gchristensen changed the topic of #nixos-chat to: NixOS but much less topical || https://logs.nix.samueldr.com/nixos-chat
<lovesegfault> Hm, my wireless headphones are finally starting to die
<lovesegfault> after 4 years
<lovesegfault> now I can't decide whether to invest in a nice desk hifi or a new pair of ANC wireless
<lovesegfault> in a normal world I do need ANC b/c of flights, subway, office, etc
<lovesegfault> should I plan for a normal world?
Dotz0cat_ has joined #nixos-chat
Dotz0cat has quit [Ping timeout: 240 seconds]
supersandro2000 has quit [Disconnected by services]
supersandro2000 has joined #nixos-chat
<infinisil> >> Trump reportedly suggested of fortifying the wall with a water-filled trench inhabited by snakes or alligators, and electric fencing topped with spikes that can pierce human flesh.
<infinisil> lmao
<infinisil> I know we're over this person, but this is hilarious
rajivr has joined #nixos-chat
<bbigras> haha
tilpner_ has joined #nixos-chat
cole-h has quit [Ping timeout: 240 seconds]
tilpner has quit [Ping timeout: 256 seconds]
tilpner_ is now known as tilpner
<ashkitten> how does he think the snakes or alligators are going to get fed???
<bbigras> with the victims maybe
<bbigras> but good point
<samueldr> natures... finds a way
<aleph-> So we feed them Jeff Goldblums
<aleph-> Got it
slack1256 has joined #nixos-chat
BaughnLogBot has quit [Quit: ZNC 1.6.2+deb1 - http://znc.in]
BaughnLogBot has joined #nixos-chat
BaughnLogBot has quit [Client Quit]
BaughnLogBot has joined #nixos-chat
BaughnLogBot has quit [Client Quit]
ky0ko has joined #nixos-chat
slack1256 has quit [Ping timeout: 265 seconds]
<hexa-> i'm well aware the issue has been known for a while now, but there's lots of feedback here
<hexa-> (swapping on m1)
<samueldr> nice how this basically means the whole motherboard has to be switched when it fails
<hexa-> my desktop ssd with nixos after ~2.5y
<hexa-> Data Units Read: 204,700,066 [104 TB]
<hexa-> Data Units Written: 124,403,415 [63.6 TB]
<hexa-> that will be >10y before I reach the max tbw (300TB)
<aleph-> heh gotta love SSD's
<jtojnar> ugh, why are the smart attributes not standardized
<hexa-> they look better on nvme disks
<lovesegfault> Mine seems to look quite good?
<lovesegfault> Percentage Used: 2%
<lovesegfault> Data Units Read: 88,762,742 [45.4 TB]
<lovesegfault> Data Units Written: 54,503,300 [27.9 TB]
<lovesegfault> this is ~2yo I think
<supersandro2000> How do I get this data?
<lovesegfault> smartctl -a /dev/yourDrive
<supersandro2000> I don't have data units there
<lovesegfault> huh
<supersandro2000> needed to calc that by hand
<supersandro2000> feel like living in a cave
<supersandro2000> 104 TB written in 2y power on
<lovesegfault> ,locate secret-tool
<{^_^}> Found in packages: gnome3.libsecret
<samueldr> uh, not so secret is it?
<lovesegfault> :D
<aleph-> haha
<elvishjerricco> hexa-: There was a reddit thread about that macbook ssd thing, and someone linked a podcast where a guy with a 2017 intel macbook reported the same kind of ratio of writes to power on hours.
<elvishjerricco> So it seems like an issue that's probably been around for a long time.
<elvishjerricco> Apple's always been very confident with their ssds though; to the point that they don't checksum user data in APFS (but they do checksum metadata, curiously).
<elvishjerricco> And to their credit, I've never heard that Apple's SSDs fail unusually often.
<aleph-> Anyone have opinions on verizon fios?
<ashkitten> it's fine
<aleph-> They do any weird price switching after your first year?
