|
psydude posted:Who else's leadership is freaking out about ShellShock today right before the weekend? So far we don't have any known-vulnerable public facing services/devices (  ) so not really! Still working on patching though. Defense in depth and whatnot.Also gives me a window to update some other poo poo I've been meaning to get to.
|
#
¿
Sep 26, 2014 19:57
|
|
|
|
| # ¿ May 9, 2024 12:09 |
|
|
That cat's already pretty well out of the bag. There's multiple articles in Information Week and The Register about it. Anyone who's paying attention at all to security news is aware that a patch and disclosure are coming. And I don't even run that product.
|
|
#
¿
Sep 29, 2014 23:36
|
|
|
MF_James posted:Got promoted to Sys Admin a few days ago. I'm sorry to inform you that you have not, in fact, been promoted. You've died. Goondolences. http://www.smbc-comics.com/comics/20140929.png
|
|
#
¿
Oct 1, 2014 03:30
|
|
|
Or even just "what would the business implications be if our executives all got trojans on their PC's and all of their communications, credentials and documents were transmitted to a competitor/some random Russian/a foreign government/etc". If you can't make even that case to them then it's probably time to YOTJ because you're working for complete morons. It also sounds like you're in a fairly large org. Are you subject to any regulations or compliance requirements? Having a bunch of unpatched machines connecting to the Internet should have you failing audits within 0.0001 seconds.
|
|
#
¿
Oct 13, 2014 23:04
|
|
|
along the way posted:WSUS is dead easy to implement. Might want to setup a small test group too so you can roll out the updates to them first before approving them for the whole office. Not fun coming into work realizing an overnight update caused every other client PC to not boot/do weird poo poo. Yes, do this. There was an update to IE ~3 years ago that changed something to do with how it negotiated SSL connections and it broke an app that literally the entire company I worked at relied on daily. That was a fun day. The takeaway from an incident like that (and every sysadmin probably has one) isn't "don't patch lol". It's to do exactly what Along the way said. Create a test OU that covers a small number of tech savvy users outside of the IT group. People in customer support or development or sales who actually use all of your supported apps and can provide useful feedback if they suddenly break. Don't make your test group consist of just yourself or the IT team; you don't use the apps the same way the real users do. They will catch issues you never would. Roll updates to the test group. If everything is good for whatever period you deem appropriate, update everyone else. If not, block the update and either wait for it to be reissued or work out a mitigation strategy. And god drat, set up WSUS and stop RDPing into everyone's desktop 
|
|
#
¿
Oct 16, 2014 06:59
|
|
|
Depends on a raft of things from whether you care about having control over the entire box, budget and what you want the site to be able to do. If you want full control, you want something like Linode or AWS/Azure/DigitalOcean (assuming from your question you don't want it in-house). If it's just a lovely Drupal or Wordpress site, you can probably host it somewhere for free. Or cheap on any of 5000 DreamHost clones. There's a pretty good Web Hosting Megathread in SH/SC. edit: heh god dammit, I also read it as "simple"  Time to go to bed.
Docjowles fucked around with this message at 04:17 on Oct 18, 2014 |
|
#
¿
Oct 18, 2014 03:49
|
|
|
the spyder posted:How well does it work out? Are they purchasing off a list of pre approved systems/phones? Does the department manager sign off? I ended up doing most of the end-user hardware purchasing at my first IT job, which was a ~80 person startup. I was technically under the IT Manager but he was incredibly lazy and happy to pawn off any responsibilities he possibly could to me (which was awesome for learning opportunities  ). It was pretty much as you said, we had a set list of 3-4 different machines and phones people could have depending on their role in the company (sales, developer, ~~Executive~~). When someone was hired, I'd fire off an order to Dell/Apple and have the finance person sign off. Wasn't a big deal.The responsibility wasn't handed to me on day 1, but maybe 6 months into the job.
