#archlinux32 | Logs for 2019-06-16

Back
[01:05:28] -!- eschwartz has quit [Ping timeout: 245 seconds]
[01:07:03] -!- ubuntuxp has quit [Quit: ERC (IRC client for Emacs 26.2)]
[01:12:09] -!- thePiGrepper has joined #archlinux32
[01:20:30] -!- eschwartz has joined #archlinux32
[01:51:28] -!- thePiGrepper has quit [Ping timeout: 272 seconds]
[01:55:13] -!- ofara has quit [Ping timeout: 244 seconds]
[01:57:47] -!- ofara_ has quit [Ping timeout: 248 seconds]
[02:10:29] <buildmaster> i686/rclone is broken (says rechenknecht).
[02:14:29] -!- thePiGrepper has joined #archlinux32
[02:19:32] <buildmaster> pentium4/rclone is broken (says eurobuild3).
[02:31:14] -!- ofara has joined #archlinux32
[02:59:12] -!- thePiGrepper has quit [Ping timeout: 245 seconds]
[03:05:52] -!- eschwartz has quit [Ping timeout: 245 seconds]
[03:16:24] -!- thePiGrepper has joined #archlinux32
[03:20:23] -!- eschwartz has joined #archlinux32
[04:09:22] -!- quequotion has joined #archlinux32
[05:05:21] -!- eschwartz has quit [Ping timeout: 244 seconds]
[05:19:16] -!- eschwartz has joined #archlinux32
[06:47:07] -!- thePiGrepper has quit [Ping timeout: 245 seconds]
[07:05:04] -!- guys has joined #archlinux32
[07:06:14] -!- eschwartz has quit [Ping timeout: 272 seconds]
[07:20:43] -!- guys has quit [Quit: A random quit message]
[07:21:25] -!- eschwartz has joined #archlinux32
[08:05:08] -!- titus_livius has joined #archlinux32
[08:06:37] <buildmaster> pentium4/libproxy is broken (says nlopc43).
[08:25:13] -!- ahuillet has parted #archlinux32
[09:06:40] -!- eschwartz has quit [Ping timeout: 248 seconds]
[09:21:05] -!- eschwartz has joined #archlinux32
[10:57:49] <buildmaster> i686/linux-pae is broken (says eurobuild3).
[11:05:56] -!- guys has joined #archlinux32
[11:06:54] -!- eschwartz has quit [Ping timeout: 272 seconds]
[11:31:23] -!- ofara_ has joined #archlinux32
[12:09:36] -!- ofara has quit [Ping timeout: 248 seconds]
[12:10:02] -!- ofara_ has quit [Ping timeout: 245 seconds]
[12:36:26] -!- thePiGrepper has joined #archlinux32
[13:06:53] -!- guys has quit [Ping timeout: 244 seconds]
[13:20:40] -!- guys has joined #archlinux32
[13:49:54] <buildmaster> i686/haskell-unicode-transforms are broken (says eurobuild3).
[13:58:12] <girls> abaumann: thinking about it, I'm not sure (anymore), that adding more regexes to the failure classification logic will be all we need to do to hav more fine-grained information about what the cause was - e.g. it doesn't make sense to check for haskell's "mismatched interface file versions ..." if the build did not fail in build() but in prepare() ...
[13:58:44] <girls> otoh it might be overkill to make the detection logic too restrictive ...
[14:18:29] <girls> elibrokeit: I had a look into how host_mirrors gets used and I think, the name is simply misleading - it /may/ be the host mirrors on archlinux, but it actually /should/ be the mirrors which should be used to install the chroot (e.g. in our case, we might have an x86_64-only mirror on the host which should not be used for the chroot)
[14:18:45] <girls> so editing that seems fine to me
[14:32:32] -!- abaumann has joined #archlinux32
[14:32:32] <buildmaster> Hi abaumann!
[14:32:32] <buildmaster> !rq abaumann
[14:32:33] <phrik> buildmaster: <abaumann> works. I receive my personal spam now ;-)
[14:32:45] <girls> Hi andreas!
[14:33:09] <abaumann> hi deep42thought :-)
[14:33:30] <girls> I should have stolen the power supply for the alix from you: now libproxy is built for i486 and I cannot test it - and my fingers itch ;-)
[14:33:57] <abaumann> oeh. the train has wifi and power, yeah. that would work. :-)
[14:34:03] <girls> :-D
[14:34:12] <girls> it would soon have multiple wifis ...
[14:34:13] <abaumann> sorry, didn't check, maybe I have a second one (like the router)
[14:34:45] <girls> lol
[14:34:54] <girls> "somewhere in this heap" ;-)
[14:34:58] <abaumann> didn't find it so far..