<lovesegfault> Not available where I am, apparently
<lovesegfault> I use AT&T and they suck ass
<lovesegfault> I hate them
<ashkitten> i had fios but i wasn't the owner of the account so couldn't say
<aleph-> Nod nod.
<aleph-> Yeah thinking of switching to them from optimum
<aleph-> $67/m for 200Mb down currently. Where as I can pay $80/m @ verizon for gigabit.
<lovesegfault> I have gigabit and was disappointed
<ashkitten> for the first year of me living here our only options were comcast and frontier, and frontier is even worse
<aleph-> lovesegfault: How so?
<ashkitten> but now we've got gigabit and i love it
<lovesegfault> aleph-: My home stuff is mostly on wifi, so the speed increase from 500Mb wasn't all that noticeable
<aleph-> Ah got it.
<aleph-> Yeah this would be killer for my torrent box and NAS.
<lovesegfault> only now that I started wiring stuff and got a proper switch etc it's starting to make any difference at all
<ashkitten> oh well if you already have symmetric it doesn't make that much of a difference
<ashkitten> for me i went from 300/15 to 1000/1000
<ashkitten> huuuge difference
<aleph-> Heck just doing some the initial backup of my NAS looks to be 60 days continuous currently. With gigabit it'd be roughly 15~ hours
ky0ko has quit [Remote host closed the connection]
Dotz0cat_ has quit [Ping timeout: 256 seconds]
Dotz0cat has joined #nixos-chat
endformationage has quit [Quit: WeeChat 2.9]
waleee-cl has quit [Quit: Connection closed for inactivity]
aminechikhaoui1 has joined #nixos-chat
aminechikhaoui has quit [Ping timeout: 256 seconds]
aminechikhaoui1 is now known as aminechikhaoui
cole-h has joined #nixos-chat
tilpner_ has joined #nixos-chat
tilpner has quit [Ping timeout: 264 seconds]
tilpner_ is now known as tilpner
lukegb has quit [Quit: ~~lukegb out~~]
lukegb has joined #nixos-chat
krkini has joined #nixos-chat
kini has quit [Ping timeout: 260 seconds]
krkini has quit [Quit: bye]
<siraben> How do I run a linux binary in Docker on macOS?
kini has joined #nixos-chat
Gaelan_ is now known as Gaelan
<AMG> hmm is there a ftp command way to see if a folder is empty?
<lunc> You can ls in ftp
<lunc> yeah but that doesnt show if the folder is empty? or does it?
<lunc> hmm
<AMG> ah don't believe that it does
<AMG> drwxr-xr-x 2 14826 501 4096 Feb 22 1999 ocean
<AMG> drwxr-xr-x 8 5460 501 4096 Jun 8 2016 oceanowl
<AMG> the last one contains folders, the first one doesnt
<AMG> can the second parm be number of dirs? hmm
<lunc> hmm not really sure. Haven't used ftp in ages, kinda forgot
<AMG> we agree a ftp with anonymous access is fair game right?
<AMG> i can snoop around here, if i dont hammer the ftp i should be good
<lunc> If it is public it is public ;)
<AMG> .. besides the 46 gb D&D books im already mooching
rajivr has quit [Quit: Connection closed for inactivity]
__monty__ has joined #nixos-chat
spudly has quit [Quit: ZNC - http://znc.in]
<etu> I've just started pulseeffects while in a video call for the first time because there was some clicking noise from something, now I can't do videocalls without pulseeffects.
<__monty__> Please stop tempting us. This is a rabbit hole I'd prefer not to fall down...
<etu> I didn't even had to tweak anything in it to make it worth it.
<cole-h> amg: That "second param being number of dirs" sounds like it could be right
<cole-h> I'd guess the 2 in the first one comes from `.` and `..`
cole-h has quit [Ping timeout: 256 seconds]
Dotz0cat has quit [Ping timeout: 256 seconds]
PhantomFutaba has joined #nixos-chat
spudly has joined #nixos-chat
tilpner_ has joined #nixos-chat
tilpner has quit [Ping timeout: 246 seconds]
tilpner_ is now known as tilpner
<ehmry> are there any languages that are slower to compile than rust?
