GNU bug report logs -
#53645
Unable to guix pull on pinebook pro
Previous Next
To reply to this bug, email your comments to 53645 AT debbugs.gnu.org.
Toggle the display of automated, internal messages from the tracker.
Report forwarded
to
bug-guix <at> gnu.org
:
bug#53645
; Package
guix
.
(Mon, 31 Jan 2022 00:08:01 GMT)
Full text and
rfc822 format available.
Acknowledgement sent
to
"D. Birch" <d.s.j.birch <at> gmail.com>
:
New bug report received and forwarded. Copy sent to
bug-guix <at> gnu.org
.
(Mon, 31 Jan 2022 00:08:01 GMT)
Full text and
rfc822 format available.
Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
As title.
I'm using a pinebook pro (arm64) and guix pull is giving up on me.
I've attached the output of stdout and stderr.
Did I do something wrong?
[Message part 2 (text/html, inline)]
[guix-pull-bug (application/octet-stream, attachment)]
Information forwarded
to
bug-guix <at> gnu.org
:
bug#53645
; Package
guix
.
(Mon, 31 Jan 2022 10:11:02 GMT)
Full text and
rfc822 format available.
Message #8 received at 53645 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
D. Birch schreef op zo 30-01-2022 om 20:36 [+0000]:
> As title.
>
> I'm using a pinebook pro (arm64) and guix pull is giving up on me.
>
> I've attached the output of stdout and stderr.
>
> Did I do something wrong?
From guix-pull-bug:
> builder for `/gnu/store/5mqlzr4gs2f28c4i64rcbxs2zw2zzfzc-gcc-10.3.0.drv' failed with exit code 1
> build of /gnu/store/5mqlzr4gs2f28c4i64rcbxs2zw2zzfzc-gcc-10.3.0.drv failed
> View build log at '/var/log/guix/drvs/5m/qlzr4gs2f28c4i64rcbxs2zw2zzfzc-gcc-10.3.0.drv.bz2'.
The log /var/log/guix/drvs/5m/qlzr4gs2f28c4i64rcbxs2zw2zzfzc-gcc-10.3.0.drv.bz2
is required to investigate the issue.
Greetings,
Maxime.
[signature.asc (application/pgp-signature, inline)]
Information forwarded
to
bug-guix <at> gnu.org
:
bug#53645
; Package
guix
.
(Mon, 31 Jan 2022 19:14:01 GMT)
Full text and
rfc822 format available.
Message #11 received at 53645 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
I took a look but I don't know what's up with
> g++ [...] -o cc1 [...]
> collect2: fatal error: ld terminated with signal 9 [Killed]
Seems like the linking step failed somehow?
Greetings,
Maxime.
[signature.asc (application/pgp-signature, inline)]
Information forwarded
to
bug-guix <at> gnu.org
:
bug#53645
; Package
guix
.
(Mon, 31 Jan 2022 19:22:04 GMT)
Full text and
rfc822 format available.
Message #14 received at 53645 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
Added. Thank-you.
On Mon, 31 Jan 2022, 10:10 Maxime Devos, <maximedevos <at> telenet.be> wrote:
> D. Birch schreef op zo 30-01-2022 om 20:36 [+0000]:
> > As title.
> >
> > I'm using a pinebook pro (arm64) and guix pull is giving up on me.
> >
> > I've attached the output of stdout and stderr.
> >
> > Did I do something wrong?
>
> From guix-pull-bug:
> > builder for `/gnu/store/5mqlzr4gs2f28c4i64rcbxs2zw2zzfzc-gcc-10.3.0.drv'
> failed with exit code 1
> > build of /gnu/store/5mqlzr4gs2f28c4i64rcbxs2zw2zzfzc-gcc-10.3.0.drv
> failed
> > View build log at
> '/var/log/guix/drvs/5m/qlzr4gs2f28c4i64rcbxs2zw2zzfzc-gcc-10.3.0.drv.bz2'.
>
> The log
> /var/log/guix/drvs/5m/qlzr4gs2f28c4i64rcbxs2zw2zzfzc-gcc-10.3.0.drv.bz2
> is required to investigate the issue.
>
> Greetings,
> Maxime.
>
[Message part 2 (text/html, inline)]
[qlzr4gs2f28c4i64rcbxs2zw2zzfzc-gcc-10.3.0.drv.bz2 (application/bzip2, attachment)]
Information forwarded
to
bug-guix <at> gnu.org
:
bug#53645
; Package
guix
.
(Mon, 31 Jan 2022 20:05:02 GMT)
Full text and
rfc822 format available.