[14:35:35] -!- ofara has joined #archlinux32
[14:37:29] <abaumann> cool: I have a libproxy on my alix. :-)
[14:37:35] <girls> \o/
[14:37:51] <abaumann> mmh. I don't have any use for it..
[14:38:02] <abaumann> ..not yet.
[14:38:07] <girls> you could connect to my universities' vpn with it ;-)
[14:38:22] <girls> well, with openconnect
[14:39:03] <abaumann> cool :-)
[14:41:14] <girls> I'll put the find-umount-rm-rf stuff into archbuild now, hopefully this solves the delete-as-non-root issues
[14:42:20] <abaumann> ok. so I'll update my slaves afterwards with new devtools32
[14:42:47] <girls> hopefully this repairs more than it breaks :-D
[14:43:01] <abaumann> or it deletes slaves.. ;-)
[14:43:22] <girls> maybe they reproduce? and we suddently have some additional ones
[14:47:07] <girls> there is even a comment before the (failing) rm -rf: "avoid change of filesystem in case of an umount failure"
[14:47:28] <abaumann> mmh.. this helps..
[14:47:45] <girls> it would help if umount didn't fail in the first place ...
[14:48:33] <abaumann> we could check, if mounts are still active, and if yes, bail out.
[14:48:57] <girls> that would stall the build slave
[14:49:05] <abaumann> yeah.
[14:49:08] <girls> I'm for inserting a 'find ... umount' there
[14:51:09] <abaumann> findmnt
[14:51:14] <abaumann> nice. didn't know
[14:51:35] <abaumann> ah. needs root privilege
[14:51:48] <girls> we need root for umount anyways
[14:53:19] <abaumann> libseccomp, valgrind tests failing, unsupported architecture (for i486). I'll ignore failures for i486..
[14:55:53] -!- thePiGrepper has quit [Ping timeout: 245 seconds]
[14:56:49] -!- thePiGrepper has joined #archlinux32
[15:07:12] -!- guys has quit [Ping timeout: 248 seconds]
[15:07:48] <girls> I just checked: we _are_ root :-)
[15:07:54] <girls> new devtools32 are on their way
[15:08:56] <girls> done
[15:09:02] <abaumann> ok.
[15:09:21] <abaumann> I'll test on eurobuild1 first, if that one dies, no big harm done
[15:10:39] <girls> there should be no big change, actually
[15:10:49] <abaumann> ok
[15:22:00] -!- guys has joined #archlinux32
[15:24:24] -!- ofara_ has joined #archlinux32
[15:28:24] <abaumann> so, all three build slaves updates.
[15:28:30] <girls> :-)
[15:31:16] -!- quequotion has quit [Remote host closed the connection]
[15:41:55] <girls> I should really get back the "package x blocks n packages" info into the database or onto the web frontend
[15:46:52] <abaumann> that would be very handy..
[15:48:08] <girls> but the info is hard to create within mysql :-(
[15:48:17] <abaumann> yeah, I know.
[15:48:27] <girls> I'll think of something
[15:48:31] <abaumann> I have some ninja/C graph library hackery which could be used
[15:48:36] <abaumann> Not finished, as usual.
[15:48:37] <abaumann> :-)
[15:48:44] <girls> but first I need to change trains :-)
[15:48:53] <girls> ah you mean calculate outside mysql?
[15:48:55] <girls> hmmm
[15:48:57] <abaumann> yeah.
[15:49:02] <abaumann> not really elegant.
[15:49:07] <girls> I had code for that once ... in ... umm ... bash :-D
[15:49:09] <abaumann> changing trains is a good thing..
[15:49:13] <abaumann> :-)
[15:52:04] -!- AndrevS has joined #archlinux32
[15:53:57] <abaumann> haskell-base-compat-batteries.. ah.. Haskell comes with batteries included.. in constrast to my Arduino box..
[16:19:01] <girls> !grab abaumann
[16:19:01] <phrik> girls: Bingpot!
[16:19:35] <girls> isn't that a typo? shouldn't it be "haskell-base-combat-batteries"?
[16:35:26] <abaumann> !grab girls
[16:35:26] <phrik> abaumann: Tada!
[16:36:57] <elibrokeit> girls: are you saying you will revert to the previous behavior, or what?
[16:37:28] <elibrokeit> Originally, your fork would disable host_mirrors entirely
[16:37:41] <girls> oh, we did that once?