|
|
#
¿
Oct 20, 2014 21:29
|
|
|
skipdogg posted:I've loving had it today. I'm running lead on a project that will involve at least a 6 figure annual spend with whomever we choose and I can't get a loving straight answer to some simple questions that I have. I'm going to write a plugin that changes the word enterprise to 'bullshit' just like the old cloud -> butt Heh. We're wrapping up a project of similar scope now, too. A couple of vendors that we were predisposed to really like ended up totally knocking themselves out of the running with bullshit like this. Ask simple questions (like... "how much does this cost?"), wait two weeks to get a response. If you can't be bothered to answer my questions even during the pre-sales "Everything about our product is AWESOME! There has never been one problem or dissatisfied user IN RECORDED HISTORY!" honeymoon phase, that doesn't speak well of what life will be like when we're under contract. I can find someone else who actually wants my money. Good job, good effort.
|
|
#
¿
Oct 20, 2014 23:09
|
|
|
Comradephate posted:This was the impetus for me to get good at the AWS commandline and write a handful of tools to interact with the API - the console is awful, and gets progressively more awful if you have a lot of stuff. Have they fixed the S3 web console yet, or is it still completely unusable if you have more than a couple dozen objects stored? But yeah, I pretty quickly moved to doing poo poo via the CLI and tools like s3cmd.
|
|
#
¿
Oct 25, 2014 02:23
|
|
|
Fiendish Dr. Wu posted:I really don't know either but I didn't wait to figure it out before CTRL+W I work for an image hosting company with a TOS banning certain content (mostly hardcore porn). The team that moderates content flagged as inappropriate... THEY are broken people. If you hear one of them laugh hysterically, or randomly exclaim "what the gently caress???", for the love of god, DO NOT look at their monitor. 
|
|
#
¿
Oct 28, 2014 19:34
|
|
|
Inspector_666 posted:Yeah, the definition of "working knowledge" will change with the position you're applying for. I feel like that's kind of inherent in the phrase. I agree with this. Given that The Dreamer is presumably applying for totally entry level work, I don't think a reasonable interviewer is going to look at "working knowledge of Linux" and assume he could stand up a 50 node OpenStack cloud on his own or something. Don't lie or oversell your experience, but it doesn't sound like he's doing that. That said, you always want to tailor your resume to the job anyway. If you're applying for a help desk role in a 100% Windows environment, there's not much point highlighting your Linux experience. Leave it off to help keep your resume under one page, and give yourself more room to talk up the skills you're actually very strong in.
|
|
#
¿
Oct 30, 2014 16:58
|
|
|
Tab8715 posted:To make sure we're on the same page, it seems that it's not so much the wording but a lot of people are misinterpreting what are basic linux skills. I think I see why the Red Hat exams are given more weight in the industry. I really don't care if someone has every possible switch to every single utility memorized. Or what units the timeout argument in a lilo config uses. That's what man pages and Google are for.
|
|
#
¿
Oct 30, 2014 18:07
|
|
|
mattfl posted:All the recruiters in the orlando area seem to be smoking hot young woman. The last one who contacted me, on her LinkedIn profile her previous job was assistant manager at an Abercrombie store. I swear everyone on LinkedIn uses a photo from 20 years ago. One woman I used to work with appeared to have her late 80's high school yearbook photo on there complete with amazing 80's hair and outfit. With ageism being a rampant thing I can't really blame people for doing so. But could you make it a little less hilariously obvious? I'm pretty sure the Senior VP of Marketing or whatever isn't 17. Misogynist posted:Remote workers, can I get a "gently caress yeah" for $150 used Aerons?  I scored a used Steelcase Leap off Craigslist for about $400 and thought I had gotten a deal.
|
|
#
¿
Oct 31, 2014 16:49
|
|
|
Hey look it takes time to grow a proper neckbeard ok. You can't rush these things or just fake it til you make it.
