#archlinux32 | Logs for 2019-09-19
Back
[00:03:32] -!- yans has quit [Ping timeout: 276 seconds]
[00:10:45] <nit-picker> abaumann: your slave eurobuild6-3 builds i686/thunderbird for more than a day, now (5 day(s) 15:06:26)
[01:39:23] <buildmaster> i486/freeglut is broken (says eurobuild6-7-i486): https://archlinux32.org
[02:01:36] -!- thePiGrepper has quit [Ping timeout: 268 seconds]
[02:08:38] -!- MrBIOS has quit [Quit: MrBIOS]
[02:19:42] -!- thePiGrepper has joined #archlinux32
[02:24:29] <buildmaster> i686/firefox is broken (says eurobuild6-6): https://archlinux32.org
[02:48:43] -!- thePiGrepper has quit [Ping timeout: 258 seconds]
[03:00:28] <nit-picker> abaumann: your slave eurobuild6-3 builds i686/thunderbird for more than a day, now (5 day(s) 17:56:08)
[03:03:52] -!- DCyrax has quit [Remote host closed the connection]
[03:04:13] -!- DCyrax has joined #archlinux32
[03:37:59] <buildmaster> pentium4/firefox is broken (says nlopc46): https://archlinux32.org
[05:36:06] <nit-picker> abaumann: your slave eurobuild6-3 builds i686/thunderbird for more than a day, now (5 day(s) 20:31:47)
[05:39:10] -!- thePiGrepper has joined #archlinux32
[06:12:55] <nit-picker> abaumann: your slave eurobuild6-3 builds i686/thunderbird for more than a day, now (5 day(s) 21:08:36)
[06:24:13] -!- nit-picker has quit [Remote host closed the connection]
[06:24:30] -!- nit-picker has joined #archlinux32
[06:24:30] <buildmaster> Hi nit-picker!
[06:24:30] <buildmaster> !rq nit-picker
[06:24:31] <phrik> buildmaster: <nit-picker> abaumann: but there are only 18 "intermediate" versions
[07:21:39] -!- thePiGrepper has quit [Ping timeout: 268 seconds]
[07:32:52] <buildmaster> i686/android-tools are broken (says buildknecht2): https://archlinux32.org
[07:43:05] <buildmaster> pentium4/android-tools are broken (says buildknecht): https://archlinux32.org
[07:48:57] <nit-picker> abaumann: your slave eurobuild6-3 builds i686/thunderbird for more than a day, now (5 day(s) 22:44:38)
[08:14:19] -!- thePiGrepper has joined #archlinux32
[08:22:42] -!- eschwartz has quit [Ping timeout: 246 seconds]
[08:24:56] -!- eschwartz has joined #archlinux32
[08:41:06] -!- deep42thought has joined #archlinux32
[08:41:06] <buildmaster> Hi deep42thought!
[08:41:06] <buildmaster> !rq deep42thought
[08:41:07] <phrik> buildmaster: <deep42thought> db-update --fuck-up
[09:17:42] -!- infides has joined #archlinux32
[09:24:13] -!- abaumann has joined #archlinux32
[09:24:13] <buildmaster> Hi abaumann!
[09:24:13] <buildmaster> !rq abaumann
[09:24:14] <phrik> buildmaster: <abaumann> nobody wrote an operating system in python yet?
[09:24:22] <deep42thought> Hi abaumann!
[09:24:28] <abaumann> hi deep42thought :-)
[09:24:43] <abaumann> nice new feature of the nit-picker :-)
[09:24:55] <deep42thought> yes
[09:25:16] <deep42thought> in the meantime, I managed to make him complain only once per package and day
[09:25:29] <abaumann> any clue, why we are stuck at 2555 pending builds?
[09:25:44] <deep42thought> not really
[09:26:09] <deep42thought> probably some packages which should be built now, do not build
[09:26:17] <abaumann> there are some blockers, mainly gettext, gnomish stuff
[09:27:06] <abaumann> this gettext version problem is annoying and I don't like to open 100nds of bugs upstream
[09:27:17] <abaumann> it happens on every release of gettext
[09:27:24] <deep42thought> hmm?