Message #17 received at 53645 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
Hi there,
Maxime Devos <maximedevos <at> telenet.be> writes:
> [[PGP Signed Part:Undecided]]
> I took a look but I don't know what's up with
>
>> g++ [...] -o cc1 [...]
>> collect2: fatal error: ld terminated with signal 9 [Killed]
>
> Seems like the linking step failed somehow?
I would guess that the build ran out of memory, that can happen with the
pinebookpro's mere 4G. Do you have some swap space allocated? I've had
to do that on my rockpro64 (same chip) for guix pull to eventually
succeed.
Here's an example to create a quick swapfile in case you're not
familiar, as root:
# Create 10G
dd if=/dev/zero of=/mnt/swapfile bs=1MiB count=10240
# Enable it
chmod 600 /mnt/swapfile
mkswap /mnt/swapfile
swapon /mnt/swapfile
Hope this helps! Ideally you wouldn't have to build so much, but I'm
afraid we don't have aarch64 substitutes ready at the moment :-/.
Cheers,
Pierre
[signature.asc (application/pgp-signature, inline)]
Information forwarded
to
bug-guix <at> gnu.org
:
bug#53645
; Package
guix
.
(Wed, 02 Feb 2022 08:47:01 GMT)
Full text and
rfc822 format available.
Message #20 received at 53645 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
Thank-you for all the help.
I left the laptop ticking away again after adding 10G of swap.
I've attached the stdout and stderr, as before.
Things got much further and the system was building for the whole day, I
left it running overnight and in the morning hit another "bug".
I can't see where the build log is this time to attach.
I don't fully understand what guix pull does. If I was to run it again,
would everything begin from the start, or is there some chance re-running
it would pick up where it left off?
I'll try it shortly and find out I suppose.
[Message part 2 (text/html, inline)]
[guix-pull-bug-03-attempt (application/octet-stream, attachment)]
Information forwarded
to
bug-guix <at> gnu.org
:
bug#53645
; Package
guix
.
(Wed, 02 Feb 2022 22:44:01 GMT)
Full text and
rfc822 format available.
Message #23 received at submit <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
Hi,
"D. Birch" <d.s.j.birch <at> gmail.com> writes:
> Thank-you for all the help.
>
> I left the laptop ticking away again after adding 10G of swap.
>
> I've attached the stdout and stderr, as before.
>
> Things got much further and the system was building for the whole day, I left it running overnight and in the morning hit another "bug".
>
> I can't see where the build log is this time to attach.
That's strange that there isn't a log file to report, it looks like
util-linux didn't build but I'm not sure. What you could try is to run
`guix pull -v3' to enable more verbosity, and it should show build
output for packages as it's building them.
> I don't fully understand what guix pull does. If I was to run it again, would everything begin from the start, or is there some chance re-running it would pick up where it left off?
Yeah, luckily guix pull should pickup where it left off and not rebuild
packages that are already built and present in /gnu/store. So normally,
you should only have to go through the couple of days of building once
(Well actually, in normal times we should have binary substitutes that
guix pull can download instead of building them, but the CI is having
issues with aarch64 at the moment :-/.).
That being said, new commits could be pushed and cause new rebuilds
between guix pull invokation, what you can do is pull a specific commit
first:
guix pull -v3 --commit=2641621 # Commit picked from the log file.
Hope that helps!
Thanks,
Pierre
[signature.asc (application/pgp-signature, inline)]
Information forwarded
to
bug-guix <at> gnu.org
:
bug#53645
; Package
guix
.
(Wed, 02 Feb 2022 22:44:02 GMT)
Full text and
rfc822 format available.
Information forwarded
to
bug-guix <at> gnu.org
:
bug#53645
; Package
guix
.
(Thu, 03 Feb 2022 10:21:02 GMT)
Full text and
rfc822 format available.
Message #29 received at 53645 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
On Wed, 2 Feb 2022 at 22:42, Pierre Langlois <pierre.langlois <at> gmx.com>
wrote:
> That's strange that there isn't a log file to report, it looks like
> util-linux didn't build but I'm not sure. What you could try is to run
> `guix pull -v3' to enable more verbosity, and it should show build
> output for packages as it's building them.
> ...
> guix pull -v3 --commit=2641621 # Commit picked from the log file.
Tried my best not to top post ;)
I've attached the output from an attempt using a slight modification to the
command recommended above. The command I ran was
guix pull -v3 --commit=2641621 2>&1 | tee guix-pull-bug-verbose-01
Thank-you
[Message part 2 (text/html, inline)]
[guix-pull-bug-verbose-01 (application/octet-stream, attachment)]
Information forwarded
to
bug-guix <at> gnu.org
:
bug#53645
; Package
guix
.
(Thu, 03 Feb 2022 10:22:01 GMT)
Full text and
rfc822 format available.
