vcunat has quit [(Quit: Leaving.)]
lassulus has quit [(Ping timeout: 260 seconds)]
lassulus has joined joined #nixos-dev
shlevy has quit [(Ping timeout: 246 seconds)]
shlevy has joined joined #nixos-dev
ris has quit [(Ping timeout: 264 seconds)]
Sonarpulse has quit [(Ping timeout: 260 seconds)]
mbrgm has quit [(Ping timeout: 260 seconds)]
mbrgm has joined joined #nixos-dev
ma27 has joined joined #nixos-dev
ma27 has quit [(Ping timeout: 255 seconds)]
jtojnar has quit [(Ping timeout: 255 seconds)]
orivej has quit [(Ping timeout: 240 seconds)]
goibhniu has joined joined #nixos-dev
ma27 has joined joined #nixos-dev
vcunat has joined joined #nixos-dev
ma27 has quit [(Ping timeout: 250 seconds)]
orivej has joined joined #nixos-dev
orivej_ has joined joined #nixos-dev
orivej has quit [(Ping timeout: 268 seconds)]
orivej has joined joined #nixos-dev
orivej_ has quit [(Ping timeout: 276 seconds)]
<vcunat> the current state of Darwin build slaves on Hydra is poor
<vcunat> it runs like seven jobs at once, only
<vcunat> This way the Darwin queue is growing persistently.
<vcunat> It seems a bottleneck that Hydra administration is so trust-sensitive.
<vcunat> But there's probably no easy way around that.
<LnL> yeah, the darwin stable idea started with the idea that we have enough build capacity now
<vcunat> as we have separate nixos-channels, the only bottle-necking I see is in staging where one may be unsure (a) if the changes make a mass breakage on Darwin, and (b) whether to merge to staging without most Darwin binaries ready
<vcunat> Fortunately, staging changes typically don't hurry, and pre-testing them on the Linux platform usually gives a lot confidence (depending on the type of the changes).
<LnL> my point is that if we're not at full capacity the 17.09 job gets priority so staging/master start lagging behind
<vcunat> Yes, that happened a week or two ago.
<vcunat> We technically *can* reverse the priorities.
<vcunat> (Instead of cancelling 17.09-darwin.)
<vcunat> 14 jobs running now, so hopefully I mostly looked at some of the worst moments.
<gchristensen> I just kicked off a GC on the macs
<gchristensen> we need to cron that
<LnL> that's not nice, but we could do that for now until the issues are resolved
<vcunat> right, some fast cron that checks free space and fires GC if under some threshold (and not running already)
<gchristensen> something decent like that could even just be a while loop + sleep + screen and I could do that this morning
<vcunat> <3
<vcunat> And Linux boxes do have some auto-GC?
<gchristensen> yeah, the standard nixos module for auto gc
<gchristensen> which I think is a cron ;)
<vcunat> nix.gc.automatic
<vcunat> We need NixOS services for more platforms :-D
<LnL> there's a nix.gc module for nix-darwin :)
<vcunat> gchristensen: ^^
<LnL> you could also just build that and copy/load it manually
<gchristensen> yeah I'm ready too merge that in to be honest
<LnL> if you don't want to switch over entirely
<gchristensen> LnL: can you do that for me and send it over?
<LnL> sure
<gchristensen> I've got a bit of a busy morning for the next bit
<gchristensen> so ... I'll be mostly AFK for a bit
<gchristensen> thanks!
<LnL> that's what I use
<LnL> you might want to tweak the 25 to something higher for the hydra macs
<vcunat> That aims to having 25 GiB of free space, right?
<LnL> yeah
<vcunat> hmm, yes, this kind of GC can be ran relatively often, I think
<gchristensen> LnL: how do I load this again?
<gchristensen> / start
<LnL> launchctl load ~/Library/LaunchAgents/org.nixos.nix-gc.plist
<LnL> or /Library for root
<LnL> not sure if it starts up immediately or if that triggers the timer
<gchristensen> ok, cool
<LnL> oh right, good idea :)
<gchristensen> hmmm I think I should wait on this until the current GC is done
<gchristensen> LnL: does that need a </plist> at the end?
