tilpner has quit [Remote host closed the connection]
tilpner has joined #nixos-dev
AlwaysLivid has quit [Ping timeout: 244 seconds]
teto has quit [Ping timeout: 244 seconds]
<abathur>
anyone know how the various platform-specific Nix installer tarballs get built and deployed? I think I'm getting close-ish on basic macos big-sur compat, but I've had to claw back some breaking changes that were already in master, so I think it'll merit broader testing
ris has quit [Ping timeout: 244 seconds]
rajivr has joined #nixos-dev
tetdim has quit [Ping timeout: 272 seconds]
p01ar has quit [Ping timeout: 260 seconds]
p01ar has joined #nixos-dev
tetdim has joined #nixos-dev
orivej has joined #nixos-dev
AlwaysLivid has joined #nixos-dev
makefu has quit [Ping timeout: 240 seconds]
makefu has joined #nixos-dev
jonringer has joined #nixos-dev
orivej_ has joined #nixos-dev
orivej has quit [Ping timeout: 244 seconds]
andi- has quit [Ping timeout: 258 seconds]
cole-h has joined #nixos-dev
orivej_ has quit [Ping timeout: 240 seconds]
justanotheruser has quit [Ping timeout: 244 seconds]
cole-h has quit [Quit: Goodbye]
orivej has joined #nixos-dev
domenkozar[m] has quit [*.net *.split]
colemickens has quit [*.net *.split]
feepo has quit [*.net *.split]
tazjin has quit [*.net *.split]
treed[m] has quit [*.net *.split]
bbigras has quit [*.net *.split]
flokli has quit [*.net *.split]
mingc has quit [*.net *.split]
andi- has joined #nixos-dev
colemickens has joined #nixos-dev
andi- has quit [Ping timeout: 244 seconds]
justanotheruser has joined #nixos-dev
orivej has quit [Ping timeout: 258 seconds]
alp has joined #nixos-dev
mingc has joined #nixos-dev
flokli has joined #nixos-dev
bbigras has joined #nixos-dev
tazjin has joined #nixos-dev
treed[m] has joined #nixos-dev
feepo has joined #nixos-dev
domenkozar[m] has joined #nixos-dev
orivej has joined #nixos-dev
alp has quit [Ping timeout: 240 seconds]
andi- has joined #nixos-dev
jonringer has quit [Ping timeout: 244 seconds]
teto has joined #nixos-dev
codyopel has quit [Quit: Idle for 30+ days]
ris has joined #nixos-dev
alp has joined #nixos-dev
alp has quit [Ping timeout: 244 seconds]
prusnak_ is now known as prusnak
prusnak has joined #nixos-dev
prusnak has joined #nixos-dev
prusnak has quit [Changing host]
peelz has joined #nixos-dev
lassulus_ has joined #nixos-dev
lassulus has quit [Ping timeout: 240 seconds]
lassulus_ is now known as lassulus
alp has joined #nixos-dev
orivej has quit [Ping timeout: 264 seconds]
alp has quit [Ping timeout: 246 seconds]
andi- has quit [Ping timeout: 244 seconds]
andi- has joined #nixos-dev
tilpner has quit [Remote host closed the connection]
tilpner has joined #nixos-dev
kloenk has quit [Ping timeout: 244 seconds]
tilpner has quit [Remote host closed the connection]
tilpner has joined #nixos-dev
Plien has joined #nixos-dev
kloenk has joined #nixos-dev
orivej has joined #nixos-dev
alp has joined #nixos-dev
AlwaysLivid has quit [Read error: Connection reset by peer]
AlwaysLivid has joined #nixos-dev
jonringer has joined #nixos-dev
cole-h has joined #nixos-dev
bennofs__ has joined #nixos-dev
bennofs_ has quit [Ping timeout: 260 seconds]
<infinisil>
Impressive, never seen this error before: error: cannot open connection to remote store 'daemon': reading from file: Connection reset by peer
<infinisil>
That's for any Nix command
<infinisil>
A build was running in another terminal, and when I stopped that build, it worked again
<abathur>
nice
<abathur>
I mentioned this in #nixos, but do you have any idea why a daemon install might be unable to find nixpkgs on path while a single-user is fine? (doing test runs against installers built from ~master, may also matter that it's on macOS)
<clever>
abathur: what about `sudo -i` then `nix-channel --list` ?