[09:27:28] <deep42thought> what's the issue?
[09:27:46] <abaumann> some autoconf stuff and gettext stuff mismatch, you have to re-gettextize the package
[09:27:58] * abaumann tries to find a bug report
[09:29:09] <abaumann> FS#63756
[09:29:18] <abaumann> https://bugs.archlinux.org
[09:29:18] <phrik> Title: FS#63756 : [libnsl] gettext 0.19/0.20 autoconf mismatch (at bugs.archlinux.org)
[09:29:43] <abaumann> upstream should go through all gettext related packages and fix it.
[09:30:08] <abaumann> alternatively not update gettext and just hope everything builds, but den you have to add gettext19, gettext20 etc.
[09:32:57] <deep42thought> yes, sounds to me, like `gettextize -f` is the right thing to do
[09:33:48] <abaumann> ok. let me see if I can unblock some more packages this way..
[09:34:38] <deep42thought> if this is common enough (say > 50 packages), then it might be worth adding this as another straw
[09:35:13] <abaumann> yeah, let's try manual first, as for CARCH -> i686 and count then. :-)
[09:35:18] <deep42thought> yes
[09:35:33] <deep42thought> for some reason, ghc is not being handed out ...
[09:35:38] <deep42thought> I'll have to take a look, why
[09:35:41] <abaumann> ok
[09:35:50] <deep42thought> this also blocks a load of packages
[09:36:33] <abaumann> Later I'll hunt the sse2 crashes (SIGILL) in newsboat, I don't fully understand, how it happens.
[09:38:49] <abaumann> oh. and fix my email reporting :-)
[09:42:41] -!- infides has quit [Ping timeout: 276 seconds]
[09:53:24] <deep42thought> hc makedepends on haskell-hscolour
[09:53:59] <abaumann> cool.
[09:53:59] <deep42thought> and apparently, our dependency database is currently so broken, that the buildmaster believes, haskell-hscolour is not available, currently and must be built first
[09:55:42] <deep42thought> I'll force it. hopefully, I'm right, and it will simply build :-)
[09:55:58] <abaumann> yeah, that's what I'm always doing :-)
[09:56:07] <deep42thought> forcing or hoping?
[09:56:11] <abaumann> both :-)
[09:56:14] <deep42thought> :-D
[09:56:19] <abaumann> in that order..
[09:56:23] <deep42thought> lol
[09:59:26] <abaumann> ah. the email reporter emails only packages from the pacman cache..
[09:59:33] <abaumann> ..bad if you cleaned up before..
[10:00:11] <abaumann> gpg: error retrieving 'buildmaster@archlinux32.org' via WKD: General error
[10:00:16] <abaumann> uh. that's new.
[10:00:26] <deep42thought> err
[10:00:32] <deep42thought> that must be on my end :-/
[10:00:42] <abaumann> mmh.
[10:00:53] <abaumann> I'm not sure which keyserver I'm accessing on the test vms
[10:01:09] <deep42thought> wkd
[10:02:28] <abaumann> mmh. I had network issues around libvirt bridges, so could also be that..
[10:02:34] <deep42thought> nope
[10:02:40] <deep42thought> I have the issue over here, too
[10:03:12] <abaumann> ah
[10:03:17] <abaumann> maybe the key expired?
[10:04:12] <deep42thought> I guess, something is wrong with the http redirection stuff
[10:04:24] <deep42thought> though, a expired key might explain it, too
[10:08:26] <deep42thought> key is not expired
[10:08:36] <deep42thought> but I do not see any http requests to that key, either
[10:08:43] <deep42thought> so some redirection does not work as intended
[10:10:41] <abaumann> 4096 bit RSA key C963C94DD204443D, created: 2017-08-29, expires: 2019-08-29 (expired)
[10:10:44] <abaumann> 4096 bit RSA key 714A1770ECE2DF62, created: 2017-06-23, expires: 2019-06-23 (expired)
[10:10:51] <abaumann> in my local keystore.
