Bug #124167 for Future-AsyncAwait: libfuture-asyncawait-perl FTBFS on s390x: test failure

Sun Jan 21 16:41:45 2018 gregoa [...] cpan.org - Ticket created

From:	gregoa [...] cpan.org
Subject:	libfuture-asyncawait-perl FTBFS on s390x: test failure

We have the following bug reported to the Debian package of Future-AsyncAwait (https://bugs.debian.org/887922): It doesn't seem to be a bug in the packaging, so you may want to take a look. Thanks! ------8<-----------8<-----------8<-----------8<-----------8<----- Source: libfuture-asyncawait-perl Version: 0.13-1 Severity: important https://buildd.debian.org/status/fetch.php?pkg=libfuture-asyncawait-perl&arch=s390x&ver=0.13-1&stamp=1516518213&raw=0 ... Test Summary Report ------------------- t/22context-foreach.t (Wstat: 11 Tests: 0 Failed: 0) Non-zero wait status: 11 Parse errors: No plan found in TAP output Files=15, Tests=71, 1 wallclock secs ( 0.03 usr 0.01 sys + 0.48 cusr 0.05 csys = 0.57 CPU) Result: FAIL Failed 1/15 test programs. 0/71 subtests failed. dh_auto_test: perl Build test --verbose 1 returned exit code 255 debian/rules:6: recipe for target 'build-arch' failed make: *** [build-arch] Error 2 ------8<-----------8<-----------8<-----------8<-----------8<----- Second log: https://buildd.debian.org/status/fetch.php?pkg=libfuture-asyncawait-perl&arch=s390x&ver=0.13-2&stamp=1516563783&raw=0 Thanks for considering, gregor herrmann, Debian Perl Group

Mon Jan 22 00:04:34 2018 leonerd-cpan [...] leonerd.org.uk - Correspondence added

On Sun Jan 21 16:41:45 2018, GREGOA wrote: Show quoted text

> Test Summary Report > ------------------- > t/22context-foreach.t (Wstat: 11 Tests: 0 Failed: 0) > Non-zero wait status: 11 > Parse errors: No plan found in TAP output

Wow, a segfault. How exciting. Yes, I know there's still something a bit wrong with foreach loop handling here. I haven't quite tracked it down. I don't imagine the failure will be s390-specific, it's probably just more apparent on some architectures than others due to minor variations in memory layout or somesuch. In any case, I'd suggest that Future::AsyncAwait is still perhaps a bit of an early implementation yet for debian to be considering packaging it just at the moment. There's still a number of things that don't really work yet, not to mention all the memory leaks I've not even looked into yet. -- Paul Evans

Mon Jan 22 00:04:35 2018 The RT System itself - Status changed from 'new' to 'open'

Mon Jan 22 13:36:17 2018 gregoa [...] debian.org - Correspondence added

Subject:	Re: [rt.cpan.org #124167] libfuture-asyncawait-perl FTBFS on s390x: test failure
Date:	Mon, 22 Jan 2018 19:18:05 +0100
To:	Paul Evans via RT <bug-Future-AsyncAwait [...] rt.cpan.org>
From:	gregor herrmann <gregoa [...] debian.org>

On Mon, 22 Jan 2018 00:04:35 -0500, Paul Evans via RT wrote: Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=124167 > > > On Sun Jan 21 16:41:45 2018, GREGOA wrote:

> > Test Summary Report > > ------------------- > > t/22context-foreach.t (Wstat: 11 Tests: 0 Failed: 0) > > Non-zero wait status: 11 > > Parse errors: No plan found in TAP output

> > Wow, a segfault. How exciting.

:) Show quoted text

> Yes, I know there's still something a bit wrong with foreach loop > handling here. I haven't quite tracked it down. I don't imagine the > failure will be s390-specific, it's probably just more apparent on > some architectures than others due to minor variations in memory > layout or somesuch.

So far only s390x is failing but yeah, who knows ... https://buildd.debian.org/status/logs.php?pkg=libfuture-asyncawait-perl Show quoted text

> In any case, I'd suggest that Future::AsyncAwait is still perhaps a > bit of an early implementation yet for debian to be considering > packaging it just at the moment. There's still a number of things > that don't really work yet, not to mention all the memory leaks > I've not even looked into yet.

