Hi Dima,
the patching that fails is only meant to make the docker image work in a
local developer setup where you have your sage source tree (with changes
and untracked files) mounted into the docker container. In the CI context
the line that reads `git status --porcelain` in the Dockerfile sho
Hi Frédéric,
docker pull source-clean is expected to fail. We should probably add a
message there that explains that. It says `docker pul ... || true` in the
script there.
On Sunday, October 25, 2020 at 1:52:48 PM UTC+1 Frédéric Chapoton wrote:
> Build failure comes from not being able to pull
On Mon, Dec 28, 2020 at 10:05 PM Sebastian Oehms wrote:
>
> https://gitlab.com/sagemath/dev/trac/-/pipelines
so docker build does some patching (why? I don't know) and there are
failures (why? I don't know),
and then an attempt to do something with these rejected patches fails:
cf. e.g. https:/
https://gitlab.com/sagemath/dev/trac/-/pipelines
On Mon, Dec 28, 2020 at 11:02 PM Sebastian Oehms wrote:
>
> Thanks to you both for the hints! Yes, instead of develop I used the
> version tag directly. Running docker as you suggested solves the
> problem. But, the version tag works on a newer CPU
Thanks to you both for the hints! Yes, instead of develop I used the
version tag directly. Running docker as you suggested solves the
problem. But, the version tag works on a newer CPU (the one causing
the crash was an i5 of the 2. generation), as well, and for 9.1.beta1
even on the older CPU.
Wha
did you run
docker run -it sagemath/sagemath-dev:develop
or something else ?
Le lundi 28 décembre 