|
|
#
¿
Oct 31, 2014 21:06
|
|
|
Tab8715 posted:Eh, that's what was I told but there did seem to be common consensus that it'll eventually get there... OpenStack is getting a lot better. We've been running about 250 VM's in production on the Grizzly release since early last year and when it works, it's awesome. But holy hell was it a full time job to maintain. We're finally upgrading from Grizzly to Icehouse and it is night and day how much more performant, stable and feature-rich it's grown in that time. We'll see if that holds true when we take it out of testing and move hundreds of VM's onto the new version but so far color me impressed. Assuming it shows well in production we'll be transitioning a lot more bare metal hosts to OpenStack over the next year. But yes, you will need some staff with serious Linux chops to properly deploy and configure it from scratch. There are some nifty tools to do point-and-click deployments but if you don't understand what's going on under the hood, god help you when something breaks.
|
|
#
¿
Nov 6, 2014 05:23
|
|
|
evol262 posted:Havana was also a huge step up. GRE/vxlans are still a little broken unless you disable gso or change the MTU in neutron/dhcp_agent, I think, but it's way better. Especially heat. And DNSaaS is coming, which will finally make neutron's dnsmasq suck less, with real dynamic DNS, but I'm getting ahead of myself... We definitely abuse it to do traditional virt stuff due to extremely tight capital budgets (startup lyfe~) plus a really bad experience with the Red Hat sales people hawking RHEV. Now that Heat is starting to suck less I'm very interested in exploring using OpenStack in a more "cloudy" way in 2015. Our primary apps are scale-out and the bulk of our VM's are identical hosts stamped out from a template and configured with SaltStack, but capacity is only added or removed manually. We're not doing any sort of autoscaling. And there's a decent number of random one-offs like an internal IRC or FTP server that have no real business being in "the cloud". Interesting to hear about DNSaaS. We've effectively rolled our own on top of PowerDNS with the MySQL backend by processing OpenStack messages as they come across Rabbitmq and then adding/deleteing/modifying DNS records in the DB as appropriate.
|
|
#
¿
Nov 6, 2014 05:55
|
|
|
Misogynist posted:10x traffic spike from Tumblr ruined my Sunday. How's everyone else doing? As an image/video host, the day all those celebrity nudes leaked a few months back was fun. 
|
|
#
¿
Nov 11, 2014 19:36
|
|
|
Erwin posted:The ones from last year don't work on macs. If you do it again, you should find better ones? I got last year's keychain and use it all the time on a variety of different Macs with no problems  Maybe you got a lemon.
|
|
#
¿
Nov 13, 2014 06:23
|
|
|
Yeah in my limited experience public school pay is all over the place. I grew up going to one of the better high schools in a state that spends a lot on education. My mom still works in that school system and the IT Director is compensated extremely well. Then I moved to a state that's among the lowest in education spending. While job hunting I looked into a posting for basically that same job in a nearby school district. IT Director for the entire district paid less than $20k/year  Not 120. 20. The loving janitor probably made more.
|
|
#
¿
Nov 13, 2014 17:49
|
|
|
Paladine_PSoT posted:Is this the year we make "gently caress printers"? Only if it can come preloaded with some Stuxnet type virus that bricks any printer you plug it into. I will take 500, please. 
|
|
#
¿
Nov 13, 2014 18:09
|
|
|
Tab8715 posted:Where the hell do you live? I'm in a town of just over 100k/y+ and plenty of companies are willing to pay six-figures starting. Yeah I am going to say that your town is not the norm. Six figures is not at all unreasonable in IT but those jobs aren't growing on trees in <random small city>. If they're plentiful it's probably because you're in a super high cost of living area (like SF/NYC where $100k is barely a living wage) or someplace no sane person wants to live without major incentives. Sounds like the latter. edit: Also you're in town where $100k/yr is easily attainable yet you're drooling over a job that pays $39k/yr (that's 3300/month x 12)?
