• thisbenzingring@lemmy.sdf.org
    link
    fedilink
    English
    arrow-up
    0
    ·
    4 days ago

    in today’s world I think you have to get in as a Windows System admin and then change your role by demonstrating how the Linux solution would be better for whatever the priority task at hand is

    • Honytawk@lemmy.zip
      link
      fedilink
      English
      arrow-up
      0
      ·
      edit-2
      4 days ago

      And then you get blamed when it breaks, so they switch back to what they know and use a ready-made solution from Microsoft with full 24/7 support package.

      Because it is more important for it to work reliably and be fixed quickly, than for it to not be proprietary or not made by some evil corporation.

      Downtime costs money.

      I know Linux can be extremely reliable, but they rarely have all the bells and whistles that Microsoft offers and are rarely compatible with all the other software they are already using.

      Having multiple different systems overlapping also adds overhead.

      So I doubt you will be able to convince many like that. It is easy looking at it from a tech perspective, but they don’t make those decisions.

      • Lettuce eat lettuce@lemmy.ml
        link
        fedilink
        arrow-up
        0
        ·
        4 days ago

        You know what’s funny about that? I can think of at 4 times in just the last year where a Microsoft outage caused significant downtime (at least 1 hour) for the company I work at.

        • Twice, Outlook/Teams was having major regional issues for hours, people at my company couldn’t log into Teams and weren’t getting emails.
        • Microsoft’s Dynamics platform, (which my company’s ERP software is built on) had some infrastructure issue that made it unusably slow for several hours.
        • Who can forget the lovely Crowd Strike kiss of death fiasco a few months ago?

        Meanwhile, the 12 year old janky Debian servers I had were running Ansible, Docker, OpenProject, and several other services without a hitch, same with all the Linux endpoints I had deployed.

        Centralization causes many of these problems and makes them more severe than they otherwise would be. When you are locked into a single vendor for everything you do, you’re completely at their mercy if anything breaks.

        The problem is that nobody, at least in the US, markets open source solutions. The big players corner the market, and IT just learn those big players. You should see the looks I’ve been given when I present IT directors with a quote from ix Systems for a TrueNAS solution to their storage needs. They have no idea who they are, even though they provide enterprise grade storage solutions at a fraction of the price of Dell or HP.

        The US tech environment is a cyber dystopia controlled by the Tech corpos of silicon valley. It’s so frustrating.

    • Korhaka@sopuli.xyz
      link
      fedilink
      English
      arrow-up
      0
      ·
      4 days ago

      Not even sure of my chances at that these days. There are very few opportunities where I live too.

      Got a few Linux VMs running on my work laptop for testing stuff though and literally no one else in the support team knows how to test what I use them for. It’s a fairly rare task that no proper process exists for. My thinking is to see if I can either roll out VMs to people or better but less likely get my own virtual server.

    • somenonewho@feddit.org
      link
      fedilink
      arrow-up
      0
      ·
      4 days ago

      Guess I was lucky. I got an apprenticeship as a Sysadmin and was told I’d mostly be in the in-house/client IT (Windows) but since I had a lot of Linux experience already, on my first day they put me into the server/hosting department (Linux) where I spent most of my apprenticeship.

      But yeah looking for a job now, most companies seem to look for “good experience with windows/Windows server required” and “experience with Linux a bonus”.

  • Possibly linux@lemmy.zip
    link
    fedilink
    English
    arrow-up
    0
    ·
    5 days ago

    The job of a old style Linux admin has been superseded by infrastructure engineering and GitOps. You write automation to deploy and maintain thousands of VMs.

      • derbolle@lemmy.world
        link
        fedilink
        arrow-up
        0
        ·
        4 days ago

        actually, yes. you test solutions „manually“ on servers and then roll them out with ansible across the whole infrastructure, neatly kept and carefully maintained in your local git instance. like raking a zen garden sometimes

    • borf@lemmynsfw.com
      link
      fedilink
      English
      arrow-up
      0
      ·
      5 days ago

      Some of us were doing infrastructure as code back when they still called us “Linux admins” and we were doing “gitops” with Subversion and Puppet. My title has changed a lot since the old days but some of the “new hotness” was around in the old days with different names. Fuck, look at LPARs.