« first day (1070 days earlier)      last day (4197 days later) » 

06:00
Porn is just tubes.
they're saying, in more words with more unnecessary squiggly lines, that the number of tries it takes to crack some ciphertext goes down if you predicate the probabilities involved on what the data might likely be, and the more information you have on that, the more optimal your bruteforce.
that's not a spectacular statement, it's obvious.
and it's also how we've cracked ciphers since like the second world war.
@FalconMomot you'd think "1+1=2" would be too, but have you seen how long the proof is??
@Andrew and this is why I didn't major in math.
anyway!
I want to show that paper to this (mildly autistic) insufferable self-proclaimed security expert, for the trololol.
it's a slight extension of the idea that the Shannon entropy per unit varies negatively as the number of constraints on the data.
What is this... I don't even...
0
Q: Assign global static ip address for my remote server

user185296I want to set a global static ip address for my remote server. My remote server is attached to a router which has its own static ip address. How do I assign a global static ip address for my remote server so that everytime I want to access my server I would just type the ip address of the server ...

06:06
Ok, a question. Does people asking stupid questions on ServerFault mean that anything is actually wrong? It's a public site on the internets, there is always going go be some noise.
The whole "ServerFault is Dying" panic seems a bit... overdone.
@Andrew my sentiment exactly
I see a lot of content. A lot of it is worth the trouble to answer.
Build a bridge and get over it.
Get a deck of cards and deal with it.
Eat some cement and harden up.
etc.
most of the people caterwauling about how the site is dying seem to be people from other sites who are angry that their dumb question got closed, or awesome people who are just bored with the site.
4
correct me if I'm wrong.
I would like an Official Determination from Stack Exchange Inc. on "scope", kthx
Well fuckers, I'm going on holiday for a week so I won't see you all next week. I'm sure I'll find all the usual in here when I get back :)
06:13
@MarkHenderson Enjoy!
@MarkHenderson Bye. =(
@FalconMomot and the people who don't realise haven't been here long enough but the time between entry and exit id shortening all the time
I think a gigantic number of drive-by users and people with very light participation is to be expected for a site like this
06:29
@FalconMomot I wasn't talking about them
@FalconMomot: driveby users don't complain
;p
They totally do!
@JourneymanGeek good point, well made
they come by, ask an inane question, complain bitterly in the comments that it gets closed, and run off.
06:31
Actually it strikes me that we close a lot of stuff on SU as well.
but we get less complaints.
Naturally, wider scope, less targetted audience, but there's obviously something in the intersite dynamic between SF and other places
hm
-2
A: Outlook 2007 Server Side rules only apply to new emails, not old emails. How do I change this?

spectranettechnologiesMaking rule for mails is a typical task as when its a important mails.Just go to tools options rules.There u can create your rules as per your requirements If you having issues to do it i can take a remote of your system n help you in that .You can add me on facebook spectranettechnologieshelp...