<eyJhb> Ehmm.... Microsoft Teams really does suck. I cannot sign out from my coorporate e-mail, and sign in with my student one...
<eyJhb> And deleting ~/.config/teams does not work
srk has quit [Remote host closed the connection]
srk has joined #nixos-chat
<eyJhb> WELL! So, restarting works, as it nuked all files. Thank god for rollback on boot <3
<etu> :D
<eyJhb> (and also, fuck you Microsoft Teams for having some magic file some place on my system :( )
spudly has quit [Quit: ZNC - http://znc.in]
kini has quit [Remote host closed the connection]
kini has joined #nixos-chat
<siraben> ehmry: er, Haskell maybe? (sad to say as a Haskll fan myself)
<siraben> Most of the time I interpret my Haskell code though :P
rajivr has joined #nixos-chat
<infinisil> ehmry: In the past at least, Swift had some real problems with compile speed
<infinisil> E.g. binary expressions took exponential time depending on the amount of operations used
PhantomFutaba has quit [Ping timeout: 272 seconds]
<infinisil> Not sure how good it is nowadays
spudly has joined #nixos-chat
spudly has joined #nixos-chat
<siraben> fun fact about Hindley-Milner type inference: it can take worst case exponential time to infer a type
bqv is now known as bao
bao is now known as bqv
spudly has quit [Ping timeout: 240 seconds]
spudly has joined #nixos-chat
nckx is now known as jorts
Dotz0cat has joined #nixos-chat
Dotz0cat has quit [Ping timeout: 240 seconds]
Dotz0cat has joined #nixos-chat
aleph- has quit [Ping timeout: 240 seconds]
aleph- has joined #nixos-chat
rj has joined #nixos-chat
jorts is now known as nckx
thibm has quit [Quit: WeeChat 2.9]
tilpner has quit [Remote host closed the connection]
tilpner has joined #nixos-chat
waleee-cl has joined #nixos-chat
<makefu> ,locate bin vi
<{^_^}> Found in packages: vim, busybox, vimHugeX
<Mic92> My new project: https://github.com/Mic92/vmsh
<Mic92> Not much too see yet
<bbigras> debug container?
<bbigras> any usecase example?
<Mic92> I renamed now from debug container might be to confusing
<bbigras> Mic92: I don't have the rights to see the slides. Should I hit the request button or do you want to change something on your side?
<Mic92> bbigras: fixed
<bbigras> Mic92: it works. thanks
<Mic92> bbigras: I recently read a paper about the authentication service google uses for link permissions like this
endformationage has joined #nixos-chat
<Mic92> They call it zanzibar
<Mic92> before vmsh I built something similar for containers: https://github.com/Mic92/cntr
<bbigras> you made cntr? I used this before
<bbigras> let see if I created an issue to ask a dumb question
<Mic92> It's part of my phd thesis
<bbigras> I only created https://github.com/Mic92/cntr/issues/11 . not a stupid question
<bbigras> nice
<{^_^}> Mic92/cntr#11 (by bbigras, 1 year ago, closed): the demo in the README doesn't work anymore
<Mic92> It's quite cool. I can dive into low-level programming. I read about kernel exploitation the last few days.
<bbigras> I'm only starting reading the slides but is it a bit like firecracker?
<bbigras> oh. containers are not lightweight. not the vms
<Mic92> bbigras: I want to use it for firecracker as well.
<bbigras> sweet
<tilpner> I wanted to stay on zfs 0.8 specifically, putting off the upgrade to 2.0 until I had time
<Mic92> firecracker still cannot be built with nix yet.
<tilpner> And now it snuck with yesterdays stable update
<Mic92> I need to fix musl rust for it.
<tilpner> And there it is, pool corruption on a day I can absolutely not afford it :(
<Mic92> important files got lost?
<tilpner> Not sure yet, scrub is stuck, zfs panicked
<gchristensen> did you reboot?
<tilpner> Not yet
<gchristensen> so you're running 0.8 in the kernel and 2.0 in userspace?
