aparcar[m] has joined #openwrt-devel
schmars[m] has joined #openwrt-devel
rmilecki has quit [Ping timeout: 480 seconds]
cmonroe has quit [Read error: Connection reset by peer]
cmonroe has joined #openwrt-devel
aparcar[m] has quit [Ping timeout: 480 seconds]
schmars[m] has quit [Ping timeout: 480 seconds]
goliath has quit [Quit: SIGSEGV]
Daanct12 has joined #openwrt-devel
n3ph has joined #openwrt-devel
asriel has quit [Quit: Don't drink the water. They put something in it to make you forget.]
Danct12 has quit [Quit: ZNC 1.9.1 - https://znc.in]
Danct12 has joined #openwrt-devel
minimal has quit [Quit: Leaving]
aparcar[m] has joined #openwrt-devel
<djfe> as a downstream fork of OpenWrt https://github.com/freifunk-gluon/gluon : how does the switch to Alpine Package Manager (apk) affect us in future releases like OpenWrt 25?
<djfe> Do we have to rewrite the packages or is it mostly repackaging and switching from the current makefiles to a new format? (adjusting metadata)
schmars[m] has joined #openwrt-devel
hexa- has quit [Quit: WeeChat 4.2.2]
hexa- has joined #openwrt-devel
Daaanct12 has joined #openwrt-devel
Daanct12 has quit [Read error: Connection reset by peer]
<slh> djfe: for the most part, after merging the base changes, it's making sure that the version numbers meet the -stricter- expectations of apk. the packaging itself needs very few (no?) changes
<schmars[m]> can't depend on virtual packages like iptables/iptables6 anymore somehow - i had to change dependencies to iptables-nft (which is recommended anyway i guess)
<slh> yes, apk will expose all kinds of dependency issues opkg didn't notice or care about
<slh> but for the most part, that aspect is fixing what was already broken before
<slh> that doesn't make it anymore fun or necessarily easier, but it's not changing the whole world
<schmars[m]> yeah i'm surprised i had to do nothing else for these ~30 packages in my feed
rua has quit [Quit: Leaving.]
Daaanct12 has quit [Quit: WeeChat 4.4.3]
n3ph has quit [Ping timeout: 480 seconds]
Daanct12 has joined #openwrt-devel
valku has quit [Quit: valku]
Sawzallz has joined #openwrt-devel
sorinello has joined #openwrt-devel
Sawzall has quit [Ping timeout: 480 seconds]
rmilecki has joined #openwrt-devel
<owrt-images-builds> Build [#7](https://buildbot.openwrt.org/images/#/builders/239/builds/7) of `openwrt-24.10_layerscape/armv8_64b` completed successfully.
rua has joined #openwrt-devel
<KanjiMonster> how do I extract a .apk? I want to inspect its content, but so far haven't found anything that works (tar -xzf does not)
<russell--> i found something that works: # CONFIG_USE_APK is not set ;-)
<russell--> "An APK file is a ZIP archive that usually contains the following files and directories: " from https://en.wikipedia.org/wiki/Apk_(file_format)
rua is now known as Guest9382
rua has joined #openwrt-devel
<KanjiMonster> russell--: as I already said, "tar -xzf" does not work. I also found that snippet ;p
<\x> just tried non of those work
<\x> and its not the zip apk on android too, so you cant just flash it on your receovery like magisk.apk :D
<\x> so yeah i also wonder now
<\x> even binwalk doesnt know about it
Guest9382 has quit [Ping timeout: 480 seconds]
<russell--> oh, wrong apk
<\x> russell--: yeah thats the android apk, like how Magisk-v28.0.apk just flashes on recovery like a recovery zip on android
<KanjiMonster> so I guess this is the custom binary format v3, not the tar-like v2
<Habbie> correct
<Habbie> i learned 'apk adbdump' recently
<KanjiMonster> so the answer is "apk extract --allow-untrusted" (using the apk from staging_dir/host/bin)
<Habbie> ah
<KanjiMonster> doesn't help that the staging apk has help disabled
<Habbie> yeah. i think we should change that
<Habbie> you're not usually tight on space there :)
<KanjiMonster> mostly I'm trying to figure out where /etc/apk/arch is coming from
<Habbie> i booted an alpine container to read help last week
<KanjiMonster> because it seems imagebuilder generated images always have this set to x86_64, regardless of the target
<owrt-images-builds> Build [#7](https://buildbot.openwrt.org/images/#/builders/227/builds/7) of `openwrt-24.10_armsr/armv8` completed successfully.
vincejv has quit [Remote host closed the connection]
<KanjiMonster> looks like these indexes apk generates (APKINDEX.*.tar.gz) aren't gz either
Stat_headcrabbed has joined #openwrt-devel
<Habbie> are they called tar.gz?