I think at least a few of these complaints are coming from desperate people who can't figure out how to do their job, maybe.
I think this is the equivilent of the 'free candy truck'
@FalconMomot: well, kinda
I'd guess this would be the 'slightly out of their depth dev-admins'
silly thing is, they could ask this sorta thing on SU or (eww) reddit's sysadmin subreddit
I'm cool with migrating some of that stuff to SU, but there is a lot of stuff that gets closed here that is just plain bad anywhere.
even on a normal forum it would probably get deleted
06:37
lol
yeah
and I don't care if those users complain, personally, because all they will ever do is generate more crappy content.
not everyone in the world has something to contribute here.
I'm internally debating whether to replace the shared rig
I should have the finances to at the moment
gah, money
there is never enough money
06:38
er.. wrong window
eh, we just had a shipment out to one of our customers
@FalconMomot but everyone thinks they deserve out attention and time
I think I'm going to move to Seattle for a month or two early next year, if I can afford to
so I have money. I'm trying to work out if its the right time, and if its worth it for me ;p
@FalconMomot just because you can ?
@Iain Yes, apparently so. I just can't wrap my head around the attitude that might make someone think that.
@Iain basically.
06:39
lol
@FalconMomot open door policy
@Iain: and a lack of proper intellectual curiocity
around that time I won't be working on development tasks at all, so it shouldn't be dodgy to get a TN visa
Someone can tell me "You need X tool to do Y" and I'll poke at it
@JourneymanGeek that's a different problem
06:40
some people however want detailed instructions
@Iain: well, part of the reason we close stuff is lack of obvious research, right?
I don't like people who want excessively detailed instructions. they won't help you unless you understand how the system works.
(amusingly, I think my most upvoted question here is redundant now, since upstart's documentation is a lot better than it used to be)
and yet there are people, who are paid as IT professionals, whose jobs are to do things like plug console cables into switches and paste in commands, and who aren't encouraged, able to, or even allowed to touch so much as the contents of those configs.
@FalconMomot: I was thinking "GIMMETEHCODEZ" vs trying to solve it, laying out the issues you have had, then picking brains.
lol
I like "why does this happen and what strategy can I use to prevent it"
or "what strategy can I use to implement this"
not "what commands"
06:43
yup
or at the very least
unless the question is asking for like... just one command because they're not sure if it exists
listing out how you went wrong, so the next guy won't do it ;p
but not a script for god's sake
like my Blackwidow question on SU
evil: spiders
06:46
lol
@FalconMomot: eh, I'm also currently lusting after one of the new nostromo clones razer is making ;p
but not likely, I don't see any practical use for it for now
also, blackwidows are reasonably lovely keyboards
I need to deep clean mine once I get my keypuller
they are
I have one at the office. it's wonderful.
the way the macro keys work is nice too
lol
I use one for search
not using the rest yet
I can reprogram the numpad to be qwerty number keys for use in weird terminals
and I have one that is programmed to pushd ~/project; make debug; popd
07:01
0
Q: How to convert lvm to ext3/4?

Mickey ShineI have an old box(CentOS 5) that was formatted as lvm but now I want to convert it to ext3/4 file system without losing data. I have some empty disks that can be used, so is there an easy way to do this?

this guy is confused
BTW morning, humans; Hail, Lord Darkstar.
Yesterday was a major holiday in this country, so today the office is almost empty. I can listen to grind death the whole day without interference while writing some puppeteer stuff. Lovely.
@dawud I looked at that and walked away, if you're that confused we can't really help
@dawud \m/
@Iain I'll leave a comment, that's all I can do
@FalconMomot http://www.youtube.com/watch?v=cneAYSbIS1U
that's not bad.
I'm also partial to symphonic stuff
I quite enjoy epica and after forever too :)
 
4 hours later…
11:00
hello
i was searching a chat bot which can be used from a terminal
having read @mileserickson's link on the star wall, I can't help thinking that "Include details bout what tha fuck you have tried n' exactly what tha fuck yo ass is tryin ta do." would actually be a huge improvement to our help page.
just the name and the method to connect to a free chat bot which can be used from a linux terminal
and then it begins...
Linux well-known-site-fw2-pri 2.6.18-53.1.14.el5xen #1 SMP Wed Mar 5 12:39:19 EST 2008 i686 i686 i386 GNU/Linux
# uptime
 10:38:44 up 1819 days,  8:09,  0 users,  load average: 0.02, 0.01, 0.05
# dmesg | tail
EXT3-fs error (device md1): ext3_get_inode_loc: unable to read inode block - inode=164453, block=327711
sd 0:0:0:0: rejecting I/O to offline device
Buffer I/O error on device md1, logical block 344161
EXT3-fs error (device md1): ext3_readdir: directory #163652 contains a hole at offset 0
sd 0:0:0:0: rejecting I/O to offline device
EXT3-fs error (device md1): ext3_readdir: directory #163652 contains a hole at offset 0
sd 0:0:0:0: rejecting I/O to offline device
EXT3-fs error (device md1): ext3_readdir: directory #163652 contains a hole at offset 0
11:15
Whatever you're looking at was probably configured by that guy who wanted to convert his file system from LVM to ext3/4
Fridays...
# mdadm --help
Bus error
Seemingly.
Today's the day I discover that my PFY, who's on vacation, has left our system centre virtual machine manager tool, the MS version of VCenter pointing to an update management server that doesn't exist...