Docjowles fucked around with this message at 04:46 on Nov 14, 2014 |
|
#
¿
Nov 14, 2014 04:42
|
|
|
e: nm shouldn't be contributing to this shitshow
Docjowles fucked around with this message at 06:24 on Nov 15, 2014 |
|
#
¿
Nov 15, 2014 06:21
|
|
|
evol262 posted:drat. You're pretty jaded for being that young. I guess that's what happens when you live near DC I assume all posters are literally exactly like their avatars (myself included, naturally). So psydude is forever a haggard-as-poo poo Joe Perry and his jaded world view makes total sense.
|
|
#
¿
Nov 16, 2014 03:36
|
|
|
Dark Helmut posted:This is totally true. I always read your posts with the voice of Wilford Brimley. FYI you have The Diabeetus. Sorry to say. Actual avatar backstory: it was some identity theft awareness ad. The caption was something like "I bought $30,000 worth of liposuction!" My friends and I wrote for a comedy newspaper in college (think a shittier 'The Onion') and one of those guys used this photo in an article as a citation from "Dr. Roland P. Jowles". 20 year old me thought it was hilarious and for some reason over a decade later that is still my online identity  You used to be able to reverse Google Image Search my avatar and find the original ad but these days it's mostly my posting history and a few other idiots who apparently also found this guy's visage hilarious.
|
|
#
¿
Nov 16, 2014 05:10
|
|
|
Was it this thread where we were talking about OpenStack last? I forget. We've moved about half our our production infrastructure from Grizzly over to Icehouse and it's NIGHT AND DAY how much better it is to work with. Stupid poo poo we had to put in gnarly hacks to deal with now Just Works. CLI commands and the Horizon dashboard are literally several orders of magnitude faster. You don't have to reissue the same command 10 times before it runs. I was worried that would go away as we upped our VM count but so far it's still super fast and reliable with 100+ VM's on the new cluster. It's still a far cry from the ease of VMware deployment but I'd actually feel good about telling someone (who understood what they were getting into) that it's a viable option now. You're still going to want at least one person on staff who understands it on a deep level but they shouldn't have to burn 80% of their time chasing issues that boil down to a "fixed in a later release" bug report anymore. Which was definitely the case in previous versions. Pretty cool to see that all the noise, money and hype around OpenStack is actually going toward making it a usable product.
|
|
#
¿
Nov 17, 2014 07:52
|
|
|
Fiendish Dr. Wu posted:c'mon man don't you know by now? ...and then come back to it drunk as hell to shitpost.
|
|
#
¿
Nov 17, 2014 19:30
|
|
|
Zero VGS posted:I am taking quotes from local Cisco techs and I have a resume that says "CCNA, CCNP, CCIE (passed written exam)". That just means he never took or didn't pass the CCIE lab? It probably means he's still studying for it or has it scheduled but it hasn't happened yet. Wouldn't count it as a black mark. e: But yes ask for clarification. I guess he could have decided "gently caress it, not gonna do the lab" which would be dumb, if unlikely. DrBouvenstein posted:Just had my first night of being on-call...already hate it. I've never worked anywhere where on-call paid extra. But I've mostly been on salary and paid enough to be above the threshold for mandatory overtime compensation. And when I wasn't, I was too young and dumb/naive to realize it . Current job lets you take a comp day with no questions asked if your on-call shift was incredibly lovely which I consider fair. Previous job I was literally on-call 24/7 with no compensation at all and it was loving awful. A big part of why I no longer work there and have no more interest in doing the "lone sysadmin" thing.Getting alerting to where it's only paging you on an ACTUAL EMERGENCY  is both extremely important and extremely difficult. There are tons and tons and tons of blog posts, conference talks, papers etc on the topic since it's by no means a solved problem. This could be a whole thread on its own, really. Some things to think about :If I get this page, is there actually anything I can do to fix it? If not, don't page on it. It's just noise. Is there a real issue but it's one that can wait til morning? Send an email or show the issue on your dashboard but don't page. Is there no problem at all? Look at your thresholds, maybe they're too sensitive. Or maybe you're monitoring something 100% pointless. Docjowles fucked around with this message at 19:48 on Nov 18, 2014 |
|
#
¿
Nov 18, 2014 19:44
|
|
|
So we JUST completed a huge rear end project to change to a new CDN vendor (EdgeCast). It has a lot more features and is literally 1/3 the cost of what we had been paying. Been a really great experience so far. Seen today: Chinese government blocks all traffic to EdgeCast lmao are you loving kidding me 
|
|
#
¿
Nov 18, 2014 20:18
|
|
|
Tab8715 posted:That's best part about corporate IT. that sounds spooky as gently caress  
|
|
#
¿
Nov 24, 2014 21:35
|
|
|
Tab8715 posted:I see what you did there Today I learned Maddox still exists. Hadn't thought about that dude since about 1998.