<owrt-images-builds> Build [#7](https://buildbot.openwrt.org/images/#/builders/38/builds/7) of `openwrt-24.10_qualcommax/ipq807x` failed.
<KanjiMonster> yes
rua has quit [Ping timeout: 480 seconds]
robimarko has joined #openwrt-devel
<ynezz> KanjiMonster: staging apk should have help enabled https://github.com/openwrt/openwrt/commit/b2a84c86, but probably its being stripped (or disabled again by other means?) when apk is being transferred into imagebuilder?
<KanjiMonster> ynezz: both the downloaded IB's apk and the one in my local build tree from yesterday say "ERROR: This apk-tools has been built without help"
<KanjiMonster> I also see no lua in build_dir/host
<ynezz> I just finished build and I've apk with help, so there is probably something else going on?
<ynezz> ah, maybe "Program lua5.3 found: YES (/usr/bin/lua5.3)" ?
<KanjiMonster> that might be it
<Habbie> i don't use imagebuilder and my apk does not have help
<Habbie> mm lua
<ynezz> staging_dir/host/bin/meson.py --internal exe --capture src/help.h -- /usr/bin/lua5.3 ../src/genhelp.lua
<KanjiMonster> no lua on my host system, so /usr/bin/lua5.3 does not exist
<KanjiMonster> and I assume neither on the autobuilders
<ynezz> yes, the same applies to buildbot
<ynezz> this is leakage of the host system into the build anyway
<KanjiMonster> I found a lua in build_dir/hostpkg, but it's lua 5.1, not lua 5.3 (and installed in staging_dir/hostpkg) - what's the difference between /host and /hostpkg?
GNUmoon has quit [Remote host closed the connection]
<ynezz> Program lua5.3 found: YES (/var/home/ynezz/dev/openwrt/openwrt.git/staging_dir/hostpkg/bin/lua5.3)
<ynezz> meson.py --internal exe --capture src/help.h -- /var/home/ynezz/dev/openwrt/openwrt.git/staging_dir/hostpkg/bin/lua5.3
<KanjiMonster> ynezz: MESON_ARGS has -Dlua_version=5.1
<ynezz> this is for target
<KanjiMonster> ah, I see
<ynezz> maybe we could do the same for the host?
GNUmoon has joined #openwrt-devel
<KanjiMonster> seems to work: Program lua5.1 found: YES (/home/jonas/opensource/openwrt/jogo/staging_dir/hostpkg/bin/lua5.1)
<ynezz> yep
<KanjiMonster> tho it also says "Run-time dependency lua5.1 found: NO (tried pkgconfig and cmake)"
<ynezz> its probably not needed for help, its just generating help.h so its probably shared by the host/target in the end
<KanjiMonster> yeah, seems to work fine without it
<KanjiMonster> since I can't find any other packages using lua5.3/host I'd say let's just set the luaversion to 5.1 for host as well
<ynezz> +1
<KanjiMonster> probably doesn't fix the broken arch in IB images tho ;p
<Habbie> i see 5.1 in my hostpkg, fwiw
<KanjiMonster> right, it is build because apk/host as it as a dependency, but nobody told apk/host to use it (and default version it tries to use is 5.3)
aparcar[m] has quit [Quit: Bridge terminating on SIGTERM]
schmars[m] has quit [Quit: Bridge terminating on SIGTERM]
<Habbie> ohh. an apk problem i can work on in -work- hours
<ynezz> lovely :)
rua has joined #openwrt-devel
<KanjiMonster> Habbie: which one, the broken arch in IB?
<Habbie> no, the "Store packages" step in multi-arch-test-build is broken
<KanjiMonster> ah, so nothing from here
<owrt-images-builds> Build [#9](https://buildbot.openwrt.org/images/#/builders/188/builds/9) of `main_realtek/rtl838x` failed.
<owrt-images-builds> Build [#9](https://buildbot.openwrt.org/images/#/builders/259/builds/9) of `main_d1/generic` failed.
<ynezz> the above failures are related to the recent apk bump (CI didn't catched it), thus going to be fixed with https://github.com/openwrt/openwrt/pull/16951
<owrt-images-builds> Build [#9](https://buildbot.openwrt.org/images/#/builders/126/builds/9) of `main_bmips/bcm6362` failed.
<owrt-images-builds> Build [#7](https://buildbot.openwrt.org/images/#/builders/238/builds/7) of `openwrt-24.10_malta/le` completed successfully.