[10:10:52] <abaumann> yep
[10:10:53] <deep42thought> yes
[10:10:58] <deep42thought> but not on the wkd server
[10:11:09] <abaumann> hope so :-)
[10:12:21] <deep42thought> not even the server running on archlinux32.org sees the wkd request O.o
[10:13:10] <abaumann> "Press Return to acknowledge the previous three paragraphs.
[10:13:18] <abaumann> seriously. in a build script?
[10:13:31] <abaumann> /usr/bin/gettextize: line 1309: /dev/tty: No such device or address
[10:13:35] <abaumann> *sigh*
[10:13:41] <deep42thought> </dev/null
[10:14:02] <deep42thought> ln -sf /dev/null /dev/tty
[10:14:08] <abaumann> gettextize doesn't work if the package is not maintained
[10:14:10] <abaumann> sure :-)
[10:14:22] <deep42thought> what do you mean by "maintained"?
[10:14:41] <abaumann> autoconf stuff is regularly updated to modern versions.
[10:14:50] <deep42thought> ah, ok
[10:16:00] <deep42thought> you seem to have discovered a bug in gpg
[10:16:16] <deep42thought> `curl -Ss 'https://archlinux32.org/.well-known/openpgpkey/hu/z4eyw18p7a9p7c9owm78fj93mqkks6q3' | gpg --import` works as advertised
[10:16:34] <deep42thought> but `gpg --auto-key-locate wkd --locate-key buildmaster@archlinux32.org` does not
[10:16:45] <abaumann> gpg uses http/3?
[10:16:54] <abaumann> ;-)
[10:16:56] <deep42thought> the server uses http2
[10:17:00] <abaumann> oh.
[10:17:01] <deep42thought> maybe this makes gpg hickup?
[10:17:22] <abaumann> it tries https and is forced into a http/2 connection maybe
[10:17:31] <deep42thought> probably
[10:17:46] * abaumann digs how http accesses are implemented in gpg
[10:18:13] <deep42thought> yes
[10:18:15] <deep42thought> http works
[10:18:18] <deep42thought> http2 does not
[10:18:39] * abaumann has personal hate feelings for HTTP protocols above 1.1
[10:18:58] <deep42thought> announcement: due to gpg wkd issues, archlinux32.org is now running without http2
[10:19:07] <abaumann> lol
[10:23:13] <nit-picker> abaumann: your slave eurobuild6-3 builds i686/thunderbird for more than a day, now (6 day(s) 01:18:54)
[10:23:57] <abaumann> nit-picker: tell Mozilla :->
[10:25:48] <nit-picker> Mozilla: your software is crappy - it builds for 6 days now and an end is not in sight
[10:25:49] -!- infides has joined #archlinux32
[10:31:12] <abaumann> ok. the generic straw for gettextize will not work, gettextize on libgnomekbd fails royally
[10:32:09] <deep42thought> ok, no straw, then
[10:39:16] <abaumann> gpg: signing failed: Inappropriate ioctl for device
[10:39:17] <abaumann> mmh
[10:39:26] <abaumann> but, it obviously worked
[10:39:31] <deep42thought> O.o
[10:39:47] <deep42thought> do you need to unlock your signing key?
[10:40:23] <abaumann> could be
[10:40:33] <abaumann> gpg: [stdin]: sign+encrypt failed: Inappropriate ioctl for device
[10:40:35] <abaumann> yeah
[10:40:46] <abaumann> the error message is, to be mild... b*
[10:41:38] <deep42thought> well, actually, it is pretty detailed
[10:41:57] <deep42thought> the /dev/stdin is not appropriate to enter your passphrase
[10:42:11] <abaumann> for techies, yes - which I'm not :-)
[10:42:26] <deep42thought> ok, maybe "detailed" was the wrong expression
[10:42:32] <abaumann> :-)
[10:42:43] <deep42thought> still better than the "general error", which we got with wkd+http2
[10:43:40] <abaumann> I have the strong suspision looking ad the wkd code, that a) they implemented their own http protocol and b) it is not http/2 capable
[10:43:45] <abaumann> they could use libcurl :-)
[10:43:55] <deep42thought> I wrote upstream
[10:44:02] <deep42thought> let's see, what comes up :-)
[10:44:05] <abaumann> ah. good.
