« first day (1134 days earlier)      last day (4133 days later) » 

00:00
yes... but this is some shit
@ewwhite Hourly billing, right? =P
@ewwhite Definitely looks like bitch work, you should farm that out to some brown dude
I just wrote a quick kickstart and am going to let this install from my repo
the client is a Windows guy
Anonymous
god damn it
Anonymous
gotta build a page for my federation's candidate campaign
Anonymous
00:02
and it is starting in Monday. Gotta work like mad.
@superpatosainz federation? like star trek?
Anonymous
As in supra-student's council
Anonymous
a higher, universal, abstraction
@superpatosainz write a wrapper class around it
Anonymous
I think i'll just go the easy way and build a mobile-only (almost everyone who consumes our content is on mobile) web app thanks to some iOS-like style framework
00:05
aah, pre-release bugfixing
I made like 2 bugs
and the fix was trivial
so 2 days of work... like 20 lines of changes.
Anonymous
The other way is to use bootstrap, I don't want to mess with it for this
Anonymous
nice
00:30
@FalconMomot Doesn't matter, got paid. =P
@pauska Thanks for the iTap RDP recommendation from a long time ago.
It fixes my RDP->ILO->console latency issues.
00:47
@ewwhite Is it on SF? link pls
Effective October 8th 2013, iTAP mobile RDP apps for iOS, Mac OS X and Android platforms are discontinued.

    We will continue to support the customers who have installed this app up to this date but the app will no longer be available at the application stores and no further updates will be released.

    We recommend you evaluate the new Microsoft Remote Desktop app available in app stores later this month.
Sucktastic, @ewwhite
That sucks. I liked iTAP.
It was a bitch trying to fix servers on an iPhone screen though.
@RyanRies That's why I got a Droid 4. Physical keyboard.
01:03
I have been devopsing my brains out all week... I'm exhausted. I'm going above and beyond to do something that I wasn't asked to do, to develop a system that does not currently exist, but there's a need for it.
As evidenced by me answering like 1 question on SF all week.
@RyanRies Nice that you have the available time at your workplace to do that.
And you're not constantly putting out fire after fire.
I think I'm being stunted really, really badly by firefighting all the time.
But hey, it's billable time, but someday I'd like to get paid and get smarter.
Well it's also that when I get my mind really focused on and passionate about something, I do that thing after hours, in the evenings, on the weekends, etc... there are definitely downsides to that kind of behavior.
And I will drop whatever else as soon as possible to get back to my one fixation of the moment.
Sup
For instance, I will never be able to have a girlfriend who requires much attention because of that...
I can only be with people who can entertain themselves for good stretches of time. :)
@RyanRies glad I'm not the only one with that problem...
Current gf is needy, luckily we only see each other on the weekends, and one night (Saturday into Sunday usually)
Although, my crap job has kinda burnt me out on that stuff.
01:16
@RyanRies I just assumed everyone that is like this found a job as a sysadmin.
Luckily it's finally over, but who knows what the new one will do for me.
@Wesley But yes, being burdened by having to constantly fight fires is a bad thing and it does stunt your growth in learning new things and developing creative solutions.
2
I'm actually terrified to start my new job.
@cole Why?
I'm not sure, I'm really nervous for some reason.
01:19
@RyanRies I love it when that happens - when you get something caught in your brain and spend all day every day for weeks to get it to come to fruition.
I do too...
@Wesley oops...
It's the thing that reminds me that I'm in the right industry... working with computers, making them do new things that no one else has and fixing weird problems that no one else can.
@cole Aww, you'll kick ass dude. No worries.
@Magellan thanks man
01:21
@Magellan @cole I second that. You're gonna' be fine.
@cole Well, they chose you for a reason. You damn well better live up to it.
True.
@jscott =P
They chose me within 3 hours of my interview
after interviewing 30+ candidates
See, don't doubt... deliver.
01:23
Under-promise and over-deliver, that's my motto. In any event that's better than over-promising and under-delivering!
I like it. You're welcome to over-promise as well... as long as you gauge your over-deliver on the same scale.
@RyanRies Stated another way, and one of my favorite quotes of a Sr admin I worked under for a while:
Decrease user expectations, increase user satisfaction.
Hah, nice!
Blah
01:41
Yet another reason not to trust cats: They smuggle drugs into jail.
Poor cats :(
Cats are evil.
10PM on a Friday night, eating apple sauce in my underwear while talking to people on the internet. I'm living the dream here, folks.
01:57
Everything but the apple sauce part sounds about right.
02:42
NORTON The anti-virus virus.
2
Norton. Because markdown has to have something to make fun of.
02:59
lawl at datacenter saying I'm capped at 20Mb
@Wesley Perhaps they mean outbound?
@MichaelHampton I can blow past 20 outbound too. It's just not what tonight's patterns are.
That would be odd if someone capped outbound only though -- I think they just made a change in their upstream router and forgot to check to see if their limits stayed in place.
I can tell you that their core is run off of MicroTik, which I'm not sure what to think of.
One minute average
Maybe you're not actually capped, but they will become annoyed with you if your 95th percentile goes over 20Mbits?
@MichaelHampton I thought about that too. Nothing was mentioned about 95th percentile billing in the contract. Just a 100Mb port with 20Mb cap. I did inquire about 95th percentile burstable billing because of a potential to go over, and they said they could do that too... but... it's all just weird. I think they dropped the ball.
I'm half tempted to push it and just download a CDN file over and over to /dev/null and see if they holler.
Or scp to another server's /dev/null
Or both
hammer it bidirectionally
03:14
Eh, if you aren't actually planning to go crazy with the bandwidth, I wouldn't mention it at all.
Yeah, I told them about it once. Emailed their NOC address. No response.
And I'm sure that hammering some server randomly and generating crap traffic is probably a TOS violation or something. So read it carefully first. :)
This was like four months ago when I noticed
@MichaelHampton I actually did just pour over my contract to see what I should renegotiate in a few weeks, and so far nothing would seem to be explicitly detailed about TOS other than nothing illegal
Oh, duh!
I forgot to mention
In spring I was most certainly hitting a 20Mb bandwidth cap
All uploads and downloads, like for instance updating servers, or just uploading files away somewhere else, were neatly capped.
And then, one day...
I was getting much faster speeds.
Yeah, they probably screwed something up.
Holy cow, two and a half pounds of pot?
Someone misplaced their dead drop.
@JourneymanGeek Heh, it can be.
And the Cardinals won the pendant!
04:42
@Wesley hehe "An officer took the marijuana and placed it into evidence for destruction."
suuuuuurrrrrrrreeeee
in seattle no less
it could well have really been a 3lb bust :P
you know... it would be really funny if they caught the dead dropper, and he gave a sworn statement that there was in fact more dope
ok, AD FS certs have been renewed, deployed to the proxy
I'm going my ass home!
well.. gym first, then home
 