<tilpner> But all IO on my main disk is stuck
<tilpner> Oh, yes, I reboot after all updates
<tilpner> 5.10.15 zfs-2.0.2-1
<gchristensen> okay, so you have 2 in the kernel and in the userspace
<tilpner> zfs-kmod-2.0.2-1
<tilpner> I think that's a yes
<tilpner> I got scared when I saw https://github.com/openzfs/zfs/issues/11559 , but it mentions rt kernel, let's hope it's not that
<{^_^}> openzfs/zfs#11559 (by rugubara, 2 weeks ago, open): zfs-2.0.1 on 5.10.1-rt-rt20 leads to pool corruption
<gchristensen> you don't have an rt do you? :P
<tilpner> I certainly didn't enable anything like that, standard linuxPackages_5_10
<gchristensen> that is spooky, tilpner :/
<tilpner> Yes, and I checked every other update, that I was still using 0.8!
<gchristensen> afaik there were no major changes at all between 0.8 and 2.0.0, if that is any consolation
<tilpner> From my dmesg output: Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./B450 Pro4, BIOS P3.60 07/31/2019
<tilpner> (unrelated, but fun-sad)
<aleph-> Reminds me I need to move my NAS from debian to nixOS finally
<aleph-> That should be an adventure.
<aleph-> Especially since all zfs utils are compiled by myself.
<aleph-> Hopefully there's no mismatch heh
<gchristensen> tilpner: off to #Zfsonlinux you go? maybe they can elicit more info
<tilpner> I don't have time to debug this, I have a deadline tomorrow and work to do
<tilpner> Just backing up unsynced progress and rebooting to see the damage
<aleph-> tilpner: Ouchhh
<tilpner> (I do have time to look into it, but not to keep the system running indefinitely, waiting for people to guide me towards more interesting debug details)
<gchristensen> it is *probably* fine ...
<tilpner> Well, the syncoid job has been failing all day
<tilpner> Which should have been alarming, but it has done that in the past and recovered from it. And no time, of course
tilpner has quit [Quit: tilpner]
<gchristensen> extreme *gulp* energy https://github.com/openzfs/zfs/issues/11443
<{^_^}> openzfs/zfs#11443 (by shuther, 5 weeks ago, open): silent corruption for thousands files gives input/output error but cannot be detected with scrub - at least for openzfs 2.0.0
lunc has quit [Ping timeout: 240 seconds]
<srk> 'Feb 10 12:21:15 ryz kernel: Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./B450 Gaming-ITX/a>'
<srk> :D
* gchristensen is concerned by the fact that tilpner hasn't joined again
<infinisil> > timeTo mars
<{^_^}> "2 days, 3 hours, 5 seconds"
slack1256 has joined #nixos-chat
tilpner has joined #nixos-chat
<tilpner> gchristensen: It seems you were right
<gchristensen> oh?
<tilpner> stage1 hung for a minute while mounting, the scrub restarted from the beginning but kept the old start timestamp
<gchristensen> hm
<tilpner> Just scrubbed 1.4T in 28min, nothing repaired, damage log didn't change
<gchristensen> phew
<gchristensen> I found something for you to know about while you were gone. it brings me no pleasure to point it out
<gchristensen> extreme *gulp* energy https://github.com/openzfs/zfs/issues/11443
<{^_^}> openzfs/zfs#11443 (by shuther, 5 weeks ago, open): silent corruption for thousands files gives input/output error but cannot be detected with scrub - at least for openzfs 2.0.0
<tilpner> Permanent error in fpool/main/var/lib@autosnap_2021-02-15_17:30:02_frequently:<0x0>
<gchristensen> yup.
<gchristensen> that is how I found the issue
<tilpner> Well, that's terrifying
<tilpner> This SSD has no business dying on me, 42 power cycles (18 unsafe), 100% spares available, 0 smart "Media and Data Integrity Errors"
<tilpner> But if it's a bug in zfs... my shiny ssd doesn't help at all
srk has quit [Remote host closed the connection]
srk has joined #nixos-chat
<__monty__> This is why I'm rooting for Btrfs.