And that Microsoft's way of announcing this is to crash the update part of the VMM console until you log onto the database server it keeps its config in, figure out what tables contain data about the erroring server, and delete them by hand.
wait a moment. How is it even possible to get a bus error from the help command on something?
@RobM the disks are gone.
this thing is running in RAM
ALL the disks? Oh. oh
directory #163652 contains a hole at offset 0
Yet the site still seems to work.
11:22
wtf!
(this is a firewall for said site)
That has 'long day at the office' written all over it
@ewwhite It's like the server's in free-fall about to crash in a flurry of bits.
11:44
@ewwhite Who leaves a system up for FIVE YEARS?
11:59
@MichaelHampton I don't know, man.
@ewwhite Hell, the firewall is all in kernel space anyway, so it could have been doing that for the last four years without anybody noticing.
it was until our disk read-only check went off
I know uptime is important but whoa.
the secondary firewall has been up 112 days
so um, yeah...
12:02
eh, firewalls don't really need to be rebooted.
Unless of course you like kernel updates...
I've had ASA firewalls up that long...
but there's no disk to deal with
seriously... an ASA firewall is $400-$600
It's a firewall, it doesn't NEED a bloody disk, let alone a whole RAID array!
I'm surprised it's still running without disks
I'd think an ASA firewall for $400-$600 has to be a better deal than any Linux server hardware you'd get at the same price.
and I wouldn't be worried about running ASA 5505's in an HA-pair setup.
I've seriously never had one fail
There was a 1U Supermicro Intel Atom server on Newegg's Shell Shocker yesterday for $200 something.
Guarantee it'll fail when it's least convenient.
12:06
@MichaelHampton "Oh, you're running a business-critical process on me? It would be a shame if I were to ...crash."
The super micro or the ASA?
@ewwhite The supermicro, of course.
so any idea why we didn't use Cisco?
We use Cisco for some clients.
The client wanted to do it themselves?
They had somebody five years ago that didn't like Cisco?
naw, we manage this stuff.
there are plenty of Cisco Pix 506e's out there, too
12:11
In that case, consult... the documentation.
Seems arbitrary
@MichaelHampton Documentation. Psssh.
A napkin with some scribbles on it
@ewwhite Better than nothing.
12:27
Effing clients...
Client emails at 4am...
We're getting errors starting django.
`web-01v: no module named _md5`
`cannot import name utils`
I look and see some fucked up VMware issues.
the VMs were being capped to 4GB RAM, but assigned 12GB
I clear it all up and rebalance the cluster.
Client messages back and says it still doesn't work... I ask for a way to reproduce the problem (run these commands)...
an hour later... Client writes
After troubleshooting these nodes extensively, we've lost confidence in them. Please destroy nodes web-01 and web-02, then clone web-03 to replace them. Please ensure that the IP addresses and hostnames are set correctly as well.
@ewwhite Clients, do they even think?
Did it work before? Sounds like they're just missing packages.
Yeah, totally.
as long as they pay the bills...
"ImportError: No module named _md5" on httpd restart
I'm troubleshooting now
RHEL 5.8
12:43
@ewwhite Developer error.
Is it?
@ewwhite rpm -V python turns up nothing?
nothing
Do they have pypy-libs installed?
They might have a virtualenv somewhere
12:50
no, they don't
mod_wsgi (pid=4452): Target WSGI script '/opt/www/dingo_blitz/apache/django.wsgi' cannot be loaded as Python module.
File "/opt/virtualenvs/defaultdjango/lib/python2.7/site-packages/django/utils/http.py", line 7, in <module>
@ewwhite Well, they borked their environment somehow. I don't think it's your problem.