3 hours later…
07:55
good morning all
08:12
Would not recommend!!

Senior Linux Engineer - DevOps Focused

Logicworks

Position Overview: Do you enjoy engaging, important work with a wide array of technologies? Is problem solving a…

Posted on Careers 2.0 on October 18, 2013

now there's a buzzword I am tired of hearing.
08:29
morning
@FalconMomot which one?
Dev... 0pp?
it's my old job
you poor thing
I like the... This is an on-site role in our NYC office, so there is no chance for remote. However, we would offer relocation assistance for the right candidate.
08:31
very direct.
@ewwhite Should I do it
@JoelESalas Just for shits and giggles...
you know what is awesome? releasing software no more often than once every 3 months.
They used to be open to remote people... I suppose I soured them on that.
@ewwhite I also like how there's a stipulation that the "right" candidate would get relocation. As if they'd hire the wrong one?
08:34
@JoelESalas best I can tell, they had a hard time closing the deal.
I was pretty turned off with what happened with you.
@FalconMomot Must be agile!!
i've been watching Lockup Raw for hours...
Lockup is a prison documentary series, produced by 44 Blue Productions, which airs on MSNBC. The Lockup franchise also includes Lockup: Raw, Lockup: World Tour, Lockup: Extended Stay, Lockup: Special Investigation, and Life After Lockup. Lockup (and the other series in the franchise) explore prison facilities throughout the United States, profiling notable inmates, incidents, and prison operations. A typical episode usually follows one or two inmates as they attend disciplinary hearings, receive visits from family, and interact with other inmates. In some episodes, inmates are provided w...
 
1 hour later…
09:42
@ewwhite you ruined it for all of us!
09:55
@DennisKaarsemaker sorry, bro.
10:40
@ewwhite @RyanRies @Wesley the iTap code was bought by Microsoft, it's now available free on all app stores (iOS/Android/OSX) as "Microsoft Remote Desktop"
5
We need to get that sticked on the star list, way too many people using it here not aware of the new client
 
2 hours later…
12:46
flags plz
0
Q: how to ask google to repair nexus 4?

javapoweredGlass on my Nexus 4 is broken (touch doesn't work) after it falled from relatively small bed (just about 50 centimeters from the floor). I purshased Nexus 4 on google play, can they fix this by warranty or for pay (how much?) I've tried to call US support team from here but it is not possible. ...

