer/SW/Unix] UID:40845 Activity:nil |
12/3 Free rootcow!
\_ Freed. --mconst
\_ What does this mean?
______________________________
< Someone may be abusing root! >
------------------------------
\ ^__^
\ (oo)\_______
(__)\ )\/\
||----w |
|| || |
| 2005/12/2-2006/1/5 [Computer/SW/Unix] UID:40825 Activity:nil |
12/2 UPCOMING DOWNTIME: Weekend of 12/2
Soda will be brought down for several hours this weekend for routine,
or not-so-routine maintenance. SSH keys will change. IRC bots will
lose their marathon uptime. Cake will be eaten. - jvarga |
| 2005/11/22-23 [Computer/SW/OS/Solaris, Computer/SW/Unix] UID:40690 Activity:nil |
11/22 Is there a native process in Unix (SunOS 5.7) that logs port traffic?
Barring that, do you know of any good software that would do the trick?
SdTFW for "logging port traffic unix" but not finding what I'm looking
for. TIA. --erikred
\_ tcpdump -w
\_ snoop. Ethereal. -John
\_ snort?
\_ You want something solaris native, 'snoop' is it.
\_ Once again, motd >> Google (and my google skills). Thank you.
--erikred |
| 2005/11/20-22 [Computer/SW/Unix] UID:40666 Activity:nil |
11/20 mac users: I just got my first mac (from a friend, wanted to check
it out) do you recommend su'ing to root and poking around to see
whats going on? apple "strongly advises against" it ... but just
wondering if thats because most of their user base is not familiar
with unix ... also, is there a default root password? the person
I got it from did a clean install before they gave it to me (not
sure if that process prompts for a root p-word) thx
\_ The root account is disabled by default. But sudo works great.
Curious why do you need to be root to see what's going on?
\_ maybe I dont? i just feel like I should have root on my
own machine ... but maybe thats not neccesary for working
w/ darwin? I'm thinking like, eg, ps -a does not seem to
be showing me all proc's
\_ er, ok, sorry. re: the above, ps -aux is doing what I
want ... so you're saying I dont need root to effectively
administer this machine? thx
\_ all you need is sudo
\_ if you really want a root shell: sudo /bin/sh
\_ sweet thanks. so all the passwords in etc/passwd are starred
and there is no shadow file ... I guess that means no user
accounts can be logged in? I was also suprised to see that
the osX user accounts have no entries in the passwd file
but I guess it was naive to assume they would be there...
so its seeming to me like the unix core does less than i
thought and there is more custom mac stuff on top ... what
do other mac users think? is messing with my system as
root just a surefire path toward instability? or can I
play with things at that level. thx
\_ osx uses netinfo to store user passwds; netinfo was
the next version of nis/ldap.
Re messing around as root: messing around as root
is a sure fire way to instability on any *nix system.
There is nothing you can't do via sudo.
\_ There is nothing wrong with suing to root a Mac OS X system if
you understand what root account is for. However, Apple has
disabled the root account becuase they were afraid that some
people would try to login as a root user into the GUI session
which can be a very dangerous thing to do. You can enable the
root account from the netinfo manager. Note that only the
users who are in the admin or wheel groups can su to root
(so if you add a non-admin account for yourself, you'll have
to add yourself to the wheel group through the netinfo manager
which doesn't seem to be documented well...) |
| 2005/11/17-20 [Computer/SW/Unix] UID:40632 Activity:low |
11/17 I'm about to buy one of the condos in a big condo complex and
I'm concerned about lighting. I'd like to know at what angle
will the sun shine through the windows and if it'll be blocked
by other units (which by the way tend to protrude out). I'm
wondering about the path of the sunlight in respect to the
season. What is a good web site that'll tell you these things?
And since there is a 15 degree difference between true north
and magnetic north, which one do people use? Thanks.
\_ Just go there at the time of day you're worried about.
\_ The sun's path through the sky is different each day.
\_ true. only USGS maps and similar care about magnetic
\_ http://www.srrb.noaa.gov/highlights/sunrise/azel.html
http://aa.usno.navy.mil/data/docs/RS_OneDay.html
The sun uses geographical (true) north.
\_ I used to have an app on my Palm that would show you a true-north
compass if the time was set correctly and you input your latitude.
All you had to do was point it toward the sun.
You could use CyberSky to view the sun's position from the city and
that would give you those details, etc.
\_ ObHousingBubble
\_ If you really want to obsess about it,
http://www.solarpathfinder.com. -tom
\_ Best solar path visualizer: http://solar.anu.edu.au/Sun/SunPath |
| 2005/11/16-18 [Computer/HW/Memory, Computer/Networking, Computer/SW/Unix] UID:40627 Activity:nil |
11/16 Your friend's kid's gadget is cooler than yours
http://www.nytimes.com/2005/11/17/technology/circuits/17pogue.html
\_ That's awesome!
\_ Wait how does it know what you clicked on? Don't you need
to store the picture somewhere (in memory) and an input
method (like a tablet)?
\_ I got to play with one of these a couple of months ago before it
came out. It is pretty cool, although the one problem it has it the
only feedback it can give is audio. Thus, it wouldn't be very easy
to use in a classroom w/o disturbing everybody. Also, in order to
use it, you need to write on special paper that costs about $1 a
sheet. Its covered by very tiny dots so the pen knows where it is
located on the paper at all times. But I will say that their future
plans for the Fly pen seem really cool (wireless sync and such).
\_ It comes with headphones and paper is about 8 cents/sheet.
\_ I think reading its manual alone will make me dizzy. |
| 2005/11/14 [Computer/SW/Unix] UID:40578 Activity:high |
11/14 Unix newbie question:
How do you read / rename a file with name "-l" (dash-el)?
\_ I've always loved this question. "./-l"
\_ Sounds like CS9E homework.
\_ Aw, crap. you're probably right.
\_ Here's a better one: list all of the files present in a directory
(which may have subdirectories) which contain the character "-"
\_ Even better. How do I delete that file named "-rf"
\_ 'find' with the right options is the answer to everything. you
just need to find the right options. :-)
\_ man 3 getopt
\_ Uh. man rm. man mv. man cp. You shouldn't need to use
find or getopt or anything.
\_ cd [dir w/ file] && mv `pwd`/-l [new name]
\_ why use pwd instead of "."? |
| 2005/11/12-14 [Computer/SW/Languages/Web, Computer/SW/Languages/Perl, Computer/SW/Unix] UID:40555 Activity:nil |
11/11 Hey MOTD, I'm looking for a good webhosting service. A coworker has
recommended http://dreamhost.com. Does anybody have any first hand experience
with that provider? Does anyone have any suggestions, warnings or
general advice? Ideally I'm looking for a place with a unix shell,
php, perl, and possibly mysql & cron -- but it doesn't have to have
heavyweight bandwidth, etc. TIA. -mice
\_ It may be more money & work than you wanted, but I've had good
luck w/ http://johncompanies.com. You get root on a jailed system: FreeBSD
w/ ports tree available or some Linux distro w/ some pkg system.
So you can run any sort of mail/web/anything server you want.
Very responsive support. But it's like $30 - $70/month. --dbushong
\_ Is that how much you pay for http://csua.org?
\_ No, currently I just run it off my home DSL. I'm thinking
about getting either a colo'ed box or something like this
one of these days just to have the option to scale a web app
to higher bandwidth on short notice should I ever actually
manage to market anything successfully (e.g.
http://floatingsheep.com/wishlist ), but for now it's not
quite worth the expense. --dbushong
\_ I have friends who use dreamhost (and one of them gave me a login).
AFAICT it seems to be pretty good (unix shell, perl, php, mysql,
and I assume cron). If I had to pay for hosting myself, I'd
probably go with them. --jameslin |
| 2005/11/7-8 [Computer/SW/Unix] UID:40465 Activity:nil |
11/6 With regard to blogs (or any other websites), is it better to
specify images using a relative path or an absolute one? Thanks.
\_ Relative.
\_ I think absolute is friendlier to RSS aggregators, especially
web-based ones such as bloglines.
\_ Now if only bloglines could handle dates that are in local time
but with the proper offset (e.g. -0700) specified, and not treat
all times as GMT. (They're valid per the RFC!) |
| 2005/11/4-8 [Computer/SW/Unix] UID:40433 Activity:nil |
11/3 Has anyone noticed how shitty citysearch is? their website is
incredibly slow, non-intuitive to use, and very inflexible.
does anyone else have this problem?
\_ I think it's great. What I do is start with an address, then
pick a food category, then a price level, then sort by distance
(for restaurants). |
| 2005/11/3-4 [Computer/SW/Unix] UID:40412 Activity:low |
11/3 What is a good way to delete all the lines in a file after
a regular expression match. For example, if I want to do a
"last" and get output only for this year or month, so I want to
delete all the lines after "Dec".
\_ sed /Dec/q file
That'll delete everything after (but not including) the first
line matching "Dec". If you want to delete that line too, you
can use sed '/Dec/,$d' instead. --mconst
\- while a fine answer, awk '/Dec/{exit};{print}' may run
faster than sed '/Dec/,$d' on large input if the match
\- while a fine answer, awk '/Dec/{exit};{print}' will run a
lot faster than sed '/Dec/,$d' on large input if the match
is early. awk may be easier to tweak to make the matching
more precise as well. The POWER of AWK. ok tnx.
more precise as well. ok tnx.
more precise as well. The POWER of awk. ok tnx.
\_ Indeed. You could also use sed -n '/Dec/q;p', if you're
not ready for the POWER of AWK. --mconst
\_ The easiest way, of course, is to assign this to an intern or the
new guy. |
| 2005/11/1-3 [Computer/SW/OS/Solaris, Computer/SW/Unix] UID:40376 Activity:nil |
11/1 what's the name of a general login machine in berkeley cs?
please email dpetrou@cs.cmu.edu. -dpetrou
\_ argus
\_ http://inst.eecs.berkeley.edu/cgi-bin/clients.cgi
That is a list of everything. -mrauser
\_ failure to read directions. |
| 2005/10/31-11/2 [Computer/SW/Unix] UID:40361 Activity:nil |
10/31 CIFS sure seems to work as advertised alot better than NFS, isn't
this a pretty big advantage of living in MS land vs. linux/solaris??
\_ I'm going to cry.
\_ please don't cry. Tell us why this makes you want to cry.
\_ I'm not the one who originally said "cry" but when I was
at Decru several times a day I would hear the engrs
implementing the CIFS side of things yell "are you
shitting me?" Apparently not even MS follows the CIFS
specs correctly. It may look like it works well but
behind the scenes it's REALLY ugly. CIFS may be decent
but it's overall ugly.
\- i was speaking to that samba carter fellow and asking why
somethign was designed the way it was. he explained
"in cifs ... locking ..." and i said "i must be missing
someting because that is insane and doesnt make sense."
then he told me "nope, you've understood it correctly". |
| 2005/10/28-31 [Computer/SW/OS/Solaris, Computer/SW/Unix] UID:40320 Activity:moderate |
10/28 Docs say I can't do dd from disk that is in use.
This seems (based on my bad memory) to be contrary to my experience.
Any opionions on duping an inuse disk (to a non-inuse one).
this is on solaris 8.
\_ live update, ufsdump, rsync, dd... you want your filesystem
as quiesced as possible though for all of the above
\_ what the above and others say about quiet drives and RO and then:
since you don't tell us what problem you're trying to solve, my
very generic advice is this: dd the drive so you get boot blocks
and other fun stuff you might want, then do an rsync after that just
for kicks in the hopes of catching a few of the files that might
have been in use while doing dd. no, this followup rsync will not
make a perfect mirror on an active disk but if you wanted to do a
real backup, you'd use real software for it or at least take the
source drive offline. you're well into kludge territory already,
anyway.
\_ You can dd from a disk that is in use. You will obviously not
get an exact duplicate and, as above, I would not try this
on a very busy filesystem. However, it works fine under Solaris
9/10 and Red Hat Linux. I've done this to make 'backup' disks
(successfully) many times.
\_ If you need to keep the filesystem mounted RW, I'd suggest
using rsync rather than dd. If the filesystem is mounted RO
it seems like dd would be fine.
\_ Rsync is slower and won't grab the boot blocks (if you want
them).
\_ rsync is NOT slower than DD unless the filesystems are
basically full. -top
\_ Yes, it is. Best case scenario (using the right
rsync options) they are about equal. "dd" can use
the raw device and bypass the filesystem entirely.
Amended: You are right that if the filesystem/disk is
not very full then "dd" will be slower merely because it
is copying "empty" space but not because "dd" is
intrinsically slower.
\_ What problem are you trying to solve?
What are your priorities?
\_ so it's not "intrinsically slower", it just takes
longer to run? Get a clue. -tom
\_ I think you need the clue, tom. "dd" is
faster. If you want to prune the list of files
you want to "dd" first it will kick the butt
of "rsync". You know that, too, but you just
like to argue. When copying the exact same
volume of data, "dd" is faster.
\_ While driving a car is faster than walking,
walking from Telegraph and Bancroft to
Telegraph and Durant is much faster than
driving. "dd is faster" is meaningless
unless the problem is well-defined. -tom
\_ Even your example is false.
\_ Alright, we're convinced: you're an idiot.
You don't have to try this hard just for
us.
\_ You may as well say that a plane
is slower than walking, too. The
statement about a 'well-defined'
problem is your own parameter but
doesn't change the fact that
planes are faster than walking.
\_ To put this in a way that even
an idiotic computer geek might
understand, if you have two
different sort algorithms,
one that's usually faster in
practice but O(N^2), and one
that's usually slower in practice
but O(N log N), it makes no sense
to say that one is intrinsically
faster than the other. It
depends on the problem you have
to solve. -tom
\_ HAMMER GOOD!
If you took a second to think, you'd realize
you're clueless. Or maybe not. It's dim.
\_ Do you know what the word 'intrinsic'
means?
\_ Yes. You seem to have trouble with the
word 'slower', though.
\_ "dd" is intrinsically faster.
\_ Not for the problem posed (ergo
"HAMMER GOOD"). If you want to
keep spouting your technically
correct and completely irrelevent
point, by all means... You will
still be wrong.
\_ You have come to know that which
is tom.
\_ hey moron: he's replying to
!tom.
\_ We don't even *know* what the
problem posed is yet. However,
we do know the guy wants to
use 'dd' to solve it. Rsync
may not even be a viable
solution in this particular case.
\_ What problem are you trying to solve? What are your priorities?
\_ try using solaris' Volume Manager (also known as disksuite).
create a one way mirror containing the current live FS as the
initial submirror, then attach the second empty FS as another
submirror. wait for the second submirror to sync up, then at
your leisure, detach the second submirror.
look at metadb, metainit, metattach, metadetach, metaclear,
and other "meta"commands. --Jon
look at metadb, metainit, metattach, metaoffline/online,
metadetach, metaclear, and other "meta"commands. --Jon
\- i have a solaris 10 sunblade 1500 which some colleages
wanted to clone to save time. i rebooted my machine and used
/usr/bin/dd if=/dev/dsk/c0t0d0s2 of=/dev/dsk/c0t1d0s2 bs=1000k
and thne fscked. all the machines work fine ( 5 clones). it
sounds like you know what the issues are so just take this an
a data point. 73gig disks ... i think it tok 40min or so.
oh i am sure i booted it su and killed off most of the
processes before running dd.