Oh, they have a stupid virtualenv. Tell them to go ask on Stack Overflow
@MichaelHampton they're asking for the servers to be blown away and reimaged from another
morning
you can always try to run the interactive python shell and try to import md5
@ewwhite They want to destroy the evidence of their developer's stupidity?
12:53
>>> import _md5
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: No module named _md5
@MichaelHampton possibly
@ewwhite full traceback please
@DennisKaarsemaker @ewwhite It's in a virtualenv
so?
virtualenv sucks for production, but generally tends to "work"
@ewwhite import hashlib works?
My point is, nothing's wrong with the server, they just fucked up their virtualenv
12:56
@dawud seems to
@MichaelHampton so that's what needs to get fixed.
@MichaelHampton could be. _md5 isn't a stdlib module and django doesn't have one either last time I checked
@ewwhite it is md5 , not _md5. Apparently the md5 module was deprecated a while back: docs.python.org/2/library/md5.html
import md5 doesn't complain
ewwhite, post the full traceback of that importerror apache gives.
Anyone here know netapp a bit?
12:57
_md5 seems to be from PyPy
tough to sanitize
2013-08-16 12:57:21,521 ERROR : [hmx-blitz-01 apache-blitz] [Fri Aug 16 12:57:21 2013] [error] [client 10.64.120.59] ImportError: No module named _md5
2013-08-16 12:57:25,267 ERROR : [hmx-blitz-01 apache-blitz] [Fri Aug 16 12:57:25 2013] [error] [client 172.18.99.123] mod_wsgi (pid=4457): Target WSGI script '/opt/www/dingo_blitz/apache/django.wsgi' cannot be loaded as Python module.
2013-08-16 12:57:25,268 ERROR : [hmx-blitz-01 apache-blitz] [Fri Aug 16 12:57:25 2013] [error] [client 172.18.99.123] mod_wsgi (pid=4457): Exception occurred processing WSGI script '/opt/www/dingo_blitz/apache/d
oh eww, they're running production on pypy?
There's no pypy package installed. At least via the normal yum.
won't be via normal yum. virtualenv.
they have hashlib in an egg? in python2.7?
what the duck is this?
I'm with @MichaelHampton on this. That virtualenv is thoroughly devops'ed.
Couldn't I just rsync it from another known-good server?
or should I do what they're asking and blow the whole server away?
13:01
could be you $PYTHONPATH in apache is fucked up
as build/* paths are odd even for virtualenv
So sweaty.
declare -x VIRTUALENVWRAPPER_HOOK_DIR="/opt/virtualenvs"
declare -x VIRTUALENVWRAPPER_LOG_DIR="/opt/virtualenvs"
declare -x VIRTUALENVWRAPPER_PROJECT_FILENAME=".project"
declare -x VIRTUALENVWRAPPER_PYTHON="/opt/python27/bin/python2.7"
declare -x VIRTUAL_ENV="/root/.virtualenvs/defaultdjango"
declare -x WORKON_HOME="/opt/virtualenvs"
oh, that's fucked up
not just vitualenv, but also virtualenvwrapper.
I'd not touch that with a 10 foot pole without permission to fix it properly and 'educate' the responsible parties. With a sledgehammer to the fingers.
If I find this '/root/.virtualenvs/defaultdjango' on a box, I set the VM on fire
13:03
@ewwhite I'd clone the known good server. There's only so much time I would want to spend poking around in a giant pile of manure.
haha
curious who this client is?
ballXXXmer?
devoppy.
They have puppet.
13:07
so? Having tools that can be used for good does not make you good.
That puppet must be on heavy drugs
The prompt says (defaultdjango)[root@hmx-blitz-02 ~]#
Over my head...
@ewwhite just a custom PS1
13:09
@dawud yep... but it's ugly.
four different colors
oh, right... I copy/pasted
colors are easy too
also blinking =]
You there @MikeyB?
laters!
I always found that dev types were more likely to do weird shit with their prompts.
-1
Q: How can i configure different multiple vlan with dhcp server with a single ip subnetting