13:31
@pauska also very cool
The feedback on our end was excellent and we would love to extend you an offer to join the team. - Red Hat
@ewwhite context?
There are a couple of offerings to work at Red Hat in my country. I'm far from ready to do so, but I'd love to.
Only one left, apparently
Oh, I interviewed for an infrastructure consultant position there.
And little did I realize that RH was advising the same healthcare company whose systems I was hosting at the last job.
14:02
@ewwhite nice. You gonna accept?
14:23
StackExchange's question filters just went down
 
1 hour later…
15:41
@Magellan ...
@ewwhite haven't made up your mind, eh?
@Magellan Just starting the search.
@ewwhite Ah.
And it's good to have options.
@Magellan where did you end up?
@ewwhite linkd.in/1a1klu7
15:56
Ah, I see...
16:09
Yeah. Looks pretty good so far. Fill in the gaps and work with some pretty good folks.
16:25
@Magellan What do you do?
@ewwhite Generalist Systems Admin. I don't have a particular specialty.
Break stuff? Fix stuff?
@ewwhite Fix stuff. Other people break it. The new place has a lot of engineering going on as they expand and they're just now building out some of the things that I want to work on. So it's a good move right now.
Puppet shop too.
@Magellan Like puppet?
@ewwhite I prefer the way Chef works, but Puppet's much more stable. I guess my primary beef with Puppet is that nodes are defined in the code instead of in the database as with Chef.
16:33
@Magellan you can use hiera or an ENC
@Magellan I use Puppet Enterprise - so Puppet's dashboard is the ENC
@ewwhite Ah. I haven't had opportunity to work with the spendy stuff.
@Magellan Wait... I supposed to pay for this?!?
@dawud True. Hiera, at least, being another bolt-on piece that last time I looked at it was even less stable than Chef.
I thought Puppet Enterprise was hosted-only.
Ah. That's right. It's free up to 10 nodes.
@Magellan Methinks that's Chef's commercial offering
16:37
@Magellan never had an issue using hiera
@Magellan Just like Microsoft CALs ;)
@ewwhite Yes, that is the case with Chef.
But once I figure out how many nodes I'll end up with and how to use this, I'll pay up.
@ewwhite Which I know nothing about. I let other people deal with MS licensing headaches.
(but otherwise, the license restriction is on your honor)
16:39
@ewwhite Ah. IMHO, $100/node per year is a bit steep for what you're getting.
@Magellan It's pretty buttoned-down... none of the webrick scale issues
and worked out of the box...
so in my case, the time savings made a difference... but yeah, for a larger environment, I'd roll my own
at job[-1] I usually only managed 40ish nodes on a daily basis. Another 200ish were used for load-testing.
16:59
@dawud: I just created an account here to vote that comment up : ) — DImension10 Abhimanyu PS 4 hours ago
Thank you for that valuable comment.
17:33
can someone else chuck a delete on serverfault.com/q/547215/9517please
18:06
Damn, SQL Server 2012 just Catch22'd me... Log file is full at 2TB on a non-GPT volume, and it won't let me add another log file because: Log file is full - FML
I dont know windows but isnt there that whole 'dynamic disks' magic?
@MIfe The file itself cannot grow beyond 2TB on a non-GPT disk, and you can't convert an MBR disk to GPT without removing all partitions :\
18:42
@MathiasR.Jessen have you tried...you know...backing it up?
@MDMarra I am, it just takes a while to backup 3.5TB of log :)
true
Playing the waiting game...
@MathiasR.Jessen wanna hear something that will make you feel better?
@MDMarra Very much
18:47
I just had a customer sign a statement of work for us to do an AD migration from their current environment with 3 forests into one environment with a single-domain forest.
So far so good
1,500 of their machines are for their mobile workforce that almost never VPN in. They are distributed throughout the country and the majority of them only have 3G/4G hotspots for connectivity.
They have no SCCM agent or any management of any kind on them other than GPOs that get enforced when they randomly decide to VPN in, which can be as infrequently as once a year
They also have no asset tracking, so no one knows what laptops are where or who has them
Also, they are shared machines that are strapped to X-Ray carts and are 6 to a van, so a user might have one laptop one day and a different one the next day
And I have to migrate them to a new domain somehow
over a VPN over 3G/4G
exactly
That's almost as bad as my Red hat interview question.
What was that?
For embedded Linux systems that need to run the Gogo airplane wifi and handle seamless switching between airport wifi, 3G/4G and Satellite networks...
I was asked to whiteboard my approach.
18:52
Just create a bridge out of all three interfaces, drop the marker, walk out
Hmmm... Remember that interfaces come and go. Depending on the flight progress.
@MDMarra At least your getting challenged, I've spent the last week trying to explain a customer why doubling the amount of RAM in a web server hosting a shitty 32-bit application will not double their visitor capacity