[16:38:04] <elibrokeit> Yes, I have a commit on my tree that came from you
[16:38:14] <elibrokeit> You rebased it into nothing. :)
[16:39:56] <girls> is there any scenario, that one needs the mirror from the host?
[16:54:12] -!- thePiGrepper has quit [Ping timeout: 245 seconds]
[16:54:58] -!- thePiGrepper has joined #archlinux32
[17:02:25] <elibrokeit> girls: it is used by archlinux internally to speed up operations because file:// is faster than https://, even if the latter is pointed at the same on-disk location but with a locally hosted web server sitting in between
[17:04:11] <girls> yes, but what operation is done in devtools which operates on the host rather than the chroot?
[17:07:32] -!- guys has quit [Ping timeout: 245 seconds]
[17:21:59] -!- guys has joined #archlinux32
[17:25:25] <elibrokeit> Bind mounting the CacheDir?
[17:26:14] <elibrokeit> It is a mirror that comes from the host rather than "somewhere in the internet"
[17:27:43] <elibrokeit> Actually it mounts the whole directory hierarchy IIRC
[17:28:13] <elibrokeit> It is of course perfectly reasonable to think that if a file:// mirror is in mirrorlist33 you could use it the same way :)
[17:28:45] <elibrokeit> ... you'd just need to have pacman-conf report the right thing via a forked host conf
[17:36:57] <girls> we take a host from /etc/pacman.d/mirrorlist32 IIRC
[17:38:08] <abaumann> ah.. soon there will be a php for i486 (a working one, I mean) :-)
[17:39:33] <girls> then you move your web server to the alix?
[17:39:53] <abaumann> no.. the archlinux32 mirror. ;-)
[17:40:05] <abaumann> needs hell of a compact flash card..
[17:40:41] <abaumann> my personal webpage, yeah. could be an option to move it..
[17:41:32] <girls> don't blame for my bad input me if it breaks ;-)
[17:42:05] <abaumann> :-)
[18:02:04] <abaumann> PHP Version 7.3.6, Linux euroalix 5.1.5-arch1-2.0-ARCH #1 SMP PREEMPT Tue May 28 13:29:01 CEST 2019 i586, Host i486-pc-linux-gnu
[18:04:59] <elibrokeit> girls: anyway the point is that using pacman-conf --config "$working_dir/etc/pacman.conf" --repo extra Server, will get you the list of mirrors known by the chroot, at which point you will... read those mirrors from the guest, then write it back to the guest? I'm not sure what the point there is.
[18:07:43] <girls> yes that's bad
[18:07:45] <girls> hmm
[18:08:06] <girls> can you point me to a line in the code where we do that? I'm unable to find it
[18:08:34] <girls> (but I'm oing offline in a few minutes anyways, as I have to change trains again - and the next one does not have wifi)
[18:17:03] <elibrokeit> girls: https://git.archlinux32.org
[18:17:04] <phrik> Title: arch-nspawn should use the correct pacman config file · bd53fd6e9a - Archlinux32 Gitea (at git.archlinux32.org)
[18:17:23] <elibrokeit> -host_mirrors=($($pacconf_cmd --repo extra Server 2> /dev/null | sed -r 's#(.*/)extra/os/.*#\1$repo/os/$arch#'))
[18:17:23] <elibrokeit> +host_mirrors=($($pacconf_cmd --conf "$pac_conf" --repo extra Server 2> /dev/null | sed -r 's#(.*/)extra/os/.*#\1$repo/os/$arch#'))
[18:52:07] -!- ofara has quit [Ping timeout: 245 seconds]
[18:52:24] -!- ofara_ has quit [Ping timeout: 272 seconds]
[18:53:14] -!- ofara has joined #archlinux32
[18:53:26] -!- ofara_ has joined #archlinux32
[19:07:32] -!- guys has quit [Ping timeout: 245 seconds]
[19:20:40] -!- guys has joined #archlinux32
[19:53:53] -!- abaumann has quit [Quit: leaving]
[20:00:52] -!- thePiGrepper has quit [Ping timeout: 245 seconds]
[20:08:58] <buildmaster> pentium4/llpp is broken (says rechenknecht).
[20:18:58] <buildmaster> i686/llpp is broken (says rechenknecht).
[20:50:37] -!- MrBIOS has joined #archlinux32
[21:07:57] -!- guys has quit [Ping timeout: 245 seconds]
[21:21:57] -!- guys has joined #archlinux32
[22:17:19] -!- samantaz has joined #archlinux32
[22:40:21] <buildmaster> i686/go is broken (says eurobuild3).
[23:10:10] -!- guys has quit [Ping timeout: 272 seconds]
[23:12:46] -!- guys has joined #archlinux32