<Habbie> is 24.10 switched to apk?
<Habbie> oh i'll just check
<ynezz> Habbie: no
<Habbie> ok cool
<Habbie> so the reusable multi arch workflow needs to support both
<Habbie> or be split again
<Habbie> fine
<ynezz> KanjiMonster: do you plan to fix that apk help by yourself?
<KanjiMonster> ynezz: sorry, I also need to sometimes do actual work stuff ;P Give me a second
<ynezz> KanjiMonster: sure, no worries, if you don't plan to do that, I'll do it, that was basically the question :P
<owrt-images-builds> Build [#10](https://buildbot.openwrt.org/images/#/builders/58/builds/10) of `main_apm821xx/sata` failed.
<ynezz> KanjiMonster: awesome, thanks!
<owrt-images-builds> Build [#7](https://buildbot.openwrt.org/images/#/builders/196/builds/7) of `openwrt-24.10_sunxi/cortexa7` completed successfully.
<owrt-images-builds> Build [#9](https://buildbot.openwrt.org/images/#/builders/175/builds/9) of `main_imx/cortexa7` failed.
n3ph has joined #openwrt-devel
<KanjiMonster> pushed
Guest9392 has joined #openwrt-devel
n3ph has quit [Ping timeout: 480 seconds]
n3ph has joined #openwrt-devel
goliath has joined #openwrt-devel
<Habbie> i love how responsive they are
<robimarko> Yeah, that is not common
goliath has quit [Quit: SIGSEGV]
totkeks has joined #openwrt-devel
<totkeks> Hello! Is there a dedicated place, like one special forum thread or something, to post issues with the new packaging system (apk)? or should I just ask here / create a new thread on forum for each issue?
<ynezz> robimarko: which "no buffer space" case exactly?
<ynezz> robimarko: just asking, because to me its not clear from that diff and/or Git commit description :P
<robimarko> I am guessing that should be the same as the revert reason: https://github.com/openwrt/openwrt/commit/dab5ca818ef46b32a9c120b5136fd5a1a34a88a5
<ynezz> you can see it in that merge request on the last line "Timo Teräs closed with commit 417a93ce 1 day ago"
<ynezz> and that commit actually contains "fixes #11038" which makes it clear
skynet2 has joined #openwrt-devel
<robimarko> Yeah, that makes sense
<ynezz> I'm actually trying to backport that in https://github.com/openwrt/openwrt/pull/16951 but it fails on CI, need to dig into this
skynet2_ has joined #openwrt-devel
<ynezz> robimarko: we currently don't test apk changes on the CI, so you need to remove/add the 'ci:target:mediatek:filogic' to see the issue
<ynezz> so the CI green status is bit misleading in this situations
<robimarko> ynezz: I was just rebasing as revert was already merged
<robimarko> I have applied that locally, lets see if it pops up
skynet2 has quit [Ping timeout: 480 seconds]
<robimarko> Well, locally qualcommax still build happily
<ynezz> yes, mediatek/filogic builds here as well, yet the CI fails
<ynezz> I assume its maybe because the build is not using s3 cache
<ynezz> so it might behave differently then the full build?
<ynezz> BTW Ansuel seems to bisect that /dev/zero no buffer issue to https://gitlab.alpinelinux.org/alpine/apk-tools/-/commit/b231eacd8121
f00b4r0 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
Daanct12 has quit [Quit: WeeChat 4.4.3]
f00b4r0 has joined #openwrt-devel
<ynezz> CI fails due to "linux-mediatek_filogic/linux-6.6.60.141efc9e/net/netfilter/ipvs/ip_vs_wrr.ko': No space left on device" is this common?
<owrt-images-builds> Build [#9](https://buildbot.openwrt.org/images/#/builders/130/builds/9) of `main_mediatek/mt7623` stopped with exception.
Sawzallz has quit [Read error: Connection reset by peer]
totkeks has quit [Quit: Page closed]
Sawzall has joined #openwrt-devel
<stintel> ugh what's with those 10GbE-PD switches where the PD-port is 1GbE only :/
skynet2 has joined #openwrt-devel
<stintel> PoE-PD*
skynet2_ has quit [Ping timeout: 480 seconds]
rua has quit [Remote host closed the connection]
rua has joined #openwrt-devel
<owrt-images-builds> Build [#7](https://buildbot.openwrt.org/images/#/builders/245/builds/7) of `openwrt-24.10_ramips/mt76x8` failed.
minimal has joined #openwrt-devel
schmars[m] has joined #openwrt-devel
skynet2 has quit [Ping timeout: 480 seconds]
goliath has joined #openwrt-devel
fakuivan_ has joined #openwrt-devel
fakuivan has quit [Ping timeout: 480 seconds]
valku has joined #openwrt-devel
tohojo has joined #openwrt-devel
fakuivan_ has quit [Remote host closed the connection]
fakuivan has joined #openwrt-devel
<mrkiko> robimarko: saw the mail - thanks...