[10:44:23] <deep42thought> though, my mail is still stuck, hmmm
[10:44:27] <abaumann> well. if the authors didn't implement it, there is not much upstream can do
[10:44:48] <deep42thought> "upstream" = "gnupg-devel@gnupg.org"
[10:45:01] <deep42thought> sry for the misleading terminology
[10:45:08] <abaumann> ah.
[10:45:18] <abaumann> upstream and upupstream is also confusing :-)
[10:45:24] <deep42thought> upperstream
[10:45:30] <deep42thought> upstream vs Upstream
[10:45:33] <deep42thought> hmmm
[10:50:24] <abaumann> gpg starts to tap on my nerves. edit key, passwd empty, then save
[10:50:26] <abaumann> same error
[10:51:01] <deep42thought> can you sign something directly
[10:51:05] <abaumann> I cannot set an empty password
[10:51:07] <deep42thought> maybe it's my script's fault?
[10:51:36] <abaumann> gpg doesn't allow to set an empty password.. nice security feature
[10:51:48] <abaumann> mmh. maybe there is another way to elimnate the password
[10:52:07] <deep42thought> maybe, we can get report-installed-packages to work with a key that has a password?
[10:52:13] <deep42thought> seems the cleaner way
[10:53:28] <abaumann> Or I generate a new key
[10:53:40] <abaumann> I fear anyway, that I used the master key for signing those emails..
[10:53:46] <deep42thought> :-D
[11:01:36] -!- infides has quit [Ping timeout: 246 seconds]
[11:07:07] <abaumann> https://bugs.archlinux.org
[11:07:08] <phrik> Title: FS#63805 : [doxygen] kdeclarative fails to install (at bugs.archlinux.org)
[11:07:16] <abaumann> "
[11:07:17] <abaumann> Repo packages are built with 1.8.15 for the time being."
[11:07:29] <abaumann> this is fine for people building packages by hand (aka upstream)
[11:07:38] <deep42thought> shim package?
[11:07:42] <abaumann> I could make a shim package doxygen1815
[11:07:45] <abaumann> exactly
[11:07:55] <deep42thought> this is, what [build-support] is *designed* for :-)
[11:07:55] <abaumann> and patch accordingly in the makedepends
[11:08:01] <deep42thought> nono
[11:08:06] <abaumann> ah, you tell me that now? ;-)
[11:08:08] <deep42thought> just make it provide doxygen
[11:08:15] <abaumann> right
[11:08:42] <abaumann> otoh you only want to build broken packages against 1.8.15, maybe others will break using a too old version of doxygen
[11:08:55] <abaumann> ototh. the straw mechanism should take care of that
[11:09:01] <abaumann> build-support is the last straw, I think
[11:09:06] <abaumann> good :-)
[11:09:13] <deep42thought> not *the* last, but a pretty late one
[11:09:23] <deep42thought> I think
[11:09:23] <abaumann> almost by design.. ;-)
[11:10:07] <abaumann> the drawback is, we don't find other packages with the same issue and cannot report upstream
[11:10:20] <deep42thought> we're not ci for upstream
[11:10:27] <abaumann> ah, not? ;-)
[11:10:41] <deep42thought> well, it feels that way
[11:10:45] <deep42thought> but actually, we're not
[11:10:46] <abaumann> :-)
[11:15:38] <deep42thought> maybe, there's an archlinuxci out there? I guess, the reproducible-builds people are pushing forward on that front :-D
[11:17:24] <abaumann> https://tests.reproducible-builds.org
[11:17:26] <phrik> Title: Reproducible archlinux ?! (at tests.reproducible-builds.org)
[11:22:34] <deep42thought> well, the graph tells me, they're doing ci for arch - and ~9% of packages ftbfs :-D
[12:14:05] * abaumann dashes away for lunch..