|
|
#
¿
Nov 24, 2014 23:52
|
|
|
There's so many players involved in OpenStack that there's definitely a lot of jostling to make it "all things to all people". At least in terms of how vendors market it But yes, when you boil it down it's about running your own cloud/IaaS platform. Most of the projects that make up OpenStack map to AWS products in some way, even if it's not exactly 1:1 EC2 == nova (compute virtualization) + neutron (SDN) S3 == Swift EBS == cinder RDS == trove CloudFormation == Heat IAM == Keystone GUI Dashboard == Horizon AMI's == Glance images The same can probably be said about Azure and GCE but I'm not as familiar with those. Docjowles fucked around with this message at 18:14 on Nov 25, 2014 |
|
#
¿
Nov 25, 2014 18:06
|
|
|
Fiendish Dr. Wu posted:So what's everybody's work load on pre-t-day?  Doing that, and handling misc chores that have no chance of breaking production. Adding or fixing monitoring for things that have been bugging me. Cleaning up documentation
|
|
#
¿
Nov 26, 2014 17:37
|
|
|
hihifellow posted:What, you mean a connection with a website that once told a victim of miscarriage that her "womb is making heaven too loving crowded" might not be a benefit to gainful employment? You can selectively hide your group memberships. That said, the SA LinkedIn group is effectively dead. The last thread is from two months ago and is literally DAF offering to buy people poo poo if they start making good posts
|
|
#
¿
Nov 29, 2014 03:59
|
|
|
psydude posted:Reneging on an offer after you've already accepted it is a pretty bad move. This. Don't do it.
|
|
#
¿
Dec 2, 2014 19:57
|
|
|
syg posted:On the topic of LinkedIn, do you guys bother filling out the summary? I feel like its definitely worthwhile but seems like it might give my employer the wrong impression. I'm not actually looking I just like networking with recruiters and having a presence so if something were to go badly in the future I could hit the ground running. Until you fill out your profile completely, you're heavily penalized in search results on LinkedIn. So if you actually want recruiters to find you, it's worth it. There's a good How To Do LinkedIn thread in BFC if you're interested. When you look at your own profile, there's a Profile Strength meter in the sidebar. You want it to say "All-Star". Adding a photo is another thing people often miss that counts toward this. Disclaimer: This was the case a year or two ago, I haven't kept up on whether it's still accurate.
|
|
#
¿
Dec 4, 2014 04:54
|
|
|
BaseballPCHiker posted:Those of you who work from home, what do you do for work and what's your job title? Right now I seem to mostly do SCCM work and help out help desk when times get busy. On the few days that I have been able to work from home I've gotten way more done. I'd like to start working from home at least half time in my next position and want to know where to focus my efforts. Honestly right now in my normal day to day work there is nothing that I do that I need to be in office for. I have a networking background and use to do plenty of ASA configurations and remote site work and wouldn't mind doing that again either. Just getting tired of getting bugged for annoying small poo poo from execs and or anyone else who passes by. I work from home 3-5 days a week depending on what I have going on. System Administrator (officially the team lead, though I was working from home before the promotion too) in an operations role running our public services, not user-facing IT obviously. We have remote hands data center guys so it's extremely rare that I actually need to go into the DC to physically do something. When I do come in it's usually for meetings, or just to get some interaction with someone besides my wife, baby and dog I could probably work from home 99% of the time if I really wanted to but I start to get cabin fever. And for meetings where I actually need to contribute, it's just harder to do that over a lovely conference phone connection than in person.I also live 60-90 minutes one way from the office. If I was closer I'd probably go in more often. There are tons of tech jobs that COULD be done entirely remote. Whether the company will allow it is the big hurdle.
|
|
#
¿
Dec 4, 2014 16:54
|
|
|
pofcorn posted:Been using DokuWiki for a few years. Works fine. Yeah I've used DokuWiki forever and it's fine. Lots of plugins if you need some extra feature.
|
|
#
¿
Dec 5, 2014 16:36
|
|
|
Tab8715 posted:What's the best explanation for OpenStack? "running your own in-house AWS". I think what evol is getting at is that it's made up of a bunch of semi-independent projects all under the OpenStack umbrella. So one company might only run Swift object storage while another runs Nova + Neutron + Glance + Cinder (which basically add up to Amazon EC2). Still another might use Neutron to do fancy SDN stuff for their bare metal servers but not manage them with Nova. And yet all of them are "running OpenStack" and "have a private cloud". But really that's no different than AWS. It's entirely possible to be an AWS customer that only uses S3, or EC2, or any of their 7127381723 other products. You're still "running on AWS" and "in the cloud" no matter which products you choose off the menu. OpenStack and AWS are both giving you API's that let you provision and manage resources (compute, network, storage). It's just that OpenStack's API endpoints are in your datacenter whereas Amazon's are at, well, Amazon. Sorry for the gratuitous use of quotes. I may have a  problem .fake edit: I am so thankful I stumbled into data center IT / web operations. It's 10000x more interesting to me than corporate IT ever was. Not knocking either one, I'm sure there are folks who are just as passionate about managing AD and Exchange as I am about cloud computing and infrastructure-as-code. It just tickles my fancy more personally. Docjowles fucked around with this message at 05:53 on Dec 11, 2014 |
|
#
¿
Dec 11, 2014 05:50
|
|
|
Dr. Arbitrary posted:I'm sort of curious if DevOps might be the direction I want to go with my career. It just is kind of a nebulous job description. Much like ~cloud~ it's been coopted to mean literally anything depending on who you ask. The guys who actually coined the term talk about it in terms of a company culture moreso than a job title or role. They summarize it as "CAMS": Culture, Automation, Measurement, Sharing. If your organization practices DevOps, you have a culture of collaboration and sharing instead of building silos and kingdoms. You automate the poo poo out of everything. You measure and monitor everything so you can make data-driven decisions. So a team practices DevOps, but you are not personally a DevOps. That being said, companies now use it to mean anything from "literally just a traditional sysadmin with a different name so it sounds cooler" to a build/release engineer to someone that writes internal tools to a syadmin who knows what Puppet/Chef are. If you enjoy scripting and automation and config management and The Cloud and collaborating with developers instead of fighting them, a job with DevOps in the title is likely to be interesting (even if it's kind of a stupid use of the term). But those are also skills a good sysadmin in 2014 should have anyway regardless of job title. Fake edit: more or less what JHVH-1 said as I was typing this Real edit: And yes, these positions generally pay VERY well, actually. Because unlike Jack of All Trades IT Rogues it's very hard to find qualified people and those who are know their value. Docjowles fucked around with this message at 18:24 on Dec 11, 2014 |
|
#
¿
Dec 11, 2014 18:20
|
|
|
|
| # ¿ May 9, 2024 12:09 |
|
|
psydude posted:Isn't DevOps just the theory that if you put the onus of running the infrastructure on the developers, both the application and the infrastructure will suck less? It's more the theory that if people and teams have empathy for each other they will not be shitlords and by extension the infrastructure (and the apps that run on the infrastructure) will suck less. Because they've been designed for each other instead of done totally separately and then mashed together like a 3 year old with some Play-Doh. "lol make devs carry a pager" is like 1% of the big picture.
|
|
#
¿
Dec 12, 2014 04:41
|
|