<owrt-images-builds> Build [#9](https://buildbot.openwrt.org/images/#/builders/65/builds/9) of `main_lantiq/ase` completed successfully.
noltari_ has quit [Quit: Bye ~ Happy Hacking!]
noltari has joined #openwrt-devel
goliath has quit [Quit: SIGSEGV]
vincejv has joined #openwrt-devel
fakuivan has quit [Remote host closed the connection]
fakuivan has joined #openwrt-devel
n3ph has quit [Ping timeout: 480 seconds]
<owrt-images-builds> Build [#9](https://buildbot.openwrt.org/images/#/builders/46/builds/9) of `main_sunxi/cortexa7` failed.
Sawzallz has joined #openwrt-devel
Sawzall has quit [Ping timeout: 480 seconds]
archcg has joined #openwrt-devel
archcg has quit []
cmonroe_ has joined #openwrt-devel
cmonroe has quit [Ping timeout: 480 seconds]
minimal has quit [Quit: Leaving]
cmonroe has joined #openwrt-devel
<Habbie> aparcar, woops. just noticed runtime_test in the matrix. my failures might be fine for now :)
<Habbie> aparcar, any idea why arm_cortex-a9 has false there?
cmonroe_ has quit [Ping timeout: 480 seconds]
cmonroe has quit [Ping timeout: 480 seconds]
<owrt-images-builds> Build [#9](https://buildbot.openwrt.org/images/#/builders/83/builds/9) of `main_bcm27xx/bcm2711` failed.
ScrewDri- has joined #openwrt-devel
ScrewDriver1337 has quit [Read error: Connection reset by peer]
ScrewDri- is now known as ScrewDriver1337
<owrt-images-builds> Build [#9](https://buildbot.openwrt.org/images/#/builders/162/builds/9) of `main_ipq806x/generic` failed.
<owrt-images-builds> Build [#9](https://buildbot.openwrt.org/images/#/builders/123/builds/9) of `main_lantiq/xrx200` completed successfully.
_whitelogger_ has joined #openwrt-devel
Sawzall has joined #openwrt-devel
Sawzall has quit [Read error: Connection reset by peer]
Sawzall has joined #openwrt-devel
Sawzallz has quit [Ping timeout: 480 seconds]
_whitelogger_ has quit [Remote host closed the connection]
_whitelogger_ has joined #openwrt-devel
rsalvaterra has joined #openwrt-devel
rsalvaterra has quit []
rsalvaterra has joined #openwrt-devel
<hauke> The build creates the /run/apk/db.lock file. we should not add it to the image
<dwfreed> yeah, image build should include the contents of /run completely
Mangix has quit [Read error: Connection reset by peer]
Mangix has joined #openwrt-devel
_whitelogger_ has quit [Remote host closed the connection]
_whitelogger_ has joined #openwrt-devel
_whitelogger_ has quit [Remote host closed the connection]
<owrt-images-builds> Build [#9](https://buildbot.openwrt.org/images/#/builders/26/builds/9) of `main_qualcommax/ipq807x` completed successfully.
goliath has joined #openwrt-devel
Stat_headcrabbed has quit [Quit: Stat_headcrabbed]
<hauke> dwfreed: /run should be on tmp fs
<hauke> I wanted to create a symlink to /var/run
<dwfreed> hauke: sure, but during an image build, you can't exactly mount filesystems without being root :)
sorinello has quit [Ping timeout: 480 seconds]
<hauke> dwfreed: Just deleting it later should be fine, see https://github.com/openwrt/openwrt/pull/16961/files
robimarko has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
vincejv has quit [Remote host closed the connection]
<dwfreed> hauke: I just realized I typoed earlier, I meant *exclude* not *include*
<dwfreed> talk about a brainfart
Obi-Wan has quit [Quit: ZNC by prozac - http://znc.sourceforge.net]
Obi-Wan has joined #openwrt-devel
_whitelogger_ has joined #openwrt-devel
_whitelogger_ has quit [Remote host closed the connection]
_whitelogger_ has joined #openwrt-devel
_whitelogger_ has quit [Remote host closed the connection]
_whitelogger_ has joined #openwrt-devel