« first day (1427 days earlier)      last day (3840 days later) » 

16:15
@MichaelHampton A quick Puppet question for you.
@ewwhite what do you use to edit your puppet stuff ?
vim with some syntax additions and a default tab of 2 spaces.
I haven't gotten into it. Since everything is really on a remote puppet master.
I guess it's installed on my Mac!
@ewwhite use sshfs to mount it locally :)
16:21
@ewwhite I was using it before puppetlabs took it onboard
@DennisKaarsemaker good vacation ?
@Iain yup.
no biking though
:) :(
@DennisKaarsemaker I'm debating how to distribute your ILO utility.
@ewwhite want did you want at 3am (BST) btw ?
Should I git clone on each target server?
16:22
@ewwhite use fpm to make an rpm #
@Iain a silly friend of mine said her front wheel was out of true...
and that she was going to "go to Byron's house so he can teach me how to true wheels"
that reminds me I need to check mine
and I think some of my ServerFault skepticism is creeping into other aspects of my life... but wouldn't you go to a pro?
I mean, I'm a trained bike mechanic...
I don't think truing wheels is a user serviceable thing these days... at least without some real experience
It depends on the wheels - my winter wheels which have loads of spokes I do myself, the DA wheels I go to a pro
16:25
@DennisKaarsemaker I had no clue... damnit
@Iain these were some factory machine-built wheels I bought for her as a gift.
so I kinda wanted to say, "I'm not sure if Byron is really qualified... "
Is Byron qualified ?
"we all take our [swim team] bikes to him... he's probably built 30 bikes!"
@Iain do you have a truing stand?
a repair stand?
Just got a call for an IT Admin position at....a manufacturing company
That's a surprise.
@cole embrace it!
go go go
16:29
/googles company
"Do you want to reboot now, or later?" >Later "Rebooting now, fuck you"
everything is better than $currentJob, isn't it?
way smaller than my current place.
@DennisKaarsemaker So should I subscribe to your repo?
@MichelZ eh not really
16:30
I would guess so from the level of complaints :D
@cole what do they make?
@tombull89 shaft collars and couplings
@ewwhite that'd be the easiest way to get updates.
and not the fun kind of shaft collars.
Dan
Dan
I've really got to stop assuming that my servers are as I fucking LEFT THEM 2 DAYS AGO. RARGH
16:32
I dont remember applying to this
@Dan ghosts?
@DennisKaarsemaker Okay. I'll put it in Puppet.
Dan
Dan
@MichelZ Other people
@Dan that's the same
Dan
Dan
@MichelZ Truth. To be fair, they're a good engineer - just they made a change and step and I've been chasing my tail because I knew it was working last time I saw it!
16:34
yeah
fucking coworkers
I applied for it like....a month ago
Dan
Dan
Right, much work has been done today and so for the first time in weeks I'm going to have my dinner, with my wife, at 6pm!
I'll just call back and see what she has to say
fuck the wife!
i mean... literally
:D
16:36
wait this isn't Google
do we need a google bot?
it's also not youporn
just in case you're wondering
@ewwhite no I borrow one from a friend
@Iain Oh, that makes sense, too.
Haha - they asked my current base salary and then said "thank you so much for your time."
16:43
@cole :) are you already getting millions?
not millions but decent
They were probably in the $50K-$65K range for that position, I'm assuming.
Check Point® Software Technologies Ltd. (Nasdaq: CHKP), the worldwide leader in securing the Internet, today released its findings of security concerns in CPE WAN Management Protocol (CWMP/TR-069) deployments, used by major Internet Service Providers (ISPs) globally to control business and consumer home internet equipment such as Wi-Fi routers, VoIP phones, amongst other devices.
uuuuuh
very nice
Manufacturing IT doesn't pay well.
Which why even though that's my background, I don't think it'll help me to stay working in IT for manufacturing companies.
16:46
well, if they pay a decent salary - why not
but yeah, in that price range... naaaaah
@MichelZ well - most manufacturing companies don't put any money into their IT infrastructure - so I'll be stuck working with/supporting old tech (NT4, 2000, 2003, etc)
@ewwhite got this in 2 weeks - nasty 3 mile climb at 83 miles but the 11 miles downhill afer that will be cool - 12k ft overall
@cole I don't think anyone but you has anything lower than win2k3 :D
@MichelZ most manufacturing companies do.
hopefully without a firewall on the internet
16:49
and I've work for enterprise level to SMB manufacturing companies.
it's alright, it isn't forever. (I hope)
@Iain damn you and your climbs... Illinois is flat.
:(
At least I'm on a plane... next to some Puppet dude
to a place with MOUNTAINS!
but no bike :(
I have cleat covers
@Iain walk it.
look or shimano?
16:54
shimano
@ewwhite What, are you @wesley now?
@voretaq7 I think @wesley is progressing much faster than me.
@ewwhite come to switzerland.. you would love it :D
Been there.
I didn't love it :)
g
lots of mountains to bike
fuck markdown
can we vote to disable it in chat?
16:59
@voretaq7 have you been absent ?
@ewwhite Oh nice, where you headed?
@ShaneMadden Portland and Seattle.
@ewwhite Nice. Work or vacation?
both
produce company
then wife is flying out... and seeing a couple of friends and some family
@ewwhite Sounds like a good plan :)
17:04
@Iain largely, yes. I've been sick.
@voretaq7 are you better / on the mend ?
0
Q: Display the number of I/O to disk (reads/writes) since the last boot on CentOS

Dr. Gianluigi Zane ZanettiniIs it possible to use iostat (or some other tool) to get the number of reads/writes operation "since the last server reboot"? I mean: I'd need to know how many reads/writes the server has done since the last boot, not in realtime. Thanks!