[12:14:06] -!- abaumann has quit [Quit: leaving]
[12:36:23] -!- thePiGrepper has quit [Ping timeout: 265 seconds]
[12:44:14] <nit-picker> abaumann: your slave eurobuild6-3 builds i686/thunderbird for more than a day, now (6 day(s) 03:39:55)
[13:11:33] <buildmaster> pentium4/haskell-hspec-expectations are broken (says nlopc46) - I rescheduled: haskell-hunit: https://archlinux32.org
[13:13:32] <buildmaster> pentium4/haskell-fclabels are broken (says buildknecht2): https://archlinux32.org
[13:15:32] -!- buildmaster has quit [Remote host closed the connection]
[13:15:44] -!- buildmaster has joined #archlinux32
[13:15:45] <buildmaster> !rq buildmaster
[13:15:45] <phrik> buildmaster: <buildmaster> I might be insane, but never confused ... ;-)
[13:19:20] <buildmaster> pentium4/haskell-cmark-gfm is broken (says eurobuild1): https://archlinux32.org
[13:26:05] <buildmaster> i486/capi4hylafax is broken (says nlopc46-i486bs0): https://archlinux32.org
[13:30:38] <buildmaster> i486/wkhtmltopdf is broken (says nlopc46-i486bs0): https://archlinux32.org
[13:36:28] <buildmaster> i486/zeal is broken (says nlopc46-i486bs0): https://archlinux32.org
[13:42:17] <buildmaster> i486/chinese-calendar is broken (says nlopc46-i486bs0): https://archlinux32.org
[13:46:52] <buildmaster> i486/texmaker is broken (says nlopc46-i486bs0): https://archlinux32.org
[13:52:54] <buildmaster> i686/haskell-glib is broken (says buildknecht2): https://archlinux32.org
[13:54:02] <buildmaster> i486/libfilezilla is broken (says nlopc46-i486bs0): https://archlinux32.org
[13:59:24] <buildmaster> i486/gsettings-qt is broken (says nlopc46-i486bs0): https://archlinux32.org
[14:03:13] <buildmaster> pentium4/haskell-glib is broken (says eurobuild1): https://archlinux32.org
[14:03:39] <buildmaster> i486/kchmviewer is broken (says nlopc46-i486bs0): https://archlinux32.org
[14:09:33] <buildmaster> i486/qml-box2d is broken (says nlopc46-i486bs0): https://archlinux32.org
[14:16:33] <buildmaster> i486/qoauth is broken (says nlopc46-i486bs0): https://archlinux32.org
[14:20:30] -!- MrBIOS has joined #archlinux32
[14:22:15] <buildmaster> i486/speedcrunch is broken (says nlopc46-i486bs0): https://archlinux32.org
[14:26:37] <buildmaster> i486/qmltermwidget is broken (says nlopc46-i486bs0): https://archlinux32.org
[14:32:48] -!- thePiGrepper has joined #archlinux32
[14:33:37] <buildmaster> i486/qterminal is broken (says nlopc46-i486bs0): https://archlinux32.org
[14:37:56] <buildmaster> i486/recoll is broken (says nlopc46-i486bs0): https://archlinux32.org
[14:42:31] <nit-picker> abaumann: your slave eurobuild6-3 builds i686/thunderbird for more than a day, now (6 day(s) 05:38:12)
[14:42:52] <buildmaster> i486/qbs are broken (says nlopc46-i486bs0): https://archlinux32.org
[14:50:35] <buildmaster> i486/python-pyotherside is broken (says nlopc46-i486bs0): https://archlinux32.org
[14:55:52] <buildmaster> i486/qt5pas are broken (says nlopc46-i486bs0): https://archlinux32.org
[15:01:14] -!- MrBIOS has quit [Quit: MrBIOS]
[15:01:53] <buildmaster> i486/drumstick is broken (says nlopc46-i486bs0): https://archlinux32.org
[15:14:59] <buildmaster> i486/faudio is broken (says nlopc46-i486bs0): https://archlinux32.org
[15:19:51] <buildmaster> i486/kqtquickcharts are broken (says nlopc46-i486bs0): https://archlinux32.org
[15:24:20] <nit-picker> abaumann: your slave eurobuild6-5 builds pentium4/sagemath for more than a day, now (1 day(s) 03:12:55)
[15:49:21] -!- abaumann has joined #archlinux32
[15:49:21] <buildmaster> Hi abaumann!