Information forwarded
to
bug-guix <at> gnu.org
:
bug#53645
; Package
guix
.
(Thu, 03 Feb 2022 15:58:01 GMT)
Full text and
rfc822 format available.
Message #35 received at 53645 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
This test is failing:
misc: swaplabel ... FAILED (misc/swaplabel)
This looks like <https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=996764>.
Debian even has a patch!
Or the cause could be different ...
Greetings,
Maxime.
[signature.asc (application/pgp-signature, inline)]
Information forwarded
to
bug-guix <at> gnu.org
:
bug#53645
; Package
guix
.
(Thu, 03 Feb 2022 16:28:01 GMT)
Full text and
rfc822 format available.
Message #38 received at 53645 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
On Thu, 3 Feb 2022 at 15:56, Maxime Devos <maximedevos <at> telenet.be> wrote:
> This test is failing:
>
> misc: swaplabel ... FAILED
> (misc/swaplabel)
>
> This looks like <https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=996764
> >.
> Debian even has a patch!
>
> Or the cause could be different ...
>
> Greetings,
> Maxime.
>
I re-ran with -K (e.g. guix pull -v3 --commit 2641621 -K 2>&1 | tee
guix-pull-bug-verbose-02) to keep build files, and a quick cat
/tmp/guix-build-util-linux-2.37.2.drv-0/util-linux-2.37.2/tests/output/misc/swaplabel.err
appears to confirm the bug
mkswap: error: swap area needs to be at least 10 pages
mkswap: <swapfile>: insecure permissions <perm>, fix with: chmod 0600
<swapfile>
mkswap: <swapfile> contains holes or other unsupported extents.
This swap file can be rejected by kernel on swap activation!
Use --verbose for more details.
mkswap: Label was truncated.
I'm still green to know how to fix this, but I presume if I check out the
latest version via git and the docs online, I should be able to update the
definition for util-linux to include the patch, then try again?
[Message part 2 (text/html, inline)]
Information forwarded
to
bug-guix <at> gnu.org
:
bug#53645
; Package
guix
.
(Thu, 03 Feb 2022 16:37:01 GMT)
Full text and
rfc822 format available.
Message #41 received at 53645 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
On Thu, 3 Feb 2022 at 16:27, D. Birch <d.s.j.birch <at> gmail.com> wrote:
>
> I re-ran with -K (e.g. guix pull -v3 --commit 2641621 -K 2>&1 | tee
> guix-pull-bug-verbose-02) to keep build files, and a quick cat
> /tmp/guix-build-util-linux-2.37.2.drv-0/util-linux-2.37.2/tests/output/misc/swaplabel.err
> appears to confirm the bug
>
> mkswap: error: swap area needs to be at least 10 pages
> mkswap: <swapfile>: insecure permissions <perm>, fix with: chmod 0600
> <swapfile>
>
> mkswap: <swapfile> contains holes or other unsupported extents.
> This swap file can be rejected by kernel on swap activation!
> Use --verbose for more details.
>
> mkswap: Label was truncated.
>
> I'm still green to know how to fix this, but I presume if I check out the
> latest version via git and the docs online, I should be able to update the
> definition for util-linux to include the patch, then try again?
>
> In fact, this was fixed upstream in commit
a02721f35b30d4ee0133446f016442f63117ddf6, is there a way I can easily
progress, by making use of that, or a later commit?
[Message part 2 (text/html, inline)]
Information forwarded
to
bug-guix <at> gnu.org
:
bug#53645
; Package
guix
.
(Thu, 03 Feb 2022 16:44:02 GMT)
Full text and
rfc822 format available.
Message #44 received at 53645 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
D. Birch schreef op do 03-02-2022 om 16:27 [+0000]:
> I'm still green to know how to fix this, but I presume if I check out
> the latest version via git and the docs online, I should be able to
> update the definition for util-linux to include the patch, then try
> again?
There's no need for that, you can use the patch from Debian
and add it to the 'patches' field of 'util-linux'.
It's odd though that util-linux fails to build for you even
though https://bordeaux.guix..gnu.org has a substitute:
$ guix weather util-linux --system=aarch64-linux
> [...]
Something else that's odd: there's only one hit for
$ grep -rF 'swap area needs to be at least ' | grep -F page
: the file tests/expected/misc/swaplabel.err. So how is mkswap
supposed to produce the message "mkswap: error: swap area needs to be
at least 10 pages' in the first place?
Greetings,
Maxime
[signature.asc (application/pgp-signature, inline)]
This bug report was last modified 3 years and 132 days ago.
Previous Next
GNU bug tracking system
Copyright (C) 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson.