|
Anyone have any experience in automating SSH to talk to Cisco equipment? I've been using Net:  :Perl, Net::Open-SSH, and a little bit of Powershell-SSHSessions but they're all kind of crap. Well, Net::Open-SSH is pretty decent but unfortunately none of the RSA keys on our networking devices are 768 bit so OpenSSH refuses to connect to them. Recompiling OpenSSH or re-generating RSA keys are not available to me at the moment (I'm the junior).
|
#
¿
Sep 20, 2014 08:13
|
|
|
|
| # ¿ Apr 16, 2024 07:49 |
|
|
hanyolo posted:RANCID works for basic commands if you're just doing a simple script. The proper way to do this these days is by using Netconf: I should have mentioned I'm querying DHCP for an array of ip and mac addresses and then doing a layer 2 traceroute for all of them, approx 800 IP addresses across 6ish subnets, and then probably another 200-300 or so in production. It was requested that I update the switchport documentation because it's all kinds of hosed. I ended up using Net: :Expect which I got from browsing around CPAN. From what I understand it's basically a perl wrapper for Expect which is good because actually writing any Expect was making me strongly consider drugs and alcohol. Hilariously enough I didn't have to regenerate any of the device keys because all of them just let you telnet right on in. That's probably something to take a closer look at in the future  .My predecessor(?), the previous networking dude, claimed to be a CCNP and actually inserted it into his company signature (which goes against policy) and then changed his title on internal documentation from Systems Administrator to Network Administrator. Turns out he had actually admitted to my boss a year or so back that he only ever took and passed CCNP: Switch. The infrastructure is about as hosed as you'd expect it to be given that this person had been in charge of it for the past 2-3 years. We had STP configured, except it was the old pvst edition so our convergence time was looking like ~40-60 seconds. Not that it mattered, anyway, because we didn't actually have any redundant switch links, which I found out when one of our fiber uplinks went down and we lost connectivity for an entire building. Half of the devices weren't configured for radius authentication because he had no idea what radius even was, and going by our e-mail logs RANCID had been emailing the department on the hour, every hour from mid-2011 to about 2 months ago. The latest network diagram was taken 1.5 years ago and, until he was fired, was pinned to his corkboard with root accounts/passwords written on it in pen. (e: yes some of them were cisco cisco) I'm fixing most of it. Unfortunately my boss doesn't really know anything about Cisco devices, doesn't trust me because I didn't lie about having a P instead of an A on my resume, and is strongly resisting my attempts to make any changes. I don't particularly blame him, as he is ultimately responsible if something were to break, but it can be frustrating at times. 12 rats tied together fucked around with this message at 05:58 on Sep 24, 2014 |
|
#
¿
Sep 24, 2014 05:54
|
|
|
BurgerQuest posted:
Same. Well, not about earning more than daf, but in approximately November 2012 I stopped working as a cook and two jobs/two years later my pay has easily tripled. Thanks a bunch goons. My resignation was not handled particularly gracefully - I asked for a $4/hr raise, they balked, I got a job offer and they counter-offered with a $30k increase. I don't really know how they expected me to take them seriously after that. New job is neat. I went from Jr Sysadmin (IIS monkey) to "Network Engineer" in a linux environment. It's been an interesting culture shock so far, even simple things like "escape ." keep blowing my loving mind.
|
|
#
¿
Nov 20, 2014 03:35
|
|
|
BaseballPCHiker posted:Those of you who work from home, what do you do for work and what's your job title? Network Engineer. I manage production/corporate networks for a software development/SaaS company. Company policy is Tuesdays and Thursdays are work from home days, but I think all of our customer support staff, HR, and finance departments are either permanently remote or show up at the office 1 day per week. C-levels apparently drop by for a few days every other month or so, so the office itself is just Dev + Ops + HR once a week. It's really, really nice to be able to just get out of bed and basically be at work, and this might be due to my relatively isolated position but it's also nice to have an entire room to myself with no interruptions to work out a problem or change request. I miss the random jokes/humor from working in an "open floor plan", but it makes sense with a smaller team of more specialized individuals to give everyone some space. I don't like how long it takes to work through a problem that involves someone else - when it's just a simple change or trace or whatever I can get it done way faster than normal, but if it's a multiple "department" spanning job, it takes forever to get the Virtualization guy, Network guy, and Database guy on the same page. I can't just walk over to their desks and be like "Yo this is broken and it looks like a VM problem". It definitely feels more like a coalition of individuals that share a work tracking system more than it feels like an "Operations Team". But, this is probably about 50% culture and 50% working from home.
|
|
#
¿
Dec 4, 2014 21:20
|
|
|
Fiendish Dr. Wu posted:Question about automation Probably yes, if it's running on a windows machine (I have no idea what Troux is). It sounds like your results you want to check are part of the GUI, so automating the monitoring of this depends on how poorly implemented the GUI is. Is there a log file or some kind of XML document somewhere on the physical machine you can check? Is the Troux "GUI" technically a webpage, maybe?
|
|
#
¿
Dec 5, 2014 21:19
|
|
|
OwlFancier posted:That... is rather more simple than I would have anticipated. Not only can you get paid to do things like that, if you live in southern california I know a guy who might have an opening for you. But, pretty much no matter where you are, if you can get an interview with a technical person and give answers like you did here, you'll probably have the job no problem.
|
|
#
¿
Dec 10, 2014 02:19
|
|
|
Fiendish Dr. Wu posted:I used them and I like them. I'm continuing to use the template they provided. My current boss told me that it stood out and made the initial selection for choosing who to interview very easy and mine was picked right away based on presentation alone. I had a similar experience with mine, and was interviewing for a Jr Systems Administrator position. Same resume plus a few bullet points got me an interview for a Network Engineer position about 9 months later. Definitely get the ball rolling on it ASAP, though, or get ready to spend extra for the expedited response time. I think it took me like 3 weeks to get the final draft, and then another 2-3 days for them to OK the final draft and return it to me in a finished format.
|
|
#
¿
Jan 22, 2015 23:36
|
|
|
Drunk Orc posted:When is an acceptable time frame from landing your first IT job to looking for something better? I think I've read 6-12 months in this thread a few times before, but I feel like I'm not really learning too much of anything useful and have a ton of down time. This is from a page or so back, but I started looking at ~6 months into my first IT job when I got a $1 raise after a stellar performance review. This was after accepting a ridiculously low (like, cashiers are making more per hour than me) wage to get in the door, so I went from "8k under the poverty line" to "6.2k under the poverty line". I found another one around the 9 month mark that I particularly liked and relocated to Chicago for a ~60% pay bump that turns into a ~75% pay bump after 6 months. None of the interviews I had were particularly concerned that I only spend 6-9 months in my first position, they were more interested in what I could do for them and where I was at technically. This was from Jr Sysadmin (to be fair, it was a real rear end sysadmin job and I learned a ton) to Networking positions, though. In my experience the bar is set quite low for non-provider networking.
|
|
#
¿
Feb 18, 2015 16:36
|
|
|
Tab8715 posted:If I can slowly read some bash/powershell/python with the help of google does that count? If I need to make a script am I allowed google, grab pieces from other scripts or do I just have a blank prompt? I put Perl / Powershell / etc on my resume and I generally expect to do some whiteboard coding. I usually admit that my syntax without an IDE is kind of rusty and I will probably miss a semicolon or something somewhere, and everyone so far who has asked me to do whiteboard coding or similar has been understanding and said "psuedocode is fine". I handle my scripts with the same level of seriousness that I would a real rear end program and it annoys me to see the stereotypical "sysadmin script" be like 6 levels of nested for loops that takes 2 hours to run and is full of hardcoded paths or config files that you have to edit manually. I've also found that, usually in programming interviews or positions that are more dev than ops, they generally want you to do some weird tricky algorithm thing. All of my ops coding interviews where I have to write something are usually really simple problems where the solution is a where loop with 2 conditions or whatever. quote:Would powershell be a good starting point to jump into coding or should you begin with something else for this framework? It depends on what your position is and what you want to do. If you are working with Linux, I would tell you to start with C (some people would say Python - personally I don't know any Python but it's probably fine too). If you are working with windows, Powershell is absolutely a good starting point but please please please excessively use Get-Member while you are playing around with stuff. A lot of "I just started Learn Powershell and..." questions on the internet are the same misunderstanding applied to different situations. e: If you are a networking guy, you should start with Expect. 
12 rats tied together fucked around with this message at 23:31 on Mar 13, 2015 |
|
#
¿
Mar 13, 2015 23:27
|
|
|
Fiendish Dr. Wu posted:I was actually going to ask about this. I started looking for new positions about 6 months into my first job when I had a stellar performance review and a pay bump from $13 to $14 dollars an hour, despite scripting almost all of our (the jr admins) duties and freeing up tons of time to help the mid/seniors with projects. I would say that its harder to land an interview in general, but the ones that respond are generally seriously considering you as a candidate. I applied to probably 15-20 jobs per month and had 6 interviews in 4 months before I accepted a position. Generally it went Screening -> Phone Call -> Skype -> In-person. Out of companies that wanted an in-person, all of them except for one offered to pay for my hotel and flight. In my experience most Skype interviewers expect you to have a webcam (I did). I didn't change my address on LinkedIn or anything, but asking why I was looking to relocate was probably the first question at each (and every) stage of all the interviews, so get ready to answer that a lot. W/r/t Cellphones, I've been offered either a monthly bonus on my paycheck or a company issued phone. The understanding is that I am available 24/7 during an on-call period and available 24/7 as a subject matter expert at all times in the event of a serious outage. The second thing never happens, the first thing happens once a month. Really, it's up to you to decide what is acceptable, but I would definitely never use my personal cell phone for business after hours. In my opinion, anyway, it's actually better for the corporation to issue you a phone and laptop because the phone can function as a hotspot so you are never unable to access things without being personally responsible for it. 12 rats tied together fucked around with this message at 17:12 on Mar 18, 2015 |
|
#
¿
Mar 18, 2015 17:09
|
|
|
evol262 posted:I have literally never seen anyone get a raise above 5% without a title change, and I wouldn't believe this if you told me in person. My previous employer offered me a 60% raise after it became clear that I was leaving, and this would have been a ~250% raise over my hiring rate after 11 months. But, this was more of a "We screwed you on your base rate, screwed you again on your performance review, gave you a knee-jerk raise when your department started quitting in droves, and then threw an arbitrarily high number at you in an attempt to get you to stay" situation. Accepting the 60% raise would have undoubtedly come with quadrupling my job duties with no warning or support (or coworkers) and then firing me after 2-3 months when they managed to find some more people. quote:I asked for an intermediate commitment until I get there, gave a range, and they said they're working on it.
|
|
#
¿
Mar 20, 2015 15:28
|
|
|
22 Eargesplitten posted:Does anyone know how to make a full screen program in general or Windows remote desktop in particular start on a secondary monitor? As it is now, I usually have to drag my ticket window to the secondary, or put the remote session in windowed mode and drag it to the secondary screen. Right after you RDP (usually I do win+r "mstsc", hostname, enter) you can move the window with win+arrow keys. If your secondary monitor is to the right, for example, tapping win+right arrow key twice will move it over there. You can also use win+up to 'fullscreen' (which might not work with mstsc depending on the resolution you configured it to use), and win+down to 'detach' from fullscreen.
|
|
#
¿
Apr 24, 2015 17:38
|
|
|
mayodreams posted:I need to scan my server subnets to find out what versions and quantities of Windows Server we have running. Anyone know of any free tools that can do this? I have probably 90% accounted for because VMware will tell me, but I've tried the trial of LanDesk and it only differentiates between Server 2003 and XP, with the rest being just Windows Server. You can do this in Powershell with WMI. I think the class is win32_operatingsystem or something, a quick google should be able to tell you - make sure that when you grab the object though you examine the 'caption' property, IIRC, unless you want to try and decipher hundreds of instances of "Windows 5.3.8.44.344882" and poo poo.
|
|
#
¿
May 1, 2015 15:46
|
|
|
In my (somewhat brief) stint as a junior sysadmin for a windows shop, I tried to powershell-ise as much of my job duties as possible. It doesn't necessarily apply to this situation specifically but my goal was to keep everything code-driven as much as possible because it increases consistency and functions as running documentation -- there is no question as to how a task is completed when you can pull up task.ps1 (actually, it was in a module so it was get-help do-task) and look at the source. Pretty much every job duty I had was in powershell and I spent a good 80% of my day in the ISE, so downloading a third party utility wasn't really an option. Having functions for everything I did encouraged me to approach problems in a manner that made codifying them easier and allowed me to react to the possible outcomes automatically and without too much extra work. Stuff like "is x part of the application running properly?" instead of being a line item on a checklist is a cmdlet in a module, and "if this stops working, check the logs for blah blah, then do blah and restart the services" became automatic tasks that I got an email about when they occurred. I definitely feel like it created a positive feedback loop in that the more tasks I automated or turned into cmdlets, the more time I had available to do the same for others. I don't work there anymore, but my boss has informed me that the "Jr" position doesn't really exist anymore because all the grunt work has been automated, they just have regular sysadmins and the desktop support (which, unfortunately, is not automatable) goes to whoever is least busy. Granted - that doesn't really matter or apply to "I just need to count our OS versions", but IMO if the option exists to have the data you need be contained in a language that your infrastructure can gather, evaluate, and respond to automatically or be something that you copy paste from an application window into an email... you should go with the former whenever possible. I had this same task and while OS version is probably the least useful bit of info to have in powershell (it doesn't change very often), the task quickly ballooned into a remote "get info from machine" cmdlet which is currently used to automatically update the server inventory checklist with info from the BIOS, tag numbers, and all kinds of other poo poo that they used to fill out manually. 12 rats tied together fucked around with this message at 17:03 on May 1, 2015 |
|
#
¿
May 1, 2015 17:00
|
|
|
My "other duties as assigned" bullshit from my first IT job ended up being becoming an on-call contact for after hours breaches of building security such as bricks thrown through windows, motion sensors tripping, general poo poo like that  .The other part was I was responsible for climbing up onto the roof and loving around with the AC unit if it stopped working. They literally gave me a checklist of "try this, this, and then this and if that doesnt work call this number". Combination Computer Janitor, Security Guard, and HVAC Technician. e: I also used to help the facilities manager fix bathroom fixtures and, every other friday, take out all the trash from both buildings. Lmao. gently caress those people.
|
|
#
¿
May 5, 2015 19:17
|
|
|
Absolutely, it's definitely more of a "I applied (and you hired me) to be a systems administrator, why am I raking dirt?" sort of situation. e: I have been an IT worker who gets paid less than the shelf stockers and landscapers, also. 12 rats tied together fucked around with this message at 20:32 on May 5, 2015 |
|
#
¿
May 5, 2015 20:28
|
|
|
Pretty much yeah. I had more to add -- I have a lot of problems with certain low pay industries, I was a cook for 7 years and spent a lot of time with illegal immigrants, convicts, or other "undesirable" elements of society working for $<10 an hour because they have no other options but... decided this really wasn't the place for it. poo poo got pretty weird pretty fast in here.
|
|
#
¿
May 5, 2015 20:34
|
|
|
I like how "no netapp experience... that's a setback" is even a thing. Like, I understand that SANs are complicated pieces of technology or whatever and that there is definitely some value in having netapp experience, but in both of the places I've worked at we ran multi-host xen or vmware clusters with ~250 VMs using netapp for storage and 99% of the day to day with them has been "expand volume, shrink different volume, possibly create new volume". Sometimes you might have to click a few dropdown menus to find the right graph so you can tell <department> that <application/service> is running at 4k more iops/sec than <previous version> and they might want to take a look at their poo poo. That's about the extent of my, admittedly pretty junior, experience with netapp stuff. They also have a pretty great powershell toolkit, so if you have even a passing familiarity with powershell you probably don't even have to open up the GUI. If homeboy is interviewing for a literal Storage Engineer position and he doesn't have NetApp experience, sure, yeah, that's a setback... but if this is a generic sysadmin position and they can't be hosed to walk you through opening up the GUI and clicking on maybe 8 different buttons I'd be really worried.
|
|
#
¿
May 6, 2015 15:31
|
|
|
Honest Thief posted:So is everyone on even remotely IT related jobs just miserable? Is there such a thing as a 'it gets better' PSA? Network engineer for a SaaS platform. It owns - no users, no customers, just use technology as a resource to accomplish business goals or whatever. The only bad parts about it are generally specific to the technology itself. Cisco IOS kinda blows (especially compared to like, bash) , your 'standard' network devices aren't really designed with mistake-making in mind so loving up can have really harsh consequences, and in general there seems to be a lack of skilled workers in small/medium business networking so you run into some really questionable design/architecture decisions that make things way more complicated and difficult than they need to be. The rest of the department is trying to, within reason, embrace the 'fail fast and often and recover quickly' mentality but that doesn't really apply to the network devices IMO -- they made sure to repeatedly state that every network engineer to date has blown something up and caused at least one outage and that "as long as I learn from it, it's okay" but... in my opinion if you gently caress up a network and cause an outage it's because you didn't spend enough time reading the documentation or examining the impact of your changes. Unfortunately for me, that means I spend more time reading documentation and examining the potential impact of my changes than I do actually making changes which can be a little boring at times. Before this I worked at a more standard IT job, about 50/50 user support and server administration. In my experience, it's fine and perfectly rewarding as long as your infrastructure isn't terrible. I had the benefit of the department having a history of remarkably competent admins before I started, so most of the usual systemic bullshit I'm seeing in other jobs was already taken care of like automatic updates, PXE images for user machines, nobody has local admin, ESET antivirus deployed everywhere automatically, group policy ran like a well oiled machine and no printer issues to be found anywhere. The remaining user support after that's taken care of is fine. Enter-PSSession owns, MSRA owns, spent most of my time working on powershell scripts which also owns when your environment is 100% windows 8 and server 2012.
|
|
#
¿
May 13, 2015 15:46
|
|
|
GNS3 + a router image will appropriately emulate most of what you need for the CCENT. I understand Cisco has a thing now that can emulate switches but I'm not sure how pricing works for it. e: f, b.
12 rats tied together fucked around with this message at 16:21 on May 13, 2015 |
|
#
¿
May 13, 2015 16:19
|
|
|
Methanar posted:I'm retarded. This is so obvious. This is going to keep happening for the foreseeable future. When it stops happening is probably a good time to start looking for a new job, also.
|
|
#
¿
May 14, 2015 01:41
|
|
|
Yeah I usually try to make sure the answer to that question is "Not me". Previous job, it was the other jr sysadmin. Current job our company is very small and we don't spend a lot of money on hardware, so most computer purchases come from the director for new employees or from the sysadmins if it's a replacement for an end user (we have about 15 end users). I actually watched a dude get fired (mostly) for slightly loving up a PC order. We ended up purchasing like 30 new desktops and 20 new laptops, but for some reason all of the desktops came with 2x DisplayPort video cards and we only had like ~6 DVI to DisplayPort adapters in the house. I guess they had to tack on the additional expense for the adapters to the purchase order for the desktops + laptops themselves for some bureaucratic reason and it "delayed the period" or something? Accounting was pissed, the CFO was pissed, he ended up leaving shortly afterwards. I try to avoid spending company money as much as possible because I'm positive that I just do not give enough of a gently caress to spot mistakes like that. To be fair, it was out of his control that we needed to order small form factor desktops (came straight from the CEO's wife) which had a tendency to blow out their power supply once every ~9 months and had to come with all sorts of other nonstandard components, but still support 2x monitors. Also apparently our Dell rep had changed 3 times in 3 months and the new guy neglected to mention that the desktops were coming with 2x displayport. Still "his fault", but yeah, definitely not something I ever want to deal with.
|
|
#
¿
May 15, 2015 15:28
|
|
|
You were supposed to ask him for a static IP for use on a laptop? Really? e: It looks like you started from the bottom (physical layer) and worked your way up. In my opinion that would be pretty much perfect - if I confirm the cable is actually plugged in my next question is usually "do i have an ip address?", so you pulled up the adapter config and noticed that no, the settings are garbage and then you set it to DHCP, got an ip address, and then it worked. Seems perfect to me. Pinging or tracerouting would also clue you into the fact that you might not have a valid configuration for the adapter, but "i can't ping google" doesn't immediately mean "i need to ask someone for a static IP address". 12 rats tied together fucked around with this message at 19:37 on May 16, 2015 |
|
#
¿
May 16, 2015 19:33
|
|
|
22 Eargesplitten posted:I noticed PowerShell in a month of lunches doesn't seem to have a version for PowerShell 4, how much am I missing by only learning v3? What you should be looking to learn from "Learn X In <Time Period>" is the basics of the language, syntax, how to do your basic If/While/For/Etc. Learn how to Do An Array (and a hashtable, and the difference between the two), and then hopefully your study material will clue you into any Gotchas! within the language that might exist (like how Python will do funny things to 1 vs "1" that might not be immediately obvious). For powershell this is probably "Always Get-Member Everything" and being aware that what the shell displays to you isn't necessarily a 1:1 mapping of what the object you are interacting with actually is. In general, powershell makes objects (and related terminology) really easy to understand, but if you have any issues with it you can ask here or in the powershell thread. Once you have that down, you can pretty much put the book away and just google whatever task you are trying to accomplish, which kind of depends on your role in your org. I'd particularly recommend checking out Get-WMIObject, Get/Set-ADUser, and you should learn how to go through a text file and change lines if they meet a certain criteria, just as general exercises. Once you have that down, and if you are able to, get powershell remoting working (plenty of blog articles on this - it probably involves a GPO change though) and you start solving end user problems with Enter-PSSession if you like. Another good project is figuring out how to get a user's mapped drives on a given machine, and that came up for me fairly often.
|
|
#
¿
May 18, 2015 16:55
|
|
|
I'm struggling to recall the exact, specific example that made me aware of this because it involved an old coworker's script that he was having a really hard time with. It's actually kind of a poor way of phrasing it anyway, my bad. The best example I can think of is Get-Content. Without doing Get-Content | gm, you would (probably) never know that instead of returning a large string with newlines (because that's what appears on the console), you are actually being returned a collection of string objects. Which could run you into some issues when you assign "get-content -file" to a variable and assume that it is one rather large string.
|
|
#
¿
May 18, 2015 17:50
|
|
|
mayodreams posted:From an O365 perspective, it is almost imperative to do a get-mailbox mayodreams | fl if you want to look at the extended attributes for the user. Once you have a few down though, its easy to pop in the ones you want. The one I use all the time is: I believe I was working with o365 sharepoint sites or something when I decided to completely stop trying to do "Get-Thing | select thing" and just started doing $Object.Parent.Child.Child.Method() because it gets pretty loving crazy in there. Fortunately, I stopped having anything to do with sharepoint pretty shortly after I discovered what a headache it is. But, if you don't like "| fl" (which I don't, but I can't really reasonably articulate why), you can probably do "| select *". Maybe. I feel like last time I was pssessioned into o365 a lot of the basic functionaly was broken and weird. 12 rats tied together fucked around with this message at 20:22 on May 18, 2015 |
|
#
¿
May 18, 2015 19:43
|
|
|
crunk dork posted:How did you network guys get started? CCNA + lots of home labbing, got a job as a jr systems administrator for a web-based SaaS company. After ~6 months or so I started applying for jobs, applied for a Network Engineer position and they liked me enough to bring me on board. Apparently the networking talent pool in Chicago is really, really limited because this environment is laughably bad and at ~8 months of experience in IT I am pointing out huge infrastructure design problems that the previous guys with 5+ years of experience on me had been making. So, I would recommend moving to Chicago I guess.
|
|
#
¿
May 23, 2015 20:15
|
|
|
ElGroucho posted:You sound like a cool dude This would probably be better than whatever that one is on HBO right now, to be honest. Maybe you guys should start a podcast.
|
|
#
¿
May 29, 2015 22:59
|
|
|
Pretty much agree with what has been said so far. I've had this happen to me twice where we don't come to a consensus in the discussion. First time was selenium plugin vs custom scripted web scraping. Selenium got up and running faster but once my script was actually working all we needed was a for loop and we went from testing 0 websites to ~450 across 3 environments. Selenium they were still trying to figure out how to click on buttons that didn't necessarily exist. Second time it was me, I said there was no way robocopying 80k files every deployment even though only 4 of them had changed was going to be faster than checking for changes. I wrote some ridiculous script that did file checksums and only copied the new ones or some poo poo, robocopy copied all the files faster than I could check them so even best-case (no changes) it still beat me by ~2 minutes.
|
|
#
¿
Jun 4, 2015 07:01
|
|
|
Methanar posted:I've never properly appreciated how much I love cisco's cli. It's nice to be able to pipe more than once in bash. That's probably my biggest problem with cisco IOS recently - I inherited an IOS load balancer essentially with 10+ years of shoddy maintenance by previous admins, I can't show any part of the running config on the device basically because cisco regex is terrible and the config is something like ~80 pages long if you print it out (my predecessor printed it out, it's in my desk). I spend a lot of time copying configs onto TFTP servers, checking them into version control, checking them out of version control onto my machine and then finally looking at them. There are also tons of nice random things in bash like "esc .", all of screen/tmux/ratpoison, the "!!" guy, and awk/scripting languages. Even something simple like "I want this line, and the line after it" is (to my knowledge anyway) not possible without writing a fairly complex regex. Compare to "grep -A1" in bash. If I could SSH into a cisco device and get a bash terminal I would immediately automate like 80% of my job. I only ever work in CLIs these days and I can't think of one that I wouldn't prefer over cisco IOS, but yes it is still way better than interacting with any kind of gui or web interface. Also, this seems like it is turning out to be a pretty loving good first job for you so far if you don't mind me saying so. A lot of people (in my limited experience, anyway), get sort of pigeonholed into jobs where 99% of their job duties are following written instructions from someone else. At my current employer we interviewed a LOT of people for a Sr position that just could not handle "What would you do if..." questions, people from fairly big name companies who have apparently spent the last 10+ years getting paid six figures to follow a checklist 40 hours a week. Fixing stuff like you are with almost no documentation might suck really hard now, but it's going to give you a lot of really good experience and stuff to talk about during interviews when you start looking for new work.
