I’m curious if whonix actually provides any value rather than just running tor locally?

Like has there ever been a case when because tor was run locally on someones computer over using whonix for tor that they were compromised?

  • ghost_laptop@lemmy.ml
    link
    fedilink
    arrow-up
    1
    ·
    3 years ago

    You need a CRAZY amount of RAM, not just more RAM. At least having 128GB RAM as a bare minimum is not something I’d ever seen in my third world country, even for people with moderately good PCs. The price for that amount of RAM where I live is 972.96USD.

    • AgreeableLandscape@lemmy.ml
      link
      fedilink
      arrow-up
      1
      ·
      edit-2
      3 years ago

      I’m assuming this is because Qubes OS allocates RAM independently for each VM, regardless of how much the VM is actually using? Does Xen not have the ability to use a pool of RAM and just overlap the empty space? I was under the assumption that enterprise level hypervisors could do that. Even if you have five VMs, if four of them are idle, only a little more than one OS’s worth of RAM should be actually filled with active data.

      • ghost_laptop@lemmy.ml
        link
        fedilink
        arrow-up
        2
        ·
        3 years ago

        Qubes does not only use compartmentalization for stuff like browsing, if I remember correctly your /home folder is also a separate VM and I think there was something similar to this happening too, so I assume there’s always at least like 3 or 4 VMs running. Either way AFAIK you still need that amount, regardless of if there’s an alternative, which makes it non usable for poor people.

        • AgreeableLandscape@lemmy.ml
          link
          fedilink
          arrow-up
          1
          ·
          edit-2
          3 years ago

          Fair enough. I said this in another comment already, but for this reason I really wish there was a viable desktop OS that uses a microkernel, since such an OS can be much more easily designed do secure compartmentalization without the need for VMs or duplicate parallel processes in general. Micorkernels still have more computational overhead than one monolithic kernel, but would be much more resource frugal than multiple VMs.