FreeBSD ate my RAM

(crocidb.com)

75 points | by theanonymousone 4 hours ago

11 comments

  • tiffanyh 1 hour ago
    If you like this kind of post, you might like this “htop explained” post.

    https://peteris.rocks/blog/htop/

  • drdexebtjl 39 minutes ago
    I don’t understand the part about using heuristics and deciding what counts as used memory…

    Used memory for the system is always total minus available.

    Heuristics? I would hope that the system knows precisely what is using every single byte of physical and virtual memory. Is this a reporting problem? Why do we have to settle for heuristics and not the exact number?

    • toast0 11 minutes ago
      > I would hope that the system knows precisely what is using every single byte of physical and virtual memory.

      Of course the system knows what is using every page. The difficulty is really in how to account for pages that are backed by disk.

      If you count all of those as free, that's not accurate. If you count all of those as used, that's not accurate either. Additionally, FreeBSD (at least) doesn't have separate queues for disk backed pages, so there's not really a good way to know how much of your active (or inactive) memory is disk backed.

      As an additional caveat that measuring active/inactive has costs. In the past, FreeBSD wouldn't really do the work for that until it needed to... I know some stuff changed, but I don't remember where it ended up; it wasn't great when it bulk marked a ton of pages as inactive and then the active ones would fault back in.

    • CrociDB 22 minutes ago
      The thing is it's easy to define free, unused memory. But a lot of the used memory is your system caching stuff that would be free if you needed more than what's actually free. So you can see you have 1g of free memory out of your 4g, but then you allocate 3g and it will do without a sweat and you'd be confused. So you have to go and dig for what those caches are and report that they're effectively free too.
    • man8alexd 15 minutes ago
      You will be surprised by how inaccurate memory measurements are.
  • duendefm 2 hours ago
    Thank you for such a quality post.
  • efxhoy 56 minutes ago
    Great job on getting the fixes merged!
  • m463 2 hours ago
    the end struck me - a picture of an os book. I wonder if students these days retain their books after college, or do they get returned as a rental?
    • linguae 2 hours ago
      I'm a professor at a community college in Silicon Valley, and my students use online textbooks. I try to use Creative Commons or other libre textbooks, but sometimes I use paid textbooks when they are heads-and-shoulders better than their libre alternatives. Some e-textbooks can be accessed on a subscription basis. I admit I prefer non-subscription materials, but a colleague advised me that often the book that students learn from is different from a good reference book that students can use once they've already learned the material. For example, my colleagues and I have had great success with an online, interactive textbook for discrete math. While the subscription is unfortunately only valid for the duration of the course, once students have learned discrete math, they could buy a used copy of Rosen's discrete math textbook as a reference.

      The nice thing about e-textbooks is not needing to carry around a bunch of heavy books. I remember the tomes I had in my college days, such as Stewart's Calculus.

      • NooneAtAll3 1 hour ago
        just hint students towards anna's archive and then sky's the limit
    • post-it 2 hours ago
      I bought as few textbooks as I could, but the few that I did buy are sitting in my parents' basement bookshelves somewhere.
  • jmclnx 3 hours ago
    Interesting post, it made me wonder. At one time FreeBSD swap usage/logic was far better than what Linux did. Is that still the case ?
    • man8alexd 1 hour ago
      FreeBSD didn’t have memory overcommit and instead used strict swap reservation - each allocated anonymous memory page was supposed to have a corresponding swap page. This required 2x RAM swap space, otherwise you would get “out of swap” when forking a large process. FreeBSD implemented memory overcommit around 2000.
    • 0x457 1 hour ago
      Yes, It's just not every tool is aware of ZFS ARC. Which is what this post is about. Author just describes in an odd way.
    • shevy-java 2 hours ago
      I remember how NetBSD promoted itself as running on many more toasters than Linux once.

      Then some NetBSD dev wrote on their mailing list that this is no longer true. Linux runs on more toasters now. (And also top 500 supercomputers, but toasters are the real metal to the petal test.)

      These fights always remind me of:

      https://www.jwz.org/doc/worse-is-better.html

      It's an interesting piece of history too. I kind of evaluate it a bit differently, e. g. my summary is "momentum beats academic perfection". Which is not completely what it is about, but it is my own imperfect TL;DR summary.

      • bee_rider 1 hour ago
        This basically fits my stereotype of BSD being a little bit more hardcore while Linux is a little more accessible… when the question was “can you install an OS on a toaster,” BSD had an advantage. Now that normal engineers have to make IOT toasters (for some reason) Linux should have the advantage, right?
        • sublinear 1 hour ago
          Normal engineers don't do that either.
      • Levitating 1 hour ago
        why did that url point me to a scrotum in an egg cup
        • UnlockedSecrets 1 hour ago
          Seems like at a glance that website if it sees a referral from ycombinator, it redirects to that image.... In a private window it loads the 'intended' page.....
  • tom2ow 1 hour ago
    [flagged]
  • tomeow 2 hours ago
    [flagged]
  • hnloser 2 hours ago
    [flagged]
  • naturalmovement 2 hours ago
    ZFS cache. The end.

    User installs an unfamiliar server OS with an enterprise filesystem and is stunned when it works differently. I fail to see a teachable moment here.

    • craftkiller 51 minutes ago
      I've noticed that the various tools report different amounts of free memory. I appreciate finally having an explanation for why.
    • toast0 2 hours ago
      Sure, but also some tools needed fixing.
  • shevy-java 2 hours ago
    This is why I use Linux. :>

    Poor FreeBSD folks though. After so many years trying to present themselves as better alternative, the road just got steeper ...

    • edoceo 1 hour ago
      The OS Crusades are over man
    • vermaden 58 minutes ago
      No ZFS no problem :)