<LnL> probably, maybe I didn't copy everything
<gchristensen> how much free space should we shoot for
<gchristensen> they have 900GB disks
<gchristensen> how about 200G
<vcunat> +0
<vcunat> +1 :-)
<LnL> I have no idea how much space they can fill up in a couple of hours
<gchristensen> I just spent like 20 minutes writing a bash script to fetch / verify / run a bash script which fetches / verifies / places a file... and then realized I could just do it in nix and be done.
<gchristensen> stupid. :P
<vcunat> :-D
<vcunat> after years of using nix(os) one may find oneself so powerless on a system without nix
<vcunat> it's just addictive
<gchristensen> no kidding
<gchristensen> amazing how good life can be when you have tools that work
<niksnut> speaking of macs, we're going to have to find a new home for them soon since this office is closing
<gchristensen> :o
<LnL> oh
<gchristensen> LnL: how can I tell if this is working?
<LnL> launchctl list should ld show 0 if it ran successfully once
<LnL> niksnut: btw. did you see my comment about the sandobx/pre-build-hook?
<niksnut> no
<gchristensen> it shows `91 0 org....` does that seem right lnl?
<LnL> I've seen you made some changes to avoid having to send all of the drvs to remote builders
<LnL> so I'm not sure what the best way to solve that would be
<LnL> gchristensen: I think so, is it running right now?
<niksnut> LnL: I agree with copumpkin's comment
<niksnut> resolve-system-dependencies could just be a function call
<gchristensen> well nix-collect-garbage is running but I'm not sure it is because of launchd
<niksnut> not relying on pre-build-hook would be good anyway since the whole build hook mechanism may go away
<LnL> right
<LnL> I'll see if I can get something working
<vcunat> niksnut: the old grant servers are still remaining at the university?
<niksnut> yes
<vcunat> (but I guess it's not a good place for the macs)
<LnL> gchristensen: that first number is the pid, so I would expect it to be running
<gchristensen> cool
<niksnut> some cache.nixos.org stats: https://nixos.org/~eelco/cache-stats/
<gchristensen> I spy an S3 outage :D
jtojnar has joined joined #nixos-dev
<vcunat> oh, thanks a lot... I'll certainly be thinking more about https://github.com/NixIPFS/infrastructure/issues/4 during December
<gchristensen> I mean not now, just in those graphs
<vcunat> To be clear, I didn't mean to react to your comment, only to Eelco's :-)
<gchristensen> ah
<LnL> s3 outage? must be a metrics bug
<vcunat> The bandwidth is like 100 Mbit/s on average now, if I look correctly.
<gchristensen> oh you're right, lnl
<vcunat> (a terabyte per day, in other words)
<vcunat> Could it be an outage in S3 metrics?
<LnL> :p
<LnL> the traffic is higher then I expected
<vcunat> the stange jump in 2016/Q1
<vcunat> ~5k unique IPs per day - that's an interesting number
<niksnut> LnL: yeah, it looks like the logs for 2017-10-12 are very incomplete
<niksnut> I don't recall any outage at that time
<niksnut> vcunat: iirc, in february 2016 we started using 1000s of spot instances at LB (probably ikwildrpepper remembers)
<vcunat> So half of that will go to LogicBlox :-D
taktoa has joined joined #nixos-dev
<vcunat> Oh, actually not. Monthly the IP counts are ~90k.
<aminechikhaoui> niksnut: I think that was before February 2016, when I joined in late 2015 it was already being used
<gchristensen> oh hi aminechikhaoui, any news on that nixops thing?
<aminechikhaoui> Hey there, the packet.net backend ?
<gchristensen> eah
<aminechikhaoui> unfortunetly, I stopped working on that as I got pulled in other stuff in work. But will likely get back to that as soon as I can ;)
<gchristensen> ok, let me know when you can
<aminechikhaoui> gchristensen: sure
vcunat has quit [(Quit: Leaving.)]
FRidh has joined joined #nixos-dev
<gchristensen> huge progress by Dezgeg today: Hydra can now run NixOS tests on aarch64-linux
<gchristensen> here is Dovecot: https://hydra.nixos.org/build/65512677
<goibhniu> fantastic!