[15:49:21] <buildmaster> !rq abaumann
[15:49:22] <phrik> buildmaster: <abaumann> it usually automatically fails. :-)
[15:59:43] <buildmaster> i486/mono is broken (says nlopc46-i486bs0): https://archlinux32.org
[16:02:38] <buildmaster> pentium4/switchboard-plug-keyboard is broken (says eurobuild6-6): https://archlinux32.org
[16:03:42] <buildmaster> i686/switchboard-plug-keyboard is broken (says buildknecht2): https://archlinux32.org
[16:03:51] -!- deep42thought has quit [Quit: Leaving.]
[16:05:02] <buildmaster> i486/qt5-location is broken (says nlopc46-i486bs0): https://archlinux32.org
[16:13:03] <buildmaster> i486/qt5-webchannel is broken (says nlopc46-i486bs0): https://archlinux32.org
[16:15:44] <buildmaster> i486/qt5-3d is broken (says eurobuild6-7-i486): https://archlinux32.org
[16:18:10] <buildmaster> i486/qt5-graphicaleffects are broken (says nlopc46-i486bs0): https://archlinux32.org
[16:19:41] <buildmaster> i486/qt5-quickcontrols are broken (says eurobuild6-7-i486): https://archlinux32.org
[16:25:05] <buildmaster> i486/qt5-quickcontrols2 is broken (says nlopc46-i486bs0): https://archlinux32.org
[16:27:30] <buildmaster> i486/qt5-scxml is broken (says eurobuild6-7-i486): https://archlinux32.org
[16:30:27] <buildmaster> i486/qt5-serialbus are broken (says nlopc46-i486bs0): https://archlinux32.org
[16:33:18] <buildmaster> i486/qt5-virtualkeyboard is broken (says eurobuild6-7-i486): https://archlinux32.org
[16:34:49] <buildmaster> i486/qt5-wayland is broken (says nlopc46-i486bs0): https://archlinux32.org
[16:36:18] <nit-picker> abaumann: your slave eurobuild6-3 builds i686/thunderbird for more than a day, now (6 day(s) 07:31:59)
[16:36:48] <buildmaster> i486/qt5-webglplugin is broken (says eurobuild6-7-i486): https://archlinux32.org
[16:37:55] <buildmaster> pentium4/haskell-gnuidn is broken (says eurobuild6-6): https://archlinux32.org
[16:39:21] <buildmaster> i486/qcad is broken (says eurobuild6-7-i486): https://archlinux32.org
[16:41:21] -!- buildmaster has quit [Remote host closed the connection]
[16:41:33] -!- buildmaster has joined #archlinux32
[16:41:33] <buildmaster> !rq buildmaster
[16:41:34] <phrik> buildmaster: <buildmaster> I might be insane, but never confused ... ;-)
[16:42:04] <buildmaster> i486/libqtxdg is broken (says nlopc46-i486bs0): https://archlinux32.org
[16:48:37] <abaumann> forced builds never end up on the slaves I force them to.
[16:52:06] <buildmaster> any/min is broken (says nlopc46-i486bs0): https://archlinux32.org
[16:54:53] <abaumann> Houston, we have a kdeclarative :-)
[16:55:04] <buildmaster> i686/kdeclarative is broken (says nlopc46): https://archlinux32.org
[16:56:43] <abaumann> ok. I take my statement back: the forced packages do end up on the right slave in the end.
[16:56:58] <abaumann> but there seems to be a race that they can also be picked up before by another free slave
[17:00:44] -!- MrBIOS has joined #archlinux32
[17:05:49] -!- MrBIOS_ has joined #archlinux32
[17:07:07] -!- MrBIOS has quit [Ping timeout: 245 seconds]
[17:07:07] MrBIOS_ is now known as MrBIOS
[17:16:27] <nit-picker> abaumann: your slave eurobuild6-5 builds pentium4/sagemath for more than a day, now (1 day(s) 05:05:02)
[17:27:03] <abaumann> sagemath hangs with 100nds of python2 tests. cool.