<__monty__> Monoculture's never good.
<gchristensen> I don't know how I feel about my motherboard's BMC runnig PAM
rj has quit [Ping timeout: 268 seconds]
* tilpner is painfully reminded of the weakpoint in his backup strategy: zfs
<tilpner> Back to 0.8 after this scrub finishes
<gchristensen> do you have any reason to think 0.8 is going to not have this problem?
rj has joined #nixos-chat
<gchristensen> it isn't like 2.0.0 was a big release, it is 2.0.0 due to the organizational change of becoming openzfs's primary zfs
<tilpner> Not really, but I have 2 data points with 2.0 panicking, and none with 0.8, and that's good enough for right now
<tilpner> Also 5.10 -> 5.9
rajivr has quit [Quit: Connection closed for inactivity]
lunc has joined #nixos-chat
tilpner has quit [Quit: tilpner]
tilpner has joined #nixos-chat
rj has quit [Ping timeout: 268 seconds]
rj has joined #nixos-chat
cole-h has joined #nixos-chat
srk has quit [Remote host closed the connection]
srk has joined #nixos-chat
<gchristensen> I guess I need a better computer: Feb 16 14:33:30 Petunia gnome-shell[5164]: libinput error: event257 - Kinesis Advantage2 Keyboard: client bug: event processing lagging behind by 21ms, your system is too slow
<samueldr> gchristensen: nah, I think libinput is just solidly impatient
Dotz0cat has quit [Ping timeout: 256 seconds]
Dotz0cat_ has joined #nixos-chat
rj has quit [Ping timeout: 268 seconds]
rj has joined #nixos-chat
<samueldr> for gnome... I'm using libinput directly for the mobile nixos early GUI things
<samueldr> and threading wouldn't gain any real advantage AFAICT
<samueldr> so I'm left with loads of spam on stdout about that :)
<gchristensen> sssh I'm working on justification here
<samueldr> your computer is plenty fine
<gchristensen> FINE.
<cole-h> lovesegfault: *requests my re-review* *merges before I can re-review* lol
<lovesegfault> cole-h: sorry, I always forget how to dismiss a stale review
<lovesegfault> :P
<cole-h> hehe :D
<elvishjerricco> gchristensen, tilpner: I thought that "silent corruption" issue was specific to TrueNAS because they merged some patches that openzfs hadn't officially released
<gchristensen> ah
<elvishjerricco> Oh but that comment you linked does specify NixOS on Linux 5.4
<gchristensen> oh wow
<elvishjerricco> So maybe that was an old suspicion that's been debunked or something
<tilpner> If you read on further, it kept on happening for rbrewer123 with 2.0 and NixOS
<tilpner> Though I think it's "only" loud corruption, not silent, for rbrewer
<tilpner> (and me)
<tilpner> (Not that I will ever know for certain if all my files survived)
evanjs has quit [Read error: Connection reset by peer]
evanjs has joined #nixos-chat
<infinisil> I wonder, for the mars landing stream, what are we actually going to see?
<infinisil> I'm kind of doubting there will be a video feed
<infinisil> Soo, just audio and some graphs? Or something else?
<infinisil> > timeTo mars
<{^_^}> "1 day, 22 hours, 8 minutes, 33 seconds"
<bbigras> infinisil: did you check the last one? I mean the last crazy one with the sky crane
<bbigras> I did.
<infinisil> The last what?
<bbigras> landing
<bbigras> with the crazy sky crane
<bbigras> Curiosity
<samueldr> I would guess 3D simulations coming from the data they get
<infinisil> I haven't
<bbigras> I watched Curiosity's landing at like 1 am. it's like people staring at screen and saying everything is ok or not.
<infinisil> bbigras: What was that, also a 3D animation?
<infinisil> Ah
<bbigras> there's was a 3d simulation on the computer we could follow in real time but there was also the real time feed in the control room
<infinisil> samueldr: Doing a 3D simulation live seems pretty hard
<bbigras> the big thing was seeing the first pictures for the rover
<samueldr> infinisil: not simulation, but you get data, plug them into unity
<infinisil> bbigras: And how long did that take?