<niksnut> o/
orivej has quit [(Ping timeout: 248 seconds)]
jtojnar has quit [(Ping timeout: 276 seconds)]
__Sander__ has joined joined #nixos-dev
<__Sander__> hehe
* __Sander__ just learned that the NPM registry always ignores package-lock.json files in the tarballs
<__Sander__> so that means I still have to keep my own implementation of the dependency resolution algorithm
<__Sander__> for packages that are directly installed from the registry
ma27 has joined joined #nixos-dev
vcunat has joined joined #nixos-dev
ma27 has quit [(Ping timeout: 255 seconds)]
__Sander__ has quit [(Quit: Konversation terminated!)]
vcunat has quit [(Ping timeout: 258 seconds)]
goibhniu has quit [(Ping timeout: 250 seconds)]
ma27 has joined joined #nixos-dev
vcunat has joined joined #nixos-dev
ma27 has quit [(Quit: WeeChat 1.9.1)]
Sonarpulse has joined joined #nixos-dev
<Sonarpulse> thanks for those normalization commits, btw
<Sonarpulse> (for GCC)
<Dezgeg> is https://hydra.nixos.org/jobset/nixpkgs/ericson2314-cross-trunk rebased on top of 703a9f93c ?
<Dezgeg> that might explain some ARM failures
<Sonarpulse> Dezgeg: checking
<Sonarpulse> Dezgeg: ah no, not yet
<Sonarpulse> I temporary set it way way back
<Sonarpulse> to the summer
<Sonarpulse> right after my commit which got rid of gcc-wrapper-cross
<Sonarpulse> I rewrote the builder script
<Dezgeg> ah, so it doesn't contain 1c12072 either? then I don't know off hand
<Sonarpulse> which got aarch4 to pass
<Sonarpulse> Dezgeg: sure no worrries. It and the other job is only differ via commits I just wrote now, unmerged
<Dezgeg> that does appear to set --with-float=hard twice for some reason, that probably shouldn't affect things
<Dezgeg> I wonder if the order of configure flags is significant?
<Sonarpulse> I did a dump of the environment variables and diffed
<Sonarpulse> I think the duplication and other stuff is the same
<Sonarpulse> except for the --with-native-include-dir=.... part
<Dezgeg> maybe the glibc is soft-float built then?
<Sonarpulse> nah checked
<Sonarpulse> it's hard float
<Sonarpulse> the missing header is the soft-float header
<Sonarpulse> echo '__ARM_PCS_VFP' | $CC -x c -o - - -E
<Sonarpulse> that should be #defined as 1 automatically by GCC
<Sonarpulse> and is in a nix-shell
<Sonarpulse> but somehow isn't when the command fails
<Sonarpulse> Dezgeg: if the PR I linked looks good to you, I'm quite confident it isn't making anything worse.
<Sonarpulse> the same 96 builds succeed early in the history
<Dezgeg> so the command is 'g++', so it's including an ARM header file when compiling native stuff?
<Sonarpulse> ahhh g++
ris has joined joined #nixos-dev
<Sonarpulse> weird
<Sonarpulse> I'll check again
<Dezgeg> it's calling g++ -I/nix/store/v1y7sdp8jd8kpg4sa5ca614bvmw05lk3-glibc-2.25-arm-unknown-linux-gnueabihf-arm-unknown-linux-gnueabihf-dev/include etc.
<Dezgeg> which sounds like a problem
<Sonarpulse> weeeird
<Sonarpulse> Dezgeg: per 5e6083b7ec703e29a75d20d12641d41e445c84f8
<Sonarpulse> in pkgs/development/compilers/gcc/builder.sh
<Sonarpulse> which is the commit that breaks it
<Dezgeg> I don't have that commit
<Sonarpulse> CFLAGS_FOR_BUILD and CXXFLAGS_FOR_BUILD should be defined the same way
<Sonarpulse> Dezgeg: it's not in the PR, only good commits in the PR. ericson2314-gcc-more-prog-envs from obsidiansystems/nixpkgs has it
<Sonarpulse> git@github.com:obsidiansystems/nixpkgs.git
<Sonarpulse> ok weird from my env var dump
<Sonarpulse> 7]="CXXFLAGS_FOR_BUILD=-O2 -g -I/nix/store/eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee-glibc-2.25-dev/include -B/nix/store/eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee-glibc-2.25/lib/ -idirafter /nix/store/eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee-glibc-2.25-dev/include -idirafter /nix/store/eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee-gcc-5.4.0/lib/gcc/x86_64-unknown-linux-gnu/5.4.0/include-fixed -Wl,-L/nix/store/eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee-glibc-2.25/lib -Wl,-rpath
<Sonarpulse> -Wl,/nix/store/eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee-glibc-2.25/lib -Wl,-L/nix/store/eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee-glibc-2.25/lib -Wl,-dynamic-linker -Wl,/nix/store/eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee-glibc-2.25/lib/ld-linux-x86-64.so.2"
<Sonarpulse> that's in makeFlagsArray
<Sonarpulse> so second -idirafter
jtojnar has joined joined #nixos-dev
<Dezgeg> from the hydra build log to me it looks like CFLAGS=-O2 -g -I/nix/store/v1y7sdp8jd8kpg4sa5ca614bvmw05lk3-glibc-2.25-arm-unknown-linux-gnueabihf-arm-unknown-linux-gnueabihf-dev/include in makeFlags
<Sonarpulse> Dezgeg: that makes sense
<Sonarpulse> it's a build != host == target build
<Sonarpulse> so host and target are arm
<Sonarpulse> it's just this build one the arm stuff is leaking in when it should be x86
<Dezgeg> are you sure that's not passed to the build gcc?
<Sonarpulse> Dezgeg: I doubt it, cause I see a problem CXXFLAGS_FOR_BUILD
<Sonarpulse> and gcc worked fine
<Sonarpulse> just g++ failed
<Dezgeg> that starts with CXXFLAGS_FOR_BUILD=-O2 -g -I/nix/store/vrr9maj9lqj2xwndlx3kh07vhnc111i2-glibc-2.25-dev/include which is correct
<Dezgeg> which does not appear at all in the failing command line
<Sonarpulse> oh sorry you are right
<Sonarpulse> Dezgeg: actually there might be a _FOR_HOST
<Dezgeg> is CXXFLAGS_FOR_BUILD something that's actually used by gcc?
<Sonarpulse> I'll check
<Sonarpulse> it is a thing in gcc 5
<bgamari> Sonarpulse, how are things?
<Sonarpulse> bgamari: pretty good
<Sonarpulse> bgamari: https://github.com/NixOS/nixpkgs/pull/32361 is some freebe improvements
<Sonarpulse> I tested don't mess up the last completely known-good commit
<Sonarpulse> is in-progress work changing build.sh
<Sonarpulse> so that gcc build != host == target continues to work
<Sonarpulse> vcunat reviewed binutils wrapper---thanks again!---so just waiting on edolstra
<Sonarpulse> orivej did point out some failures with that one that I'll fix before merging, but review's are on the concept anyways
<bgamari> righty
<Sonarpulse> I think based on what Dezgeg just found
<bgamari> alright, so I will wait a bit longer before trying a rebase I guess
<Sonarpulse> issue is gcc 5 may not be happy with CXXFLAGS being defined
<Sonarpulse> sure
<bgamari> hmm
<Sonarpulse> old build.sh just did CXXFLAGS_FOR_BUILD and CXXFLAGS_FOR_HOST
<Sonarpulse> *_FOR_TARGET
<Sonarpulse> there is no _FOR_HOST, that would be the plain one
<Sonarpulse> but I guess that messes things up
<bgamari> sounds plausible
<Sonarpulse> basically trying to pre-fix it right before the regression, then merge with future bit by bit crossing fingers
goibhniu has joined joined #nixos-dev
<gchristensen> niksnut: may I create a jobset later today for grahamcofborg?
goibhniu has quit [(Ping timeout: 248 seconds)]
MoreTea has quit [(Ping timeout: 255 seconds)]
MoreTea has joined joined #nixos-dev
adisbladis has quit [(Remote host closed the connection)]
mbrgm has quit [(Ping timeout: 260 seconds)]
hedning[m] has quit [(Ping timeout: 252 seconds)]
regnat[m] has quit [(Ping timeout: 240 seconds)]
stites[m] has quit [(Ping timeout: 264 seconds)]
florianjacob has quit [(Ping timeout: 264 seconds)]
adisbladis[m] has quit [(Ping timeout: 255 seconds)]
copumpkin has quit [(Ping timeout: 248 seconds)]
sphalerite has quit [(Ping timeout: 248 seconds)]
adisbladis has joined joined #nixos-dev
mbrgm has joined joined #nixos-dev
nocent has quit [(Ping timeout: 246 seconds)]
moredread[m] has quit [(Ping timeout: 264 seconds)]
olejorgenb[m] has quit [(Ping timeout: 252 seconds)]
rycee has quit [(Ping timeout: 255 seconds)]
<mog> im trying to install signal-desktop but it seems to be getting tripped up by the dash
<mog> how do i pass that into nix-env -i
<mog> nevermind i got it working, bash was just being odd
goibhniu has joined joined #nixos-dev
<bgamari> Sonarpulse, do you have a branch that would have a reasonable chance of working if I were to rebase?