It's like they think extra hardware makes up for shitty application and data access design
yeah
that sucks
18:55
But I finally found a solution to the strange isRecycled rollback issue
I'm doing a SCSM/SCORCH architecture/design/implementation for a customer doing away with their 15 year old home-grown change management/ticketing system that has an Access 2000 backend that doesn't run on Windows Vista or later
@MathiasR.Jessen Oh, do tell
@MDMarra Well, not a solution, I tried having Microsoft PS test whether we could change the isRecycled schema class to either non-SystemOnly, or removing the replication flag, but 3 PFE's gave up after a few days, and the only viable solution is: un- and rejoin all affected machines
aka. rejoin 1700+ Windows XP machines scattered over 500 branches
wow
that sucks
gotta run to the in-laws for dinner, but I want to hear more later
@MDMarra Sure, I'll tell you about a clever workaround i've been fiddling with ;)
Gotta get something to at too
19:57
'ello sysadmuppets
Is it always this dead on a weekend? Its as if some people have other stuff better to do.
Some of us have lives. We discourage it, but can't enforce it.
Heh, I find it pretty ironic that people must be doing a lot of this from where they work (or during a working hour).
And about 50% of the conversations in here revolve around career and working conditions!
Yeah, I chat from work too in quiet times
which isn't too often :)
I actually dont.
Wait till I get home.
20:03
Weekends are the only chance I get to get any work done!
20:19
October is a mess
too many birthdays and other festivities, hardly any time left in the weekend
I think tomorrow is my first free day in october
20:32
It's a sign of underemployment. I'm not in SF when I'm in deep focus.
Any of you use ESXi host here?
I predict LARTing
@I'll-Be-Back I do
meh, docker is stupidly bound to Ubuntu. I dislike upstart. Had to ln -s /sbin/start /bin/true just to keep dpkg from complaining.
@dawud Nobody likes upstart; everyone but Ubuntu has either dumped it or is planning to.
20:41
I don't see any reason to use upstart when there is systemd, which is much powerful, easier to use and flexible.
@GioMac Right. systemd is what everyone is moving to - even Debian, eventually.
# hostnamectl
Static hostname: pris
Pretty hostname: Pris Stratton
Icon name: computer-laptop
Chassis: laptop
Machine ID: de9a6f8147226d7b3d55f1750000001e
Boot ID: d05b132751154a988508ea735729af96
Operating System: Debian GNU/Linux jessie/sid
Kernel: Linux 3.10-3-amd64
Architecture: x86_64
I already did. Not a single problem yet.
You might be interested in this: coreos.com
I had few projects few months ago and because we needed fast-booting system, fresh software, no security at all - we've used Fedora 19 as the base. Before I wrote sysvinit scripts for our applications and it was terrible from the point of view of time it took, dependency issues, logging, scenarios etc.
@dawud Oh, so that's what happens when devs get root access.
Then for stable release we've redesigned our system to complete with systemd. Service unit file contains only 10 lines, and I've got everything I need - stderr logging, startup dependency, status checks etc.
this sucks: daemonize blablablah 1>>/asdasd/asdsad.log 2>&1
20:51
@MichaelHampton Arguably, devs gave root access to sysadmins
@dawud Yeah, this seems a bit backward :)
@GioMac What version of mpt2sas you using?
No LSI in here
I am getting slow write speed (Raid 1)
`~ # vmkload_mod -s mpt2sas |grep Version
Version: Version 10.00.00.00.5vmw, Build: 799733, Interface: 9.2 Built on: Aug 1 2012
`
should i upgrade mpt2sas?
I don't think it's a driver problem
Probably, it's a write cache or disk
20:56
Latest driver is 17 though
I have LSI 9211-4i Raid Controller Card
I don't know which version is bundled with 5.0 and 5.1 and what was provided by LSI/HP at that moment, but we had no problems in 2012 with any on HP DL G7 and G8's.
It is clean ESXi installed, it seem mpt2sas version 10 is default version
And same with some Dells
Update your ESXi version. It's unreasonably old.
It should work anyway
21:01
I am using 5.1
ESXi 5.1
There are patches.
What about disks, configuration, caching etc? :)
and what does it mean "slow"? :)
@dawud Wow that looks interesting.
21:12
@I'll-Be-Back So, how many disks, what disks do you have, what about cache configuration policy on controller, on the disks etc?
@GioMac 2 x 2000GB SATA (6Gb/s) WD SE 7200rpm 64MB cache (24/7 WD2000F9YZ SE) in Raid 1
I don't remember cache configuration policy on controller - i think it set to default
@I'll-Be-Back Pretty expected performance.
It can't be
It is. It's one-disk performance. All resources say that
What do you expect from 7200 rpm SATA drive? 300 MB/s?
21:33
Meh.
21:46
@Wesley ! If you're not going to buy a goddamn bike at least be like this cat and learn a useful skill (barking)!
@voretaq7 Shuttup I'll buy a bike someday!
woof
@Wesley then hell will have frozen over ;)
@Iain Tell Apollyon to sharpen up his ice skates. =P
 
1 hour later…
23:43
@Wesley suuure you will...

« first day (1134 days earlier)      last day (4133 days later) »