\_ The thing with the Solaris LVM method is that you don't
really need to have the filesystem quiescent. You get a
valid, consistent snapshot at the time of metaoffline or
metadetach. --Jon
\_ As long as you understand this isn't considered "production
quality", if that worked for you, great. I wouldn't do that
for something going into a data center 24x7, etc.
\- i bet lot of production operations were running on
less reliable windows installations (i dont know how
reliable windows OS is today). anyway, these are
cheep sun workstations with ide disks, so what do you
think. anyway, it seemed to be the OP was aware of the
"issues" and was just interested if it was "crazy talk"
[you'ld be lucky if it worked] vs. 99%chance it should
be fine. hence the "datapoint x5" comment.
\_ who said that just because you put a windows box in
production means its production quality? thanks for
pointing out they're workstations. that was so unclear.
i was pointing out that this isn't production quality.
sheesh.
\- are you a frequent visitor to Casino Troll?
\_ And what is "production quality"? This is essentially
what tools like "System Imager" do. What are you
afraid of? 'cksum' the OS if you want to. Bits
is bits.
\_ PQ: built correctly, not a hack job with dd
from another box. Something you'd bet your
career on. Workstation for the guys? Whatever.
Server in the colo running critical db? No way
is that getting built from a dd clone off a hot
drive. Building a few dozen/hundred/whatever
web/whatever servers? Jump/Kick Start. I've
done the clone thing for workstations. I'd
never do it for servers. If you want to discuss
it in detail, post your name and we'll take it
to email. |
| 2005/10/27-29 [Computer/SW/Languages, Computer/SW/Unix] UID:40293 Activity:nil |
10/27 http://news.yahoo.com/s/ap/20051027/ap_on_bi_ge/earns_oil Exxon Mobil, Shell Post Record Profits \_ Exxon first U.S. company to have $100 billion in quarterly sales! USA USA USA! |
| 2005/10/27-29 [Computer/SW/Security, Computer/SW/Unix] UID:40291 Activity:low |
10/27 Okay, is ftp completely gone? I'd search the motd archives but,
wait for it, there are none.
\_ try sftp or scp. most sftp clients that I've used have scp
support for transfering multiple files or directories.
\_ Hm, does WS_FTP do sftp?
\_ Use Filezilla.
\_ putty has a free command line scp binary that I use all
the time. I've never tried their sftp client, but it can
be found here: http://www.putty.nl/download.html -sax
\_ See section 3 of last week's minutes. -gm |
| 2005/10/25-26 [Computer/SW/Unix] UID:40270 Activity:nil |
10/25 Suppose I have directories foo, foo/bar, and foo/bar/{a,b,c}. (In
actuality, there are a lot more files.) I'd like tar to archive
everything under foo, except for foo/bar/*, except I do want to keep
foo/bar/a. Unfortunately, tar's exclude patterns seem to take
precedence over any include patterns, so if I exclude foo/bar, I can't
then include foo/bar/a. The easiest workaround I see is to just exclude
foo/bar, and then call tar a second time to append foo/bar/a. Is there
a way to do this with a single call to tar?
\_ Use find to generate a file or dir list, your favorite filtering,
and tar --files-from=<file> to archive the exact list? append
will yield basically the same thing, due to lack of indexing in
tar format. |
| 2005/10/25-26 [Computer/SW/Unix] UID:40267 Activity:nil |
10/25 Emacs question: after I do "M-x grep -n -e foo -e bar myfile", how do I
repeat the last grep command? The help says there's a history list.
Thx.
\_ hit the up arrow
\_ Thanks! I only tried C-p but didn't think of the up arrow.
\- who are you? --psb
\_ Are you just looking for M-x repeat-complex-command or something
more specific to grep. apropos is your friend.
\_ Just for grep. For example, I want to pull up the last command,
add "-e baz" in the middle, then run it again. |
| 2005/10/16-19 [Computer/SW/Security, Computer/SW/Unix] UID:40126 Activity:nil |
10/16 I accidently overwrote a file in my home dir. Is there a process
where I can request the version of this file from, say, 1 month
ago? Or are there even backups/archives like this at all?
\_ mail root
\_ Yes, backups do exist. Right now, they are not mounted, so
you will need to email root. Be aware that backups do rotate out,
and are currently being sporadically manually done, so email sooner
than later -- njh (the guy who runs backups)
\_ Thanks!! Now that I think about it, I might actually have my
own backup from the time I want, though it would be on a PC
that I don't have access to today. I'll check for my own
backup before emailing root, but it's good to know that root
can help me if necessary. Thanks! -op |
| 2005/10/15-16 [Computer/SW/Security, Computer/SW/Unix] UID:40104 Activity:nil |
10/15 Here is a proposal, a compromise for both parties. Split
/etc/motd.public into two files-- one is /etc/motd.civil
which is logged and viewable by root only, and is viewed in
default .login. The other one is /etc/motd.wild, which is
unlogged and is pretty much like our current motd.