I'm aware that the distro is not yet finished, and I've kept your big fat WARNING in the package description. Since it's already in unstable, it's too late to not package it, but I'll happily file a serious bug to keep it from migrating to testing until you think it's ready for general usage or you remove the warning. (Until then it at least gets some exposure, if only on the build daemons ...). ... Done: https://bugs.debian.org/888024 I'm looking forward to the future of this distribution :) Cheers, gregor -- .''`. https://info.comodo.priv.at -- Debian Developer https://www.debian.org : :' : OpenPGP fingerprint D1E1 316E 93A7 60A8 104D 85FA BB3A 6801 8649 AA06 `. `' Member VIBE!AT & SPI Inc. -- Supporter Free Software Foundation Europe `- NP: Bettina Wegner: Mutter Rock

Download signature.asc
application/pgp-signature 963b

Message body not shown because it is not plain text.

Sun Mar 18 15:08:46 2018 gregoa [...] cpan.org - Correspondence added

FWIW, 0.15 built successfully on s390x. All build logs: https://buildd.debian.org/status/logs.php?pkg=libfuture-asyncawait-perl&ver=0.15-1 Cheers, gregor

Mon Apr 02 14:08:45 2018 leonerd-cpan [...] leonerd.org.uk - Correspondence added

At this point I think this isn't anything s390x-specific, and is just another occurrence of https://rt.cpan.org/Ticket/Display.html?id=124353 -- Paul Evans

Mon Apr 02 14:08:46 2018 leonerd-cpan [...] leonerd.org.uk - Status changed from 'open' to 'rejected'

Thu Jan 17 07:30:31 2019 leonerd-cpan [...] leonerd.org.uk - Correspondence added

FYI, I believe this issue ought to be fixed since 0.17. In addition, I've been making quite a few more releases and bugfixes lately - 0.20 should hopefully be a lot more stable. Give that a try and let me know how you get on. -- Paul Evans

Thu Jan 17 10:33:41 2019 gregoa [...] debian.org - Correspondence added

Subject:	Re: [rt.cpan.org #124167] libfuture-asyncawait-perl FTBFS on s390x: test failure
Date:	Thu, 17 Jan 2019 16:33:24 +0100
To:	Paul Evans via RT <bug-Future-AsyncAwait [...] rt.cpan.org>
From:	gregor herrmann <gregoa [...] debian.org>

On Thu, 17 Jan 2019 07:30:33 -0500, Paul Evans via RT wrote: Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=124167 > > > FYI, I believe this issue ought to be fixed since 0.17.

Right, actually it built already successfully with 0.15, and I've closed the Debian bug already: https://bugs.debian.org/887922 Sorry for not reporting back here. Show quoted text

> In addition, I've been making quite a few more releases and bugfixes lately - 0.20 should hopefully be a lot more stable. > Give that a try and let me know how you get on.

0.20 is in Debian ... oh no, that's 0.19; I'll upload 0.20 shortly. BTW, do you think that 0.20 is mature enough to be included in (testing and then) in the upcoming Debian release? After our previous discussions here I've filed https://bugs.debian.org/888024 to keep it out of testing. Cheers, gregor -- .''`. https://info.comodo.priv.at -- Debian Developer https://www.debian.org : :' : OpenPGP fingerprint D1E1 316E 93A7 60A8 104D 85FA BB3A 6801 8649 AA06 `. `' Member VIBE!AT & SPI Inc. -- Supporter Free Software Foundation Europe `- NP: Kings of Convenience: Mrs. Cold

Download signature.asc
application/pgp-signature 963b

Message body not shown because it is not plain text.

Thu Jan 17 13:22:02 2019 leonerd-cpan [...] leonerd.org.uk - Correspondence added

On Thu Jan 17 10:33:41 2019, gregoa@debian.org wrote: Show quoted text

> Right, actually it built already successfully with 0.15, and I've > closed the Debian bug already: > https://bugs.debian.org/887922

Ah good :) Show quoted text

> 0.20 is in Debian ... oh no, that's 0.19; I'll upload 0.20 shortly.

Yes I already noticed 0.19 in Debian which lead me to remember here again. Show quoted text

> BTW, do you think that 0.20 is mature enough to be included in > (testing and then) in the upcoming Debian release? > After our previous discussions here I've filed > https://bugs.debian.org/888024 > to keep it out of testing.

I think it's certainly getting there. There's basically four known bugs currently. Two of them are "unsupported older perl versions" that don't affect Debian since we're already at 5.28. Third is a slow memory leak which accumulates unreachable SVs over time, but otherwise doesn't cause too much trouble, and fourth is that it doesn't handle named loop controls. Apart from those, it's getting more stable and reliable now; I'm considering adjusting the wording in the docs some more to reflect that fact. -- Paul Evans

Thu Jan 17 13:33:52 2019 gregoa [...] debian.org - Correspondence added

Subject:	Re: [rt.cpan.org #124167] libfuture-asyncawait-perl FTBFS on s390x: test failure
Date:	Thu, 17 Jan 2019 19:33:29 +0100
To:	Paul Evans via RT <bug-Future-AsyncAwait [...] rt.cpan.org>
From:	gregor herrmann <gregoa [...] debian.org>

On Thu, 17 Jan 2019 13:22:02 -0500, Paul Evans via RT wrote: Show quoted text

> > BTW, do you think that 0.20 is mature enough to be included in > > (testing and then) in the upcoming Debian release? > > After our previous discussions here I've filed > > https://bugs.debian.org/888024 > > to keep it out of testing.

