>"What you're really formating is "123 thousand bytes", which in short form should probably have a space: "123K B". You're not formating "123 kilobytes"

this is silly. SI is the standard. having an API to do tech munged KB = kibibytes is fine, but normal SI units shouldn't add a space.

if you're mocking it, it isn't an integration test. mocks are for unit tests.

[fs/ext4/super.c:735]

if (test_opt(sb, ERRORS_PANIC) && !system_going_down()) {
panic("EXT4-fs (device %s): panic forced after error\n",
sb->s_id);
}

looks like it's still in there. seems like it would be reasonable to issue a warning before mounting it. an annoyance, but no danger.

>they learned to love Ruby on Rails

ah, yes. stockholm syndrome

interesting.hadn't seen the use of ephemeral addresses for privacy before.that's a solid idea. I see dhcp assigns a range to the device instead of a specific address to account for this sort of thing

till my ISP foregoes IPv4 completely, I'll keep my NAT security blanket, but thanks for the pointer

it would be about five minutes after the "no NAT revolution" that ad companies and others would start selling lists of active IP addresses. buy one or get one from a shady site and then scan the blocks around them for active hosts.

a firewall adds latency and isn't free as well.

so very sorry you can't port scan my internal networks. hopefully, you'll find a way to come to peace with this in time

the only thing that NAT requires is two brain cells to poke a hole or for application developers to have a UDP rendezvous server available to help bridge the double NAT

no biggie

did some basket weaving in the scouts as a kid, so this phrase always amused me. oh, so you're going to just start weaving with a bunch of dry reeds that snap every time you try to twist them around or pull them in the slightest funny way trying to tuck an end or something? those reeds need to soak