|
|
#
¿
Jun 13, 2015 23:31
|
|
|
evol262 posted:Half the poo poo in here isn't bash. [...] "hey sepist/psydude/network person, what's the best way to do foo?" Yeah, my mistake. I'm aware that <x feature> isn't always necessarily "just bash" but, in my haste, neglected to specify "on linux servers with usability tools installed". As far as the best way to do foo, all of the things I mentioned are pretty much widely known issues. There are a lot of things to complain about with regards to cisco IOS and "its not like <linux shell>" is definitely one of them. For example, you can "more" commands instead of "show" to get additional features. The featurelist is, essentially, 3 things: you can use the ridiculous regex engine that doesn't handle multiple lines. You can 'begin' - start showing the output of the command at the first occurence of your input. You can 'include' - only show lines that match your input. You can also / - basically your forward search in Vi/Vim/Less/hatever the official name of this thing actually is. You can't ? backwards search. There's no cut/awk and then print the Xth item in this array delimited by Y character. There's no screen or similar. If you want to show specific information about things in cisco IOS you pretty much have 2 options. Show run <thing> and hope that A: this command exists, B: that your specific software version supports this command, and C: whoever arbitrarily decided the output of this command included the information that you want. Or you can do what I do and manage all of your config files on a linux server that has BSD coreutils or GNU coreutils or busybox or whatever else installed, so you can check the configs out of version control and then do your awks and seds and greps and what-have-you. If <other network guy> in this thread has any corrections or suggestions I would be very interested in hearing about them!
|
|
#
¿
Jun 14, 2015 19:36
|
|
|
The Fool posted:So, is my approach of showing the entire config, exporting my buffer, and pasting it into sublime text the wrong way to do things? It would be a little "better" if you set up a tftp server and did your show _____ | redirect tftp. Then you can check your configs into version control for easy rollbacks/diffs between versions (unless rancid is doing that for you). Given that you only touch it twice a year, I guess it's not really a big deal though! Then yeah, by all means paste your stuff into sublime text. If you don't have it already there is an excellent cisco ios syntax package available.
|
|
#
¿
Jun 14, 2015 20:27
|
|
|
Shaocaholica posted:Looking for a recommendation on a simple free remote process monitor for windows. I basically want to be alerted if a particular process on 4 remote machines is either killed or stops using CPU time. You can do this with like, 6-30 lines of powershell depending on how generous you are with spacing/formatting and how much info you actually want.
|
|
#
¿
Jun 24, 2015 22:45
|
|
|
Mrit posted:Holy poo poo. At least most sane people will have all in-band management on a separate vlan, which should prevent this from being a big issue.(disclaimer - I'm still working on my CCNA) I think most sane people aren't using any of the affected products, honestly. Who would buy a Cisco email security virtual appliance? Probably the same people that would need the device to have default root ssh keys so cisco support can configure it for them. Also, yeah, linux is generally pretty good about telling you what is wrong.
|
|
#
¿
Jun 27, 2015 05:37
|
|
|
Antioch posted:I use 'Skilled'. I feel like it sounds better. Similar vein, I actually dropped out of highschool. I leave anything education related off of my resume and, when asked about my education, I usually go into "well, college wasn't really an option for me financially..." and talk a bit about being a cook from 18 to 25. Nobody has ever actually asked me directly if I've graduated from highschool. But, yeah. It's kind of sad that we're still in an era where "being able to intelligently discuss everything under skills" puts you above 90% of the competition. We were interviewing candidates for a senior linux admin recently and the talent pool is really, really dry around here. We had one guy with C/C++ on his resume and I didn't even get a chance to ask him simple C questions because he immediately cut me off to tell me how he doesn't really know that much about programming languages. Hmm. That this guy made it past the phone screen says something about the talent pool, I guess. We ended up way overpaying a midlevel candidate for the senior position because he was the only person who didn't completely bomb every portion of the technical interview, at he was at least up front with me (and on his resume) about not being very strong with the network side of things. For all of our other candidates I did the networking portion of the technical interview and being able to describe a subnet mask was about as far as anybody got, despite having a variety of networking related terminology proudly displayed under "skills and technologies".