@Iain mostly but the meds im on are causing trouble sleeping which sucks 'cuz I'm tired and irritable
bring me a spammer so I can shove my foot through their chest and delete their account!
go nuke some crap to me yourself happier
<bombs the middle east into an atomic wasteland>
"@Iain said I could!"
17:14
@MichaelHampton @ShaneMadden I need some puppet help
@ewwhite Fire away
@ShaneMadden The things I'm working on for this flight are: 1) adding some default sshd parameters to all of my servers 2) finally implementing ctrlaltdelete disable through the Puppet Dashboard and 3) pushing some of @DennisKaarsemaker scripts to the HP servers I have.
and I'm not sure if it makes sense to replace the entire sshd_config file or just add the 2-3 lines I need
this is across 3 major versions of RHEL/CentOS, and there are some sites that already have some sshd_config cusomization
@ewwhite Well, it's either dump the whole file in or augeas it.
I've been using ini_setting a lot more, actually.
@voretaq7 What's been wrong dear one?
17:25
@ewwhite Yeah, I've got our puppet.confs completely managed with it these days. But no help there for the sshd config
oh, you don't think it's capable?
@ewwhite Nope, not ini format
hmmph so probably push the entire file out or learn augeas?
Push the entire file out
I don't use augeas at all. Files get fully managed.
augeas is a workaround for lazy people who don't want to properly manage their files.
I've used augeas in one or two places, but for the most part I just do the whole file. There's not much reason not to.
17:31
why did you use augeas on those two places?
@ScottPack Not sure
I use templates often. That falls under pushing the entire file out, right?
@DennisKaarsemaker Adding a line to inittab to start a console on the serial port, then add that console to the securetty file
@ewwhite Yes
@ScottPack those fall under laziness. Those files can easily be puppetized :)
@DennisKaarsemaker Meh. I didn't want to risk fucking anything else up. It seemed legit. :)
@voretaq7 So just blindly trying things to try to fix it?
17:43
14
A: How to list installed packages from a given repo using yum

James AntillOn newer versions of yum, this information is stored in the "yumdb" when the package is installed. This is the only 100% accurate way to get the information, and you can use: yumdb search from_repo repoid (or repoquery and grep -- don't grep yum output). However the command "find-repos-of-inst...

SO how about Puppet for the ctrlaltdelete disablement?
Maybe I can get an internship at Booking.com
@ewwhite Same deal, manage all or part of inittab?
17:58
@ScottPack isn't that what all of medicine is?
@ewwhite would also require a move.
18:28
@DennisKaarsemaker I'm flying to Portland, then Seattle... isn't that good enough?
LinkedIn wants me to connect to Elvira.
Only if you stay there.
god news
Just secured a PT contract job.
that'll help :)
@DennisKaarsemaker :(
@DennisKaarsemaker Why can't I have it all!??!
@Pants there's cars now!?
18:34
@cole that video man.
the RP is 50/10
that guy is a pro
@ewwhite must be the black thing
@DennisKaarsemaker :(
@Pants "just tryin' to get her salt!"
Oh, this place still exists..
2
@Pants "right in her poopshute, right in her god damn poopshute"
18:40
@cole Wait until he gets in a car accident.
@MatthewIfe You've missed so much
I bet.
I've been ass-deep into NDMP for the last 3 weeks.
@Pants "SEVENTY SIX?!"
@cole hahahahahah
@Pants dying
18:42
@MatthewIfe you mean, balls-deep?
No, ass deep. I'm taking NDMP like the little bitch I deserve to be.
2
@Pants this guy's voice
Its my race track.
@Pants ohhh my back..my neck
Laughing so hard.
Sometimes I want to play this game.
That was great
18:54
@ewwhite thats a good question. Actually has lots of useful information.
Theres some cool correlation you can make from that question.
If you take his block address, multiply by four and add one (this gives you a LBA address)
Then convert that address to hex you get 1F32DE89. And look, 1f32decd is an address that smart produces as a block that might be damaged in some way. Only 34KB from that location.
I assume a whole load of blocks around that area are damanged, and the disk has ran out of sectors to reassign.
Basically replace the disk. Although -- you cant tell where that LBA belongs seeing as that value is an abstraction of a RAID array.
@voretaq7 Just remember that the more you dilute it the more effective it is.
@ewwhite Not shoulder deep?
19:18
0
A: input/output error at clean ext3 partition - how to check what wrong with data block

Matthew IfeSo, to figure this out I did the following. Take your block number, multiply by four and add one (130856866 * 4) + 1 = 523427465 This represents the sector reported as producing an I/O error. The block size being 2k, sectors being 512 bytes. The additional one extra accounts for the starting ...