user185856Hello every one out there!! I got a issue i would like to share with you, expecting to have it sorted out. I want o set up vlans on my network, in order to separat same department,which can be about 7 to 10 different Vlas. lets say right now im using 192.168.1.0 network. the thing is, i dont ...

"But...I want vlans!" VLANS!...and a cloud! Buzzword!
13:18
Is it 430PM yet?
Amazon put together a list of the funniest reviews. I think it needs some new nominations...
@Cole 4:30? I think I'm checking our around 3:00 PM today. =]
Lucky.
@TheCleaner "I got a issue i would like to share with you, expecting to have it sorted out."
@ChrisS we have an internal version of that as well
including some very much NSFW reviews...
@DennisKaarsemaker Perhaps that was part of the reasoning - those are all quite tame
13:24
including a guy offering his slutwife (his words, not mine) for nasty gangbang activity...
@ChrisS I think he's locked away somewhere without windows and visitors. He's way too excited that people exist. "Hello every one out there!!"
@TheCleaner That sounds like a direct translation or something
"I purchased this product 4.47 Billion Years ago and when I opened it today, it was half empty." - witty...I like
How to Avoid Huge Ships by John W. Trimmer
"I read this book before going on vacation and I couldn't find my cruise liner in the port. Vacation ruined." -hahahahaha
@TheCleaner that one got me.
Good morning.
13:37
sup @mossy
cool, on Monday we plan to migrate our most complex site to a new datacenter but today is a bridgeday here (not sure if that is a term in English?) and nobody is here to help me figure out the last minute problems
@Basil Yep
@MikeyB Have you ever removed a lun from a mirrored plex in an aggregate?
@RobM That's just disturbing
@ChrisS then it belongs here for sure
The guy who posted this answer is a friend of mine. I'm sure he's going to be thrilled with what I did to it.
0
Q: Installing couchbase in ubuntu 12.04

Wondering CoderI havent been able to start my newly installed couchbase-server_2.0.1. I've spent more that 2 hours trying to get couchbase-server work in my local machine. By the way i'm using Ubuntu 12.04. I tried removing it using this command dpkg -r couch-server and even removing it using my Synaptic Pack...

@TomO'Connor What answ- ...oooh.
I don't get it
You have to have 10K+ rep to see what happened there.
14:06
@Cole it got nuked from orbit
He posted an answer which wasn't.
So i moved it to a comment, and left him a comment on it. (which i believe he can see).
Doh
@MichaelHampton I should answer that question
wtf acne
General question... How can I tell customers to check with me before doing stupid things?
14:14
@ewwhite "Check with me before doing stupid things."
Damn @MichaelHampton types faster than me
@ewwhite Hit them with a stick everytime they dont!
@Cole I know that feel.
Customer got water in a network switch.... decided they needed to get a new PoE switch. This is an HP shop... but they went and got a Cisco WS-C2960-24PC-S for $1300.
@ewwhite Threaten with a clue-by-four
14:14
@NathanC random honker of a zit on my jawline. I think it may be an ingrown hair
to run some wireless access points.
I'd like to go home.
Suddenly, after a power outage, anything connected to that switch can't get a DHCP address
@Cole I get ones that randomly appear in the most visible of places (side of my eye, on my nose of all places)
I ask client what the switch's IP is...
14:15
@ewwhite Gee...I wonder why.
I don't know
...
Ouch
I finally find it... and telnet right in.
ip dhcp pool 10.0.0.0
   network 10.0.0.0 255.255.255.0
   lease 0 0 10
interface Vlan1
 ip address 10.0.0.1 255.255.255.0 secondary
 ip address 10.1.1.207 255.255.255.0
 no ip route-cache
So the switch had DHCP enabled...
and the client chose a random-ass address to assign to it
...a switch with dhcp enabled? lolwut
I run DHCP on switches sometimes... BIG SWITCHES...
14:18
@NathanC Usually get them on the sides of my nose.
@Cole I get them everywhere no matter how much i try, so I stopped trying :P
:(
I had yet another recruiter call me this morning...
(while I was at work, awkward)
I think I should hide my dice profile...
I get about one a month right now. They all want me to move to NYC or LA. Ain't gonna happen.
Yeah I get them for like Texas and Tennessee. yeah...NOPE.
14:22
I told the guy I've had it since I started looking for a job previous to this one and I keep it updated just in case :p
Casually printing a 21 page cisco configuration plan..
lalalala
WHY DO YOU HATE TREES?!
@ChrisS Because they fall on stuff and break them.
Like my hopes and dreams.
@ChrisS I had to print 6 X 164 Full duplex pages yesterday.
@MichaelHampton literally say that?
14:27
For people with multi-tiered IT departments. (help desk > desktop > sys admin, etc) Where do you draw the lines for responsibility?
@ewwhite Perhaps not. I can get away with saying that to my clients...
@t1nt1n I send half a million faxes a day. Fuck trees.
@Cole Normally sysadmins handle tickets titled "HALP THE SKY IS FALLING AND ALL THE SERVERS ARE BURNING"
and not "I'm a dumbass and can't figure out my printer"
I'm trying to see how much shit I can pawn off to the help desk and desktop
@MichaelHampton maybe this?
14:31
@Cole But really, unless the desktop/helpdesk folk don't have the ability to handle the ticket they should be...
@ewwhite You want to hire me to tell your clients that they're morons?
(Permissions, etc)
Yeah should help desk be doing permissions on the file server or us?
The Cisco 2960 had a DHCP server configuration on it. This was dangerous.