[17:29:15] <buildmaster> i686/thunderbird is broken (says eurobuild6-3): https://archlinux32.org
[17:40:53] -!- MrBIOS has quit [Read error: Connection reset by peer]
[17:42:47] -!- deep42thought has joined #archlinux32
[17:42:48] <buildmaster> Hi deep42thought!
[17:42:48] <buildmaster> !rq deep42thought
[17:42:49] <phrik> buildmaster: <deep42thought> because you're a pessimist with no hope - or you're a realist afraid of change ... pick your evil
[17:43:15] <deep42thought> abaumann: I consider this correct behaviour: forcing a package to slave x should not prevent slave y from getting it if it asks first
[17:43:54] <abaumann> ah ok. :-)
[17:44:05] <abaumann> then it's a feature, not a bug. ;-)
[17:51:05] -!- ofara has quit [Remote host closed the connection]
[17:53:09] -!- ofara has joined #archlinux32
[17:56:02] <buildmaster> i686/haskell-lucid is broken (says eurobuild6-3): https://archlinux32.org
[18:05:42] -!- MrBIOS has joined #archlinux32
[18:09:44] <eschwartz> abaumann: I'm curious how you managed to clone gtk-engine
[18:11:44] <abaumann> eschwartz: I just entered my username and password every time the build slave was asking for it. :-)
[18:12:03] <deep42thought> you did *what*??
[18:12:03] <eschwartz> abaumann: but, the repository was deleted from github...
[18:12:09] <abaumann> mmh..
[18:12:15] <abaumann> that's interesting indeed..
[18:12:18] <abaumann> a cache somewhere?
[18:12:20] <eschwartz> it's in gitlab.gnome.org/Archive/
[18:12:32] <eschwartz> Upstream killed it off.
[18:12:34] <abaumann> deep42thought might have added a Archlinux32 cache for git checkouts.
[18:12:41] <abaumann> so the build slave fetches it from there.
[18:12:44] <deep42thought> I don't think so
[18:12:49] <abaumann> mmh.
[18:12:59] <abaumann> then I don't know how I managed it. :-)
[18:13:00] <deep42thought> https://archive-server.archlinux32.org
[18:13:11] <deep42thought> there you can see, what we have mirrored
[18:13:32] <abaumann> mmh. indeed.
[18:13:40] * abaumann shrugs
[18:13:47] <deep42thought> you're magic :-)
[18:14:13] <deep42thought> git clone repo-I-just-thought-about
[18:14:33] <abaumann> dolphin genes, I have the sixt sense. ;-)
[18:14:44] <eschwartz> I wonder, again and again, if it makes sense to have makepkg support a MIRRORURL that can download tarballs if sources are 404
[18:15:13] <deep42thought> crux has that :-)
[18:15:19] <eschwartz> so does gentoo
[18:15:41] <deep42thought> and arch is derived from crux (in terms of PKGBUILDs) - and more convenient, too
[18:15:50] <abaumann> otoh if more and more authors _know_ that distributions do that, the sluggishness in releasing tarballs will increase
[18:16:12] <deep42thought> well, we do have a source mirror, too
[18:16:22] <deep42thought> https://sources.archlinux32.org
[18:16:22] <phrik> Title: Index of / (at sources.archlinux32.org)
[18:16:34] <abaumann> yeah, but it acts more like a cache.
[18:16:39] <deep42thought> yes
[18:16:51] <deep42thought> eschwartz' thing would do something similar, wouldn't it?
[18:17:48] * abaumann sees why newsboat has SSE2 in it. it builds a rust (cough) part and the one in rust-bin contains sse2 opcodes.
[18:19:06] <deep42thought> I wonder which infrastructure we're going to set up to iron out flaws in foreign stuff - so far we have a tar ball server, a source mirror, a key server ...
[18:19:23] <eschwartz> allan wants to eventually have reponame.src.tar.gz repositories to go with reponame.db and reponame.files
[18:19:42] <deep42thought> this would be a huge file
[18:19:45] <deep42thought> wouldn't it?
[18:20:01] <eschwartz> So, we will eventually end up mirroring all sources, not just the licensing ones -- which means we don't need to make sources.archlinux.org have --allsource copies
[18:20:23] <eschwartz> no, like db.tar.gz it would have links to the *.src.tar.gz files
[18:20:28] <deep42thought> this would roughly double the size of mirrors, right?