\_ The problem with this "solution" is that it does not fix
the problem of threats, slander, etc, from the point of
view of the politburo. They are still responsible for
hosting it. -ausman
\_ Your welcome to create ~user/cesspool.motd if you really want a
place where you can be threatened at will by anyone. Root will
not breach the anonimity of the logs unless there is a specific
post which requires it. -mrauser
\_ I have a better idea. We'll have one file called /etc/motd.public
which is an open forum for discussing politics, fundamental
computer science, the computer industry, general science, sex,
and the meanining of life in a lively, free form, while also
posting timely links about current events and giving recent grads
a leg up on their careers. Then we'll have another file called
/etc/motd.jamf, where a small group of people can have a
carefully logged and moderated discussion of vi/EMACS, the
latest linux kernel and monty python. Anyone who mentions
politics, sex, violence, industry, uses a swear word, or says
anything remotely useful or interesting on /etc/motd.jamf
will recieve a demerit. Three demerits will banish them forever
from /etc/motd.jamf. |
| 2005/10/14-2010/9/30 [Academia/Berkeley/CSUA, Computer/SW/Security, Computer/SW/Unix] UID:40095 Activity:nil |
**/** Do not mail individual members of root for assistance. You will be
ignored! Your root staff are: steven, edilaic, mconst, jvarga,
mikeh, mrauser, kimbrel, toulouse, vaheder
Your Politburo are: kimbrel (P), steven (VP), toulouse (S),
yns88 (T), vidya (L), steven (E), bordicon (A)
Your new Politburo are: toulouse (P), steven (VP), eyung (T),
stevenk (S), sakura (L), dw5ight (E), scotspin (A)
The uncensored messages below this line may not reflect opinions of the CSUA. |
| 2005/10/14 [Computer/Theory, Computer/SW/Unix] UID:40092 Activity:nil |
10/14 If only we had some sort of forum to discuss having some sort of
forum so we could disc---... heh sorry. :-) |
| 2005/10/13-14 [Computer/SW/Unix, Computer/SW/Security] UID:40061 Activity:very high |
10/12 [moved to top]
I strongly suggest everyone read the minutes from the last
meeting. Both changes to the motd and soda itself were
discussed. -jrleek
\_ (Put up front since it's relevant) One thing that was left out of
the minutes is this: although we decided to enable logging of the
MOTD, we would like the implementation to be put in place by the
users OF the MOTD. The decision stands and is not debatable, but
the flavor of it is up to you guys. The current proposition is
to enable kernel auditing, such that only root can view the logs.
If you have a more palatable idea, you're welcome to submit your
opinions to root@csua. Of course, 'ideas' are not nearly as
useful as 'implementations', if you propose something non-trivial.
\_ I don't have any complaint on any of this. I just like to
ask if you guys can consider making the list of people who
have root public, and tighten access control to only those
who should have root. Secondly, I'd like to ask if you guys
can make all user complaints and requests to expose offenders
public. I'd hate to see root exercising power under the hood
without any form of auditing. Without public auditing
there is no check and no balance.
\_ Why perpetuate the scam and make us lend the logging an air of
respectability?
\_ I am amused by the fact that this was posted anonymously. -gm
\_ Exact proposition: "To allow, when necessary, root-types to
identify exactly who posted any message in the MOTD"
\_ If I read these correctly, the change that will be implemented is
a foolproof way for root-types to know who is posting to the motd,
so that people who make direct threats can be found. Somehow I
doubt this is gonna raise the quality of the discourse around
here.
\_ The problem is we've apparently seen root-type people abuse their
root in the past to un-anon people on the motd they simply don't
like. I'd like to know who the root-type people are and that
there is some official (as official as the csua can get) process
in place to a) make sure no one else has root and b) make sure
the very limited set of people with root are known and c) revoke
root privs of abusers. I was once in favor of a totally anon
motd, but given some of the vicious and excessive personal
attacks, threats, and named posts clearly intended to destroy
other people, I've changed my mind on the topic. Free speech
is a good thing but yelling fire in a theatre is not ok nor is
abusing anonymity to harass or ruin others.
\_ The root list has been getting cleaned up, and I have made
sure that the only people with root on any of the new
machines are active, trustworthy root types. Furthermore,
abuse of root power by anyone to un-anonymify someone for
any reason other than official business is an immediately
squishable offense in my book. If I caught someone using
root logs to spite someone on the motd, I would not hesitate
to not only revoke the root cookie, but also sorry that
person's account. I would even take such action on a
current member of Politburo if they did such. I consider
the privacy of the people on this server, and the
professionalness of those who have access to priviledged
information on this server very important. - jvarga
\_ You are a thin-skinned idiot.
\_ Haven't been around here that long, huh?
\_ Only about 8 years. What'd I miss?
\_ Pfft. n00b! -meyers
\_ You missed the part where not abusing root is a good
thing, and an obvious thing. Where have you been?
\_ Vicious and excessive personal attacks? Perhaps, but the
motd is not for the faint of heart. This is less "fire" in a
theater and more theater of the absurd. More Sproul Plaza than
debate club. Keep it anon. How else am I supposed to make my
snide "yermom" comments without looking like a total sleeze?
\_ yer mom doesn't mind looking like a total sleaze.
\_ You're correct that too many people have root. We'll get
an automatic reset when we switch to new soda, we should
set up some new rules then.