19:29
@MatthewIfe See, I work with this equipment often. I've never had real disk issues that weren't visible to the RAID controller. That's why this is interesting.
I've had it before.
Bit of a shame the RAID card isn't willing to evict the disk.
Could be a faulty controller though.
These controllers are rarely faulty. I've replaced two in 5000 or so.
but I bet the controller is bitching...
20:04
Seen on AAL115 to Portland - "Nerd Life" #nerd #sysadmin #puppetlabs http://t.co/LzVxLZE4rj
2
I should add that the BOFH sitting next to me in first class not only had "Nerd Life" and puppet stickers on his laptop, but sat in an EMC Unisphere window for a good hour and kept slamming Rum and Coke drinks.
@ewwhite well that's nice
how is portland?
Green and granola.
Coming to Seattle tomorrow.
@ewwhite Seattle is awesome
be sure to hit all the tourist traps in PDX (by which I mean Powell's)
@ewwhite Rofl that's new relic, the NERD LIFE sticker
20:14
welp
Well, this guy watched me write puppet manifests, check my new relic, answer a few SF questions, do some HP magic and everything... Jigga didn't even say anything or acknowledge my sysadminness
Officially a Monitoring Engineer for <company redacted> part time :)
he was intimidated
He was getting crunk.
you should have got crunk with him
whatever crunk is
20:15
"oh yeah, moar steak, please" NOM NOM NOM.
It was like a horrible Oatmeal character come to life..
perhaps he's undernourished?
@cole Oh shit, congrats!
@ShaneMadden thanks, it's just part time/after hours.
@cole was that the one you didn't want to interview?
@RyJones no, didn't even have to interview for this.
For a company that has a dragon that speaks naturally.
20:22
oh well, very nice
so can't hurt
plus I can do all the work remotely
20:40
nice habitrails for Amazonians
20:55
@cole can't ping it? reboot it.
I just found out about Hetzner's Server Auction. Shit's cheap.
Course not sure what I'd do with an i7, 32GB RAM and 2x 3TB disks, but I'd find something.
21:36
I should just say... I hate iSCSI.
21:56
Finally was approved for a card that i can balance transfer to for free...and 0% interest. Bye-bye being gouged for $100+/month in fees...
I think I may be done with serverfault - I can't even be arsed to cast close votes any more
I just opened a box of software I got in 2005. It came with a bunch of gift cards! I wonder if they've expired?
OMG... I'm at a customer who can't find me an IP address for my new server
22:05
@ewwhite in Portland? I thought that was a vacation not a workcation
I needed to come here and rack a server
I budgeted 2 hours to unpack the HP server, rack it, download CentOS kickstart + puppet.
I assume it's in the rack, so your job is done!
and I've spent the past 50 minutes trying to get a fucking DHCP address
this is BAD
I've spent the last 50 minutes drinking wine and eating chips
@Iain gross dude
22:09
why ?
22:20
interesting commentary: news.ycombinator.com/item?id=8153798
OK fuck markdown
@JourneymanGeek ...no. it's for drinking sake out of.
@ewwhite we having a SE/SF lunch when you're here?
since Jacob will be here Monday as well
22:48
Just got a bug report in through Microsoft Premier Support... we'll see if anything comes out of it
Has to do with Remote Desktop services trying to use archived certificates, causing TLS negotiation to fail
back up and watch it change!
23:06
@MichaelHampton I'm getting weird issues with doing something like this in systemd:
INSTANCE_ID=$(/bin/curl -s http://169.254.169.254/latest/meta-data/instance-id)
that doesn't store anything in INSTANCE_ID
@JoelESalas Is the network up?
@MichaelHampton I.. don't know. here's my service file:
@MichaelHampton pastebin.com/XgmRWfC7
I assume INSTANCE_ID Is declared in the script?
Because that wont work.
2
A: How to set systemd service dependencies?

Michael HamptonYou need, at minimum, After=network.target in the [Unit] section of your unit file, to ensure that the network is up before starting nginx. I have no idea why your unit file doesn't have it. Here is a complete example from my handy Fedora system, as shipped by Fedora: [Unit] Description=The ngi...

@MatthewIfe yes let me provide the script...
23:10
Well, wont work even with the whole script shown
You don't even need to click through. Everything you need is in the one box.
Environment variables are inherited to the calling process and all its children.
you cant acquire a environment variable and assign it to the parent (such is systemd in this case)
Perhaps you can return a value however.
@MichaelHampton @MatthewIfe thanks to both of you!
23:42
weee
@cole Poop on anyone's desk today?
Nope
@cole Go poop on someone's desk.
@Wesley but I'm home..
@cole That poses challenges.
I'm not aware of the feasibility of long range pooping, but something tells me the blood pressure involved could pose some risks.
Long Range Poop as a RESTful Service
23:58
On the contrary, there is no rest when poop is involved.

« first day (1427 days earlier)      last day (3840 days later) »