<bgamari> unfortunately it turns out that my fallback plan to cross-compilation isn't going to work
<Sonarpulse> bgamari: hmm
<Sonarpulse> bgamari: I have rebased branches, but IIRC they were sort of broken
<Sonarpulse> but that's resolved now
<Sonarpulse> I can get you something soon
<Sonarpulse> Dezgeg: still there? thanks again for the help earlier
Nadrieril has quit [(Ping timeout: 240 seconds)]
<Sonarpulse> trying without CXXFLAGS now
<bgamari> Sonarpulse, cool
<Sonarpulse> ah the CXXFLAGS thing didn't work; it seems correct but causes other problems
<bgamari> :(
<Sonarpulse> bgamari: ah wait!
* bgamari waits
<Sonarpulse> I didn't do my test right
<Sonarpulse> I thought I had seen this other error before
<Sonarpulse> could still work after all
<Sonarpulse> I need to merge first
<Sonarpulse> Dezgeg: thanks for all your help before on debugging that other stuff, but did you have an opinion on that PR itself?
<bgamari> Sonarpulse, have you confirmed that 32361 and the CXXFLAGS thing together work?
<Sonarpulse> bgamari: that's what I've been checking
<Sonarpulse> the first commits there
<Sonarpulse> before the merges
<Sonarpulse> are me preparing for my builder.sh rewrite (with the CXXFLAGS that I now regret)
<Sonarpulse> preparing in that there is no native hash change
<Sonarpulse> it's just nix-level cleanup
<Sonarpulse> so I want to merge that PR
<Sonarpulse> then do the same chain of merges with the builder.sh change after
<Sonarpulse> then look for the next regression (if there even is one!)
Nadrieril has joined joined #nixos-dev
orivej has joined joined #nixos-dev
goibhniu has quit [(Ping timeout: 276 seconds)]
<Profpatsch> dtzWill: Saw your ping
<Profpatsch> The tests implementation is burning hot on my soul. :D
<Profpatsch> Maybe I should just create a PR with the current implementation.
<Profpatsch> The docs are not yet what they should be, it’s more of a list of patches I cherry-pick around.
<Sonarpulse> orivej: see my followup comment in https://github.com/NixOS/nixpkgs/pull/32361 ?
<orivej> Sonarpulse: yes, I'm trying to understand your reasoning, so far I don't see why merging old commits into master is more convenient than tegging them
<Sonarpulse> orivej: oh those are *new* merge commits
<Sonarpulse> they don't currently exist
<Sonarpulse> they are me forward porting my gcc fix
<Sonarpulse> through various conflicting gcc changes
<gchristensen> vcunat: I think we should merge https://github.com/NixOS/nixpkgs/pull/32365 soonish lest it get out of date
<gchristensen> preferably with that handleEvalIssue patch :)
<orivej> Sonarpulse: hm, and why do you need these changes applied prior to old commits? to see at what point between the base and the head of the PR they broke something?
<Sonarpulse> orivej: yeah the first old commit is a known good state
<Sonarpulse> the last one
<Sonarpulse> to validate that these changes indeed are entirely safe
<Sonarpulse> the second merge commit has issues that my next PR will hopefully fix
<Sonarpulse> but I need that PR to depend on this one
<Sonarpulse> I don't want to throw away my testing commits
<vcunat> gchristensen: I don't think it will bit-rot quickly
<Sonarpulse> but rather commit them
<Sonarpulse> so I can reference them later
<Sonarpulse> sort of a "history proof of correctness"
<vcunat> I had to fix one small conflict in a few hours, but that was probably just bad luck (and it took under a minute anyway)
<gchristensen> oh
<gchristensen> well
<gchristensen> I guess maybe I'm excited to do it :P
<vcunat> well, the high-churn changes can actually be merged fast
<vcunat> what I'm not very certain yet is how to do the check-meta.nix change
<gchristensen> yeah
andi- has joined joined #nixos-dev
vcunat has quit [(Quit: Leaving.)]
ma27 has joined joined #nixos-dev