\_ So let's say some user provides a web- or e-mail based front-end to
let anonymous types modify motd. The soda log will show that the
creator of the interface is making changes, even though it could be
Joe Loser off the Internet. I suppose at the first abuse then that
interface should be shut off?
\_ Before the first abuse; it's against policy to share your
account. -tom
\_ Has this specific example been tested yet? ("share your
account" encompasses providing a web/e-mail interface for
people outside soda to anonymously modify motd)
\_ "share your account" means whatever they want it to mean.
\_ This would also qualify under "don't be a hoser." -tom
\_ Just curious, but how many of you outraged motders are actual
csua voters?
\_ I'm disappointed that the CSUA would run Linux, I'm not sure what
the issue was with BSD. There was a big push to get it working
at the end of last year, and as far as I know it was. What
happened? --jwm
\_ How competent is the vp? This is not intended to be a put
down as such, but failing to get bsd to boot may be
\_ How competent is the vp? Failing to get bsd to boot may be
meaningful or meaningless, depending on vp cluefulness.
\_ As freebsd developers have trouble getting 5.4 to run on
certain amd boxes, I wouldn't use this as a guide to VP
cluefulness
\_ You do know that my question regarding vp cluefulness
still applies until you show (boot_bsd(clueless admin)
== 1) for all values of clueless admin.
\_ What's wrong with Linux these days? (Aside from TRADITION!)
\_ If you have to ask, you don't know.
\_ Yes, that would be why I'm asking.
\_ Install the 2.6 kernel and see how long it lasts
under load.
\- can you elaborate on this a little. i have some
crunching farms and the people who run them for
me appear to slowly be moving toward 2.6. tnx. |
| 2005/10/13 [Computer/SW/Unix, Computer/SW/Security, Politics/Domestic/President/Bush] UID:40060 Activity:nil |
10/12 root, please do not squish me for posting this treasonous
url anonymously. also the picture is wrong, p bush
was funding them until 1951.
http://www.indybay.org/uploads/p1090147a.jpg - danh
\_ It's been nice knowing you danh, I shall miss you after your
mysterious disappearance.
\_ Huh, I didn't realize we were into punishing the sons for the
sins of the fathers. |
| 2005/10/9-11 [Politics/Foreign, Politics/Foreign/Europe, Computer/SW/Unix] UID:40031 Activity:very high |
10/9 got a speeding ticket.. a while back someone posted a
website for a school on east bay that is easy or online..
what was it? thanks
\_ http://traffic101.com is easy and online --dbushong
\_ they're morons. they ask for a state first.. then
ask for a country after you pick the state....?
\_ ...they were relatively inexpensive, you don't have to go
in to a physical testing center, and the test was pretty easy.
I wouldn't discount them solely on suboptimal interface design.
--dbushong
\_ There's nothing wrong with the interface design. It's
"county" not "country" which makes sense that they ask
you after the state.
\_ why do people go to traffic school? the points still appear on your
record and raise your insurance rates. it only matters if you got
so many you might lose your license, right?
\_ wrong, the points don't appear. the catch is that you can't
attend traffic school more than twice in an 18-month period.
\_ even then you can still appear before a judge and cry and
get multiple chances to keep going to TS to avoid more points
or having your license revoked. there was someone in my TS
session a few year ago who said she was one point from losing
hers (8 i think to lose it) and had been to traffic school
*several* times in the last year alone. fear. stay off the
streets if you want to survive with people like that around.
\_ Getting points is different than driving dangerously. I
have seen people driving below the speed limit who were
driving more dangerously than people who exceeded the
limit. Case in point, German drivers tend to drive much
faster than Americans but suffer fewer fatalities.
\_ uhm, yeah.... Points are for driving dangerously.
Where do I mention anything about the woman's speed?
I said she had a lot of points. Points are not just
for going over the speed limit. *boggle*
\_ hi mudder!
\_ Police in the US have a lot of leeway to give
tickets based on subjective criteria, which is why
you have recourse mechanisms such as going to
court. -John
\_ dura lex, sed lex
\_ OK wiseass, so where do you draw the line
between the law and its application? I.e.
do you believe in a 100% literal
application of the law (cop gives you a ticket
even though there's a burglary down the
street, because hey, you're breaking the law)?
Hey, sed lex, right? The law must always be
obeyed, citizen. Always! -John |
| 2005/9/22-23 [Computer/SW/Unix, Computer/SW/OS/OsX] UID:39826 Activity:nil |
9/22 Help. The NFS on my Mac OSX is sick and I can't find /etc/init.d
or any /etc/rc.X to do nfs restart. I can't even do init 2 or
other things. How do I do this on the Mac? Thanks.
<what's up with the censorship?>
\_ Use kill, the way God and Bill intended. You might be able
to (re)start NFS using the following script:
/System/Library/StartupItems/NFS/NFS
<what's up with the censorship?> |
| 2005/9/22-23 [Computer/SW/Unix] UID:39811 Activity:moderate |
9/22 What's the safest way of restarting a sun machine when someone has
the screen locked and you can't find that person? At least on these
suns, a stop-A; sync sometimes results in the machine asking for
an fsck upon reboot. Also, I realize that a different screen locker
with a logout option or maybe a script that kills the current X
session would be more ideal, but the administrators here are lazy
or something.