<bbigras> but I think there was also signal for "it didn't crash"
<samueldr> or something cheap like that
<bbigras> hmm. there's the 7 mintutes or whatever delay but I think it took a while. like 30 mins or more to get the first picture. not sure
<samueldr> ignore the delay
<bbigras> maybe you can still watch the feed
<samueldr> they can't get the data any faster than that
<samueldr> so for all intents and purposes, "accounting" for the delay is not useful
<bbigras> I mean I'm guessing the first picture delay is not just for the radio delay.
<samueldr> ah
<samueldr> yeah, ignore only the 7 minute delay for their telemetry
slack1256 has quit [Ping timeout: 264 seconds]
<bbigras> > The first images taken by NASA's Curiosity rover moments after landing on Mars.
<{^_^}> error: syntax error, unexpected ')', expecting ID or OR_KW or DOLLAR_CURLY or '"', at (string):476:1
<bbigras> yeah computer simulation
<bbigras> I guess it must have taken a while to get that high def landing video.
<bbigras> ^ that eyes thing is a simulation thing
<infinisil> Oh nice
<bbigras> I'm not 100% sure but I think we were able to set it in real time
<bbigras> the first curiosity picture was a shitty grayscale one
<samueldr> you're demanding!
<bbigras> me? I'm just saying it was from a collision detector camera or something.
<infinisil> Just watched that eyes.nasa.gov simulation
<bbigras> I'm guessing you guys saw: https://www.youtube.com/watch?v=Esj5juUzhpU
<infinisil> It apparently inspects the terrain before it even touches down so it can find a good place to land
<bbigras> nice. I hope it will all go well. the crane thing blows my mind.
<infinisil> Yes
<infinisil> Same
<infinisil> The number of stages is amazing
<bbigras> ^ a live feed of something
<bbigras> some woman was speaking
<bbigras> ah, maybe it just ended
<infinisil> That is, something
<bbigras> 2026 mars sample return. nice
<infinisil> Oh right there's that too
<infinisil> Damn
rj has quit [Quit: rj]
LnL- is now known as LnL
evanjs has quit [Ping timeout: 240 seconds]
evanjs has joined #nixos-chat
rj has joined #nixos-chat
<bbigras> "7 minutes of terror". I guess the landing takes 7 mins.
rj has quit [Client Quit]
rj has joined #nixos-chat
rj has quit [Remote host closed the connection]
rj has joined #nixos-chat
<infinisil> Cool mars fact: It can get 35° C (95 °F) hot
<bbigras> uh. it can get a lot cold, right?
<infinisil> Indeed, check https://en.wikipedia.org/wiki/Mars for some more numbers
<infinisil> -143° C low, -63° C mean
<samueldr> cool
<infinisil> There's a surprisingly big overlap with temparatures on earth
<samueldr> another nice fact, since it takes 687 earth days to go around the sun, you age slower
<bbigras> just the age number right? you'll still die as fast
<infinisil> Oh that's a nice infographic of the temparature overlap: https://mars.nasa.gov/files/mep/facts/Temperature-Mars-Facts.jpg
<samueldr> bbigras: I wasn't being serious
<infinisil> xD
<bbigras> haha
<infinisil> It's kind of sad
<infinisil> Even if we were to travel near the speed of light, we'd still live just as long
<infinisil> In our own timeframe at least
<infinisil> If you only had 1 hour to live, but wanted to meet your grand grand grand children before you die, you could (if you travel fast enough)
<infinisil> But you'd still only have 1 hour to live
<infinisil> You could outlive literally every other human being within an hour
<bbigras> It's humbling to see the scale of the universe and how much it doesn't give a fuck about us.
<infinisil> Yet here we are :o
<infinisil> Somehow alive
<bbigras> Are we? /s
rj has quit [Quit: rj]
rj has joined #nixos-chat
__monty__ has quit [Quit: leaving]
tilpner_ has joined #nixos-chat
tilpner has quit [Ping timeout: 240 seconds]
tilpner_ is now known as tilpner