ip dhcp pool 10.0.0.0
network 10.0.0.0 255.255.255.0
lease 0 0 10

When it comes to equipment purchases, it makes sense to check with me ahead of time. It's difficult to work backwards and troubleshoot things like this.
@MichaelHampton good enough?
@ewwhite It won't offend anybody, but it probably doesn't get the point across strongly enough.
14:32
@MichaelHampton then what do I need to say?
@Cole Well, do you trust they won't make broad changes as some people like doing? :P
@ewwhite Um, let the invoice do the talking?
@MichaelHampton I'm already $200/hr to this place.
"Oh, you need access to just that subfolder? here, have "full control" over the entire share!"
but why would you just go and buy a Cisco 2960 when you have HP chassis switches...
14:34
@ewwhite Then yeah, the above combined with your invoice is probably all you can do.
@NathanC so most of our shares have a group assigned to them - it's just adding the user to the group.
Ah, then they should be easily able to carry out that task.
I thought so
I think we need a list of certain shares that need authorization before handing it off though
@Cole the people owning the file should. You should build a tool that allows them to, in a safe way.
@DennisKaarsemaker That'll never fly here.
14:50
Stupid account summary screen uses 15 minute delay prices... Buy a stock and it instantly shows a $13 loss - and 15 minutes later it's broke even. >=|
For anyone who didn't notice we're two weeks out from quarter end, so stock prices are more volatile than usual... Can be a good time to get in on "underrated" stocks.
@DennisKaarsemaker I seem to remember @ewwhite and MDMarra discussing that. Are users capable of managing permissions?
Users shouldn't have to manager ACEs/ACLs... They could, however, be delegated rights to alter group memberships.
You'd have to create a workflow for it. User submits request > request goes to owner > owner clicks "yes" or "no" > No sends an email to user stating that they cannot get permission and to contact owner or "yes" adds the user to the group then send them an email stating they have access and need to log off and back on to Windows for it to apply.
That would be ideal.
0
Q: What were your server room pain points, big wins and must haves?

PhilLike many people on here I suspect, our server room has evolved over time. At the beginning we had a single server running Microsoft Small Business Server 2000 and it sat in the corner plugged in and happily did all our emailing and firewalling. Over time we have deployed a few more servers a...

14:58
Don't we have something like that already?
I think everyone will agree with "a budget". — Nathan C 9 secs ago
=p
lol
Its worse not having a budget ;p
What's a budget?
How curious...this server "was" restarting, then just stopped.
Is it that really tiny amount of money that we have allotted to us that the Applications and R&D group keep eating up on us?

« first day (1070 days earlier)      last day (4197 days later) »