\_ There is no safe way, without root. On my clusters, rebooting
the machine will get your account turned off. -tom
\_ Does anybody else laugh at tom upon reading things like this?
\_ no kchang. go stick your head in a pig.
\_ As long as there's no serious disk activity going on,
there should be. Even if there is, there should still be, if
your file system doesn't suck. I'm not sure how much solaris's
UFS implementation sucks.
\_ Rebooting multi-user machines has implications other than
whether the filesystem needs to be fsck'ed or not. -tom
\_ True, but these users know they shouldn't stay logged
in anyway. These are essentially single-user machines
that interface to lab hardware. All I care about is that
the machine doesn't ask for the root password to fsck
upon reboot.
\_ just accidentally trip over the power cord
\_ Are you, by any chance, thinking of the Soda 2nd floor labs?
There I think Ctrl-Shift-Blackspace will usually log them out.
There I think Ctrl-Alt-Blackspace will log them out. -jrleek
\_ Nope. If only.
\_ On the SunRays, there's some key combination (Ctrl-Meta-Moon?)
that will reboot the terminal. This is also occasionally useful
for getting rid of the dreaded newt cursor. -gm
\_ On my Sparc Ultra 5 at work, the power botton at the front
gracefully shuts down the machine. |
| 2005/9/19-20 [Computer/SW/OS, Computer/SW/Unix] UID:39750 Activity:nil |
9/19 So I have this proprietary logger that writes into a temporary
directory, and to rotate it, we simply rename it, gzip it, and
the logger is smart enough to create a new log file. However,
I've noticed that even after we rename it, the logger still
APPENDS to the log file, and during the time we gzip and delete
it, the stuff that is appended, is lost. So I'm wondering if
there is a command in UNIX that'll CHANGE the inode so that
I can tell the logger to create a new file as soon as I rename
it? ok thanx.
\_ That is because the logger already has the file descriptor open.
Renaming it doesn't change that. Typically you would rename file,
send logger the -HUP signal which should cause it to reopen the log
file, creating a new one if it doesn't exist.. Then you can gzip the
old one.
\_ That is because the logger already has the file descriptor
open. Renaming it doesn't change that. Typically you would
rename file, send logger the -HUP signal which should cause
it to reopen the log file, creating a new one if it doesn't
exist.. Then you can gzip the old one.
\_ And if your logger doesn't handle -HUP, you will have to kill
and restart it. |
| 2005/9/19-21 [Computer/SW/Unix] UID:39746 Activity:nil |
9/18 I just installed Red Hat Enterprise Linux 2.1 and everything
seems to work ok except my httpd cannot be reached outside
of the box. I have no problem pinging and sshing into the
system, but http connection would not work. However, it does
seem to work when I launch firefox within (same machine).
What could cause this? I'm not familiar with PAM and I'm
wondering if that's the culprit, and how I can check/fix it. Thanks.
\_ PAM is a system for authentication and has nothing to do with httpd
(barring certain login mechanisms, but that's another story). Your
problem sounds like a firewall issue. You probably need to poke a
hole in RedHat's default firewall to allow inbound connections to
whatever port httpd is running on (probably 80 if it's started on
boot or 8080 if you start it manually, but it's configurable). -gm
\_ I'm going to guess you're a fucking idiot, but that's just a
guess.
\_ U = teh ghei. Check firewall rules, and whether the listening
addresses in httpd.conf are correct. -John
\_ Yes I'm an idiot, I've never done this before. So where do I
configure firewalls? What files?
\_ Ignore dim, he's just some out-to-lunch math guy at JPL.
\_ I apologize for my harsh language. I had a bad day
yesterday. I think RH 2.1 still uses ipchains, but it might
use iptables. Look to see which one you are running, if any. You
can do this with '/sbin/chkconfig --list'. Post your findings.
\_ Ok I looked up and it's due to ipchains. I tried to correct
it by "ipchains -A input -s 0.0.0.0/0 http -p TCP -j ACCEPT"
and other crap and after about 35 minutes fiddling and
reading horribly written man-pages without a godamn simple
example, I just turned it off. /etc/init.d/ipchains stop
did wonders. I'm just learning so I don't really care about
firewalls. People need to learn to WTFM (write the fucking
manpage) better. And btw you're a mean, mean guy, dim. You
almost made me cry. -op
\_ There are some good examples and tutorials online. Would
you like links to some? -John
\_ yeah yeah, please! Thank you John! -op, newbie
\_ Google says:
http://www.linux.org/lessons/beginner
http://www.linux.org/lessons/beginner
http://www.tutorialized.com/tutorials/Linux/Getting-Started/1
http://www.serverwatch.com/tutorials/article.php/3496046 |
| 2005/9/15 [Computer/SW/Apps, Computer/Theory, Computer/SW/Unix] UID:39696 Activity:nil |
9/15 I have a stupid Excel question. It's easy to obtain the product of
any two cells, but what is the syntax to specify that I want the
product of, say, the previous two cells? That is, if I am in row
A I want the product of A1 and A2 entered into A3, in row B I
want the product of B1 and B2 entered into B3, and so on - sort of
"relative to my position I want to do this". The rows can change
lettering as things are inserted or deleted, but the idea is that
I always want, say, column 1 and 2 multiplied into column 3 of the
same row. How do I do this? I've RTFM. |
| 2005/9/6-8 [Computer/SW/Security, Computer/SW/Unix] UID:39525 Activity:nil |
9/7 Is there a way to change passwords on windows from the cmd line?
