bug#29537: Core updates broken

[Gábor Boskovits]

 8dd35a0e2 is good

2017-12-02 21:12 GMT+01:00 Gábor Boskovits <address@hidden>:

It seems, that we have a breakage in current core-updates. m4, gettext, and at least a few other packages fail to build.

We had a discussion about that on irc, here are the important points:

[09:34:53]	* g_bor has joined #guix
[09:34:58]	<g_bor>	Hello guix!
[09:37:37]	<g_bor>	I just tried to build something no core-updates and m4 and python-boot0 does not build, I get a message that a decoding error occured in exception dispatcher. It seem like it's locale related, I'm on hu_HU.UTF-8
[09:38:03]	<g_bor>	I'm tring this now on master to see if the problem persists.
[09:38:10]	<g_bor>	Anyone seen that?

[16:46:05]	<civodul>	mb[m]1: you were right about the failing installation tests
[16:46:16]	<civodul>	it did work a couple of days ago though, so there's hope
[16:47:16]	<mb[m]1>	civodul: Any idea what caused it? Haven't been paying attention lately.
[16:47:24]	* kristofer has joined #guix
[16:53:03]	<civodul>	mb[m]1: no idea yet, let's see
[16:54:11]	<mb[m]1>	I don't understand why core-updates is failing. I've bisected glibc down to the first two commits on the 2.26 branch, and also tried reverting the libatomic-ops update.
[16:54:35]	<civodul>	how's it failing?
[16:54:41]	<civodul>	well maybe i should focus on one thing at a time :-)
[16:54:46]	<mb[m]1>	So the culprit is likely one of these: https://sourceware.org/git/?p=glibc.git;a=commitdiff;h=665ce88d68fd13c5c... and https://sourceware.org/git/?p=glibc.git;a=commitdiff;h=dc258ce62ae0bbb45...
[16:55:27]	* jonsger has joined #guix
[16:55:31]	<mb[m]1>	civodul: "m4" and a few others fail like this: https://paste.debian.net/998765/
[16:55:36]	<civodul>	you'll find yourself being a libc hacker without noticing
[16:55:41]	<mb[m]1>	Hehe.
[16:56:24]	<civodul>	in this case it's the 'configure' file that leads to a decoding error
[16:56:28]	<civodul>	could you check its encoding?

[16:57:52]	<mb[m]1>	I don't currently have the files. But I had built much further before the glibc update (with the C++ fixes only).
[16:57:56]	* jonsger has joined #guix
[17:00:55]	<mb[m]1>	I unpacked the m4 source and ran `file` on it:
[17:00:57]	<mb[m]1>	configure: POSIX shell script, UTF-8 Unicode text executable
[17:01:24]	<civodul>	so it could be an issue with iconv in the new libc
[17:06:59]	* ryanwatkins has joined #guix
[17:11:08]	<g_bor>	hello!
[17:11:18]	<g_bor>	Anyone seen this?
[17:11:30]	<g_bor>	It's on core-updates:
[17:11:51]	<g_bor>	starting phase `patch-usr-bin-file' Backtrace: 16 (primitive-load "/gnu/store/4jf0jcpvjn6r2warav5xmxhmddf?") In ice-9/eval.scm: 191:35 15 (_ _) In srfi/srfi-1.scm: 863:16 14 (every1 #<procedure 9b0b40 at /gnu/store/yifqwmdxc4pmd?> ?) In /gnu/store/yifqwmdxc4pmdpm73diq03lqkprnrizn-module-import/guix/build/gnu-build-system.scm: 711:27 13 (_ _) 171:4 12 (patch-usr-bin-file #:native-inputs _ #:inputs _ # _) In srfi/srfi
[17:13:38]	<civodul>	g_bor: mb[m]1 was just reporting the exact same issue :-)
[17:13:45]	<mb[m]1>	g_bor: Yes, I'm currently trying to track it down.
[17:14:36]	<g_bor>	I've also seen it on python-boot0, slightly different message.
[17:14:55]	<g_bor>	Can this be locale related?
[17:15:05]	<g_bor>	I have hu_HU.utf8 here.
[17:16:04]	* civodul has quit (Read error: Connection reset by peer)
[17:17:26]	<g_bor>	mb[m]1: can I be of assistance?
[17:18:19]	* civodul has joined #guix
[17:18:42]	<mb[m]1>	g_bor: The failure happens in the build container which uses a fixed locale (C?).
[17:19:17]	<g_bor>	Ok, then it's not that.
[17:19:38]	<g_bor>	It just seemd to be some character conversion error...
[17:20:08]	<g_bor>	Maybe some invalid charaters leaked in somehow...
[17:20:15]	<g_bor>	Wait a sec...
[17:22:51]	<g_bor>	It seems, that a lot of packages are broken.
[17:23:01]	<g_bor>	I laso got that for gettext.
[17:26:26]	* xkapastel has joined #guix
[17:30:38]	<mb[m]1>	Can it be ncurses related?
[17:31:12]	<g_bor>	Might be...
[17:31:21]	<g_bor>	did not checked that.
[17:32:01]	<mb[m]1>	g_bor: Can you see if reverting 667082d59104d4b964dce878f5e8c0f8ad1be958 makes a difference?
[17:32:24]	<g_bor>	Ok, i will try that

[17:36:36]	<g_bor>	Do we know about what was the last time it was working?
[17:36:58]	<g_bor>	We could do a git bisect on core-updates?
[17:37:13]	<g_bor>	Do we have a good commit to start from?

[18:40] <g_bor> Ok, now it's building.

[18:40] <g_bor> I think this will take a time...

[18:41] <g_bor> (guile does take some time :) )

[18:42] <g_bor> I'm trying to help to switch the default icedtea to icedtea-8, and we'd like to that on core-updates...

[18:43] <g_bor> I've just written on the mailing list, that it would be nice if we could have substitutes for the core-updates gnu-build-system.

[18:43] <g_bor> That could speed things like this up.

[18:50] <efraim> mb[m]1: just popped in for a second, core-updates is failing on aarch64 also

[18:50] <efraim> civodul too ^

[18:52] <g_bor> Oh, well.

[18:52] <g_bor> Do we have a bug report on that already?

[18:53] <g_bor> Now I'm trying what mb[m]1 suggested, by reverting the ncurses commit.

[18:54] <g_bor> It might be easier to track related information in a bug...

[20:52] <g_bor> reverting 667082d59104d4b964dce878f5e8c0f8ad1be958 did not help

Does anyone know of a last working commit id?

I'm trying to seen if it's working after last merge of master, but if someone knows a later point to start investigating the issue would help.

It would also help, if someone with a more powerful computer could help, as building on mine is very slow.