|
|
#
¿
Jul 3, 2015 21:55
|
|
|
goobernoodles posted:It helped me show my boss (CFO) visually what all I was working on, and how it's literally impossible for me to get all of the projects that need to get done when I'm constantly being interrupted I'm actually not a huge fan of Jira, especially for ops tasks because it tends to be pretty (really) dense. Well, at least, our implementation of it at my current employer is really unnecessarily dense and there are a lot of buttons to click, text boxes to fill, and then "submit changes" processes to wait for. It was used at a dev job that I had, and I thought it was actually pretty loving great there because we were using cloud jira with hooks to our bitbucket repo and the "denseness" of it works well when you are dealing with requests from multiple people, correspondence with non-technical people, suggested changes, code review, and then actual changes. For ops tasks, I had excellent results implementing a Trello setup at my last job, and I even use it just for my stuff at my current one. It's a kanban board, so the entire point is basically to visually represent work (especially work in progress) and it's very, very, extremely light and easy to use. You can pull up the site, make 4 tabs for Planning / In Progress / Documenting / Done and be "using trello" in about 2 minutes. You can invite a fairly significant amount of people to a trello board and I'm not sure what the pricing is like, but we ran a 5 person ops team + one direct report on it for at least 6 months prior to my departure completely free. I think a handful of features like exporting to XML or other "api-esque" things are gated behind a paywall but not much else.
|
|
#
¿
Jul 9, 2015 01:23
|
|
|
In my mind there's basically "Jr. / Entry Level" aka, we will need to teach this person what to do but they have some understanding of a technology that is critical to their job role: Networking, Linux, AD administration, etc. You don't need 2 years of experience to qualify for an AD role where you right click on GPOs, manage user accounts, and make sure people are in the right groups, you just need pen+paper at the very least and someone to show you how to do it the first time. Senior/Mid level are basically "how long is this person going to need before they can start to contribute value, and what (if any) skills are they bringing with them that we either do not already have or could use some more expertise in?". The smaller the first number is and the bigger the second number is, the more "Senior" you are. It's all relative to the position and the organization, and IMO anyway years of experience have nothing to do with it, although there is usually a pretty strong correlation there's always the chance that you end up hiring a Sr Linux Sysadmin with 15 years of experience and it turns out he's very, very good with system internals and he nailed all of your bullshit trivia questions, but that ends up only being 10% of his job. It's up to you as a department/hiring manager to know what you actually need and how to determine if candidates match your needs.
|
|
#
¿
Jul 22, 2015 15:05
|
|
|
e, fb. I generally work with smallish SaaS data centers, but in my recollection the collapsed core generally refers to using a single (paired) l3 switch that handles a bunch of vlans and some access switches and hopefully not much else. You default, static route into an asa which is hopefully externally managed so they can't possibly gently caress it up, and then the asa terminates an sslvpn which they also can't possibly gently caress up. Basically you want to keep them from touching layer 3 as much as possible because it will eventually turn into a rats nest of static, asymmetric routes. In my experience anyway.
|
|
#
¿
Jul 28, 2015 05:52
|
|
|
That's pretty much my current full time job, too, so I feel your pain. I'm working against 8 years of institutionalized misunderstanding of layer 3. No routing protocols, its a Web service so the admin team thinks they know what they're doing but really it's a mess of virtual machines where the VM hosts don't participate in STP, and some poor bastard from before my time was coerced into intentionally implementing asymmetric routing for about 1/2 the network. My first day on the job I checked the logs for one of the vmhost switches and found 2.5 years straight of warnings that all of the VMs on one link were mac flapping between the other link for that host. I guess that's just fine. That's what we do here -- gently caress layer 3 we're still working on layer 2. You can look at the mac table on the core switches and watch it completely change every 20 seconds. Tracing a layer 2 path is something of a probability equation in our "cloud LAN". It's really dangerous to be smart enough to make changes but not smart enough to know what you're doing. E: of course I ran into the classic "directly connected route has lower AD than configured static route and everyone is confused". Really just poor decision making all around -- I'm going to start asking to see configs before I accept employment offers. 12 rats tied together fucked around with this message at 06:16 on Jul 28, 2015 |
|
#
¿
Jul 28, 2015 06:14
|
|
|
|
| # ¿ Apr 16, 2024 07:49 |
|
|
Methanar posted:I want to be able to say I've done this huge list of things over the summer so I can be qualified for a big boy job when I'm finished school. Honestly you can probably just lie and say you actually did it. You've done all of the hard work already and it's super impressive (to me anyway) that you, as a junior something-or-other? understand the concept of domain functional levels, migrations, replication and especially all of the gotchas involved in splitting a server apart. You're pretty much already qualified. I've worked with people in windows sysadmin roles who've been working in IT for 5 years plus who would struggle to even explain what a domain controller is except for some amalgamation of "its important". Maybe the bar is higher in whatever non-US country you are in, but I kind of doubt it. I actually know a couple dudes right now in the US who would hire you on the spot for a Junior position if you even mentioned the phrase 'functional level' in an interview.
|
|
#
¿
Jul 29, 2015 15:07
|
|