> > I think it's certainly getting there. There's basically four known > bugs currently. Two of them are "unsupported older perl versions" > that don't affect Debian since we're already at 5.28. Third is a > slow memory leak which accumulates unreachable SVs over time, but > otherwise doesn't cause too much trouble, and fourth is that it > doesn't handle named loop controls.

Sounds good. Show quoted text

> Apart from those, it's getting more stable and reliable now; I'm > considering adjusting the wording in the docs some more to reflect > that fact.

If 0.20 (or a newer release) should be included in Debian buster we have to close #888024 before the soft freeze starts at 2019-02-12 to let the package migrate to testing. Cheers, gregor -- .''`. https://info.comodo.priv.at -- Debian Developer https://www.debian.org : :' : OpenPGP fingerprint D1E1 316E 93A7 60A8 104D 85FA BB3A 6801 8649 AA06 `. `' Member VIBE!AT & SPI Inc. -- Supporter Free Software Foundation Europe `- NP: REM: Belong

Download signature.asc
application/pgp-signature 963b

Message body not shown because it is not plain text.

Sun Jan 26 10:56:16 2020 gregoa [...] cpan.org - Correspondence added

On Thu Jan 17 13:33:52 2019, gregoa@debian.org wrote: Show quoted text

> On Thu, 17 Jan 2019 13:22:02 -0500, Paul Evans via RT wrote:

Show quoted text

> > Apart from those, it's getting more stable and reliable now; I'm > > considering adjusting the wording in the docs some more to reflect > > that fact.

> If 0.20 (or a newer release) should be included in Debian buster we > have to close #888024 before the soft freeze starts at 2019-02-12 to > let the package migrate to testing.

One year later, same question :) We're still blocking libfuture-asyncawait-perl from getting out of Debian unstable. What do you think - is it ready for prime time by now? (One reason I'm asking is that Mojolicous now optionally uses Future::AsyncAwait.) Cheers, gregor

Sun Jan 26 13:04:14 2020 leonerd-cpan [...] leonerd.org.uk - Correspondence added

On Sun Jan 26 10:56:16 2020, GREGOA wrote: Show quoted text

> One year later, same question :) > > We're still blocking libfuture-asyncawait-perl from getting out of > Debian unstable. > What do you think - is it ready for prime time by now? > (One reason I'm asking is that Mojolicous now optionally uses > Future::AsyncAwait.)

Oh, I think almost certainly it can be migrated to testing by now yes. It's been sat at the current version for quite a while and I think it's had increasingly more "battle testing", with no major issues coming to light. Now could be the time to migrate it indeed. -- Paul Evans

Sun Jan 26 15:25:47 2020 gregoa [...] debian.org - Correspondence added

Subject:	Re: [rt.cpan.org #124167] libfuture-asyncawait-perl FTBFS on s390x: test failure
Date:	Sun, 26 Jan 2020 21:25:34 +0100
To:	Paul Evans via RT <bug-Future-AsyncAwait [...] rt.cpan.org>
From:	gregor herrmann <gregoa [...] debian.org>

On Sun, 26 Jan 2020 13:04:15 -0500, Paul Evans via RT wrote: Show quoted text

> > We're still blocking libfuture-asyncawait-perl from getting out of > > Debian unstable. > > What do you think - is it ready for prime time by now?

Show quoted text

> Oh, I think almost certainly it can be migrated to testing by now > yes. It's been sat at the current version for quite a while and I > think it's had increasingly more "battle testing", with no major > issues coming to light. Now could be the time to migrate it indeed.

Great news, and thanks for the swift reply. I've closed the blocking bug now. Cheers, gregor -- .''`. https://info.comodo.priv.at -- Debian Developer https://www.debian.org : :' : OpenPGP fingerprint D1E1 316E 93A7 60A8 104D 85FA BB3A 6801 8649 AA06 `. `' Member VIBE!AT & SPI Inc. -- Supporter Free Software Foundation Europe `- NP: James Taylor: Something In The Way She Moves

Download signature.asc
application/pgp-signature 963b

Message body not shown because it is not plain text.