I only have a telnet session... Thx.
\_ Your google fu is weak.
http://support.microsoft.com/?kbid=149427 -John |
| 2005/9/6-7 [Computer/SW/Unix] UID:39515 Activity:nil |
9/7 When I used wget -k -r to get stuff recursively, it doesn't translate
URLS like "http://hello.com/Application?key=val" I think it is
confused by the ? and other stuff. Also I need to convert them to
HTML. Do you know how to make wget translate properly? ok thx. |
| 2005/8/31 [Computer/SW/Unix] UID:39387 Activity:nil |
8/31 Do any legitimate user-agents make HTTP requests without
an HTTP Host: header these days?
\_ Not since Netscape 1.1. -tom
\_ What about search engine spiders?
\_ They don't if they want to index correctly.
\_ Thanks. When GoogleBot makes requests with HTTP/1.0
does it always send a Host header?
\_ What about search engines spiders? |
| 2005/8/23-24 [Computer/SW/Security, Computer/SW/Unix] UID:39241 Activity:nil |
8/23 Looking for a good backup software for Windows. Preferably free,
or something cheap with encryption. I'm sick and tired of manually
using MS's backup software to create a tar-like file and then
using my pirated Nero to burn it on the DVD. ok thx.
\_ Check out the backup reviews first.
http://www.backup-software-reviews.com
I downloaded a copy of Genie Backup Manager, trial version.
It is very good. I got it from Bittorrent with serial keys
but I liked it so much that I decided to buy it from them.
\_ I use Acronis True Image to back up my Windoze disk to an
external hard drive. It's fast: 1 gigabyte / minute over
FireWire or an efficient Hi-Speed USB 2.0 interface.
Image is password-protected (though not encrypted, but I think
the password protection is good enough). Buy the download version
off http://newegg.com. |
| 2005/8/19-22 [Computer/SW/Unix] UID:39195 Activity:nil |
8/19 I have files full of  s that I NEED to squash, nothing I've tried
works though. (shy of going in with an editor and deleting each one).
They all exist at the beginning of a line and are followed by another
strange charcter "»" if that helps. OS = Linux.
\_ sed is your friend.
\_ tr -d
\_ PERL baybee PERL! It'll even copy the old file before modifying,
from the command line!
\_ awk me awk me awk awk awk
\_ Any of many substitution processes
\_ Have you tried emacs?
\_ Do you know the hex/octal value of the strange character? od or
hexdump might be of help to you. Then use tr -d to get rid of it.
\_ perl -pi -e 's/^\xc2(?=\xbb)//' file1 file2 file3 ...
--dbushong |
| 2005/8/8 [Computer/Networking, Computer/SW/Unix] UID:39038 Activity:moderate |
8/8 any equivalent to arp on soda?
what about tracert (dos)? thanks
\_ Hrm, maybe, /usr/sbin/arp?
and traceroute or mtr
\_ sadly, typing arp didnt work, should we put
/usr/sbin/ path in .cshrc?
\_ No. Do some research on why things in /usr/sbin aren't
in /usr/bin. People are free to add it on their own.
\_ any urls for the clueless? google didn't turn
up anything for me :(
\_ Googling for "introduction to unix" without quotes
turned up approx 7,820,000 results for me. Or you
can look for the Nemeth book. -John
\_ i know how to use unix fairly well. i
just want to know why things in sbin
aren't in bin.
\_ funny thing is that arp/traceroute used to
be in /usr/etc/
\_ On freebsd?
\_ no. on SODA about 2 years ago
\_ Seriously doubt it. Maybe on the sequent
but that's like 7 years ago now.
\_ tom holub, you have my permission to flame this newbie. |
| 2005/8/5-6 [Computer/SW/Unix] UID:39019 Activity:moderate |
8/5 Whis is UNIX time stamp messed up on NFS?
echo hello > hello; ls -l hello; date
-rw------- 1 bozo csua 6 Aug 5 17:31 hello
Fri Aug 5 17:26:08 PDT 2005
\_ Whis is? What are you trying to say????
\_ take a Whis..?
\_ clearly because the NFS server has the wrong time.
\_ Fixed NTPD on keg. - jvarga
\_ Instead of working at internships, getting paid lots of
$$$, and learning important corporate skills like
attending long endless meetings, politicking, and
backstabbing... skills that you REALLY need when you
join us bitter engineering alumni in Silicon Valley,
WHY are you spending time doing CSUA stuff with absolutely
nothing in return? WHY?
\_ Because I like pain. A lot. - jvarga
\_ Ranga is thanking him all night long!
\_ Because if he ever says he needs a job, I'll certainly look
around to help him out as best I can, because I appreciate
people who're dedicated enough for this kind of shit. -John
\_ Ob man touch |
| 12/25 |