[18:20:33] <deep42thought> ah, ok
[18:20:35] <deep42thought> :-)
[18:20:56] <deep42thought> and those src.tar.gz would reside where?
[18:21:03] <eschwartz> and you wouldn't need to download a .srcdb, just like you don't need to download a .files
[18:21:37] <eschwartz> They would reside, I suppose, in mirror.example.com/archlinux/$repo/os/$arch/
[18:21:37] <deep42thought> if it's only a db of sources, the size is negligible
[18:21:48] <deep42thought> hmm
[18:21:54] <deep42thought> anyways: dinner-time!
[18:22:05] <abaumann> have a nice dinner :-)
[18:22:26] <eschwartz> the idea is that you could do pacman --sources and download a source package
[18:22:44] <eschwartz> (obviously generated with makepkg --source, not makepkg --allsource)
[18:23:35] <eschwartz> So, if we get that, then we no longer need to generate --allsource tarballs for sources.archlinux.org, and we might want to instead just host a SRCDEST cache as a mirror
[18:23:55] <abaumann> sounds good :-)
[18:44:30] * abaumann dashes to a dinner
[18:44:31] -!- abaumann has quit [Quit: leaving]
[18:59:13] -!- deep42thought has quit [Quit: Leaving.]
[19:04:28] -!- eschwartz has quit [Ping timeout: 245 seconds]
[19:27:26] <buildmaster> pentium4/haskell-httpd-shed is broken (says eurobuild6-4) - I rescheduled: haskell-network-bsd: https://archlinux32.org
[19:57:58] -!- MrBIOS has quit [Ping timeout: 258 seconds]
[20:04:49] -!- MrBIOS has joined #archlinux32
[20:20:54] <nit-picker> abaumann: your slave eurobuild6-5 builds pentium4/sagemath for more than a day, now (1 day(s) 08:09:29)
[20:36:12] -!- lodger has joined #archlinux32
[20:37:50] -!- lodger has quit [Client Quit]
[20:52:35] -!- MrBIOS has quit [Read error: Connection reset by peer]
[20:54:06] -!- MrBIOS has joined #archlinux32
[20:59:03] -!- MrBIOS has quit [Ping timeout: 265 seconds]
[20:59:55] -!- skydrome has quit [Ping timeout: 246 seconds]
[21:02:04] -!- MrBIOS has joined #archlinux32
[21:04:53] -!- eschwartz has joined #archlinux32
[21:06:45] -!- MrBIOS_ has joined #archlinux32
[21:08:26] -!- MrBIOS has quit [Ping timeout: 276 seconds]
[21:08:26] MrBIOS_ is now known as MrBIOS
[21:08:46] -!- skydrome has joined #archlinux32
[21:24:29] -!- infides has joined #archlinux32
[22:27:27] <buildmaster> pentium4/haskell-sbv is broken (says eurobuild6-3): https://archlinux32.org
[22:33:19] <buildmaster> i686/haskell-sbv is broken (says eurobuild6-1): https://archlinux32.org
[22:35:53] <buildmaster> i686/hedgewars are broken (says eurobuild6-6): https://archlinux32.org
[22:45:48] <nit-picker> abaumann: your slave eurobuild6-5 builds pentium4/sagemath for more than a day, now (1 day(s) 10:34:24)
[22:52:34] <buildmaster> pentium4/hedgewars are broken (says eurobuild6-1): https://archlinux32.org
[22:55:32] -!- infides has quit [Ping timeout: 265 seconds]
[23:27:58] <buildmaster> i686/haskell-http2 is broken (says nlopc46) - I rescheduled: haskell-time-manager: https://archlinux32.org
[23:30:02] -!- thePiGrepper has quit [Ping timeout: 245 seconds]
[23:30:54] -!- thePiGrepper has joined #archlinux32
[23:37:53] <buildmaster> i686/shellcheck is broken (says eurobuild6-6): https://archlinux32.org
[23:42:54] -!- thePiGrepper has quit [Ping timeout: 265 seconds]