<abathur>
during the install it did run `sudo <vars> .../nix-channel --update nixpkgs`, and it did report that it was unpacking channels
<abathur>
clever: that does show it
rajivr has quit [Quit: Connection closed for inactivity]
<clever>
abathur: your sudo isnt updating $HOME, so `sudo nix-channel --list` shows the channels for the wrong user
alp has quit [Ping timeout: 244 seconds]
<clever>
abathur: darwin does that by default
<abathur>
gotcha
Plien has quit [Quit: Connection closed for inactivity]
<abathur>
not really sure, if root has a channel, why the error, though? not terribly familiar with daemon installs
<clever>
the daemon doesnt care about $NIX_PATH
<clever>
but your user must source the right nix.sh to setup $NIX_PATH
<clever>
and darwin doesnt have a /etc/bashrc to simplify it
<abathur>
mmm, there was a commit a while back that intentionally pulled NIX_PATH out? but I assumed from the commit date that it must already be live/released and that it probably wouldn't be the issue here
<abathur>
I guess that is unreleased, though; anyways--the bashrc and zshenv do get set up to source /nix/var/nix/profiles/default/etc/profile.d/nix-daemon.sh, but it doesn't export a NIX_PATH anymore
<abathur>
inclined to just call it out-of-scope for my update if that's actually why it's breaking, but trying to understand it well enough to make sure I'm not causing the problem
alp has joined #nixos-dev
__monty__ has joined #nixos-dev
orivej has quit [Ping timeout: 244 seconds]
orivej has joined #nixos-dev
cole-h has quit [Quit: Goodbye]
jonringer has quit [Ping timeout: 240 seconds]
orivej has quit [Ping timeout: 244 seconds]
justanotheruser has quit [Ping timeout: 244 seconds]
AlwaysLivid has quit [Ping timeout: 244 seconds]
__monty__ has quit [Quit: leaving]
jonringer has joined #nixos-dev
MichaelRaskin has quit [Quit: MichaelRaskin]
andi- has quit [Quit: WeeChat 2.8]
andi- has joined #nixos-dev
bennofs__ has quit [Remote host closed the connection]
<samueldr>
aminechikhaoui: I think it's part of a bigger issue
<samueldr>
that's just a symptom
<samueldr>
though I have no proof really, and need to coordinate with people that can verify my assumptions
<aminechikhaoui>
what's the bigger issue ?
<samueldr>
I'm pretty sure the reason it fails is the same that firefox fails on aarch64
<samueldr>
so let me explain with firefox for which I have made more observations
<samueldr>
when the queue is basically empty for aarch64, firefox builds in less than 1 hour
<aminechikhaoui>
oh ok so you're suspecting build sharing is the problem ?
<samueldr>
when there is a lot in the queue for aarch64, firefox will not achieve a build for the timeout
<samueldr>
yes
<samueldr>
and timeout I'm not sure if it's 4 hours or 10 hours
<samueldr>
but still, if it was distributed "better" (how?) we'd be able to build 4 firefoxen in the time it takes for one to fail
<samueldr>
and possibly making other "big" builds fail!
<samueldr>
and maybe even not-big-builds take order of magnitudes longer!
<aminechikhaoui>
yeah I agree, it would need a much smarter queue runner and maybe some hints in package definitions like a weight or something for it to be better distributed I think
<aminechikhaoui>
but this particular fonts package I think is slow eitherway
<aminechikhaoui>
it's been running for a while on my rpi4 on its own
<samueldr>
pretty sure the contention issue is entirely filesystem-based
<samueldr>
and that font is likely filesystem-heavy!
<samueldr>
I'm saying that only has a gut feeling
<samueldr>
only as*
<aminechikhaoui>
it seems all what it's doing is optimizing pngs with zopflipng
<aminechikhaoui>
that's all what I can see in `top`
<samueldr>
quite possible
<aminechikhaoui>
though it's interesting it's faster on x86_64 or fast enough to not trigger a problem
<samueldr>
the ARM machines might be oversubscribed because they have way-many-cores
<samueldr>
in an I/O free world that would possibly be fine
<samueldr>
IIRC 64-96 cores
<aminechikhaoui>
I guess it depends how we set the number of jobs per aarch64 box in the queue runner config
<aminechikhaoui>
yeah we have a config file that sets that but Graham might know better about it :)
<samueldr>
yeah, I know he could help, but I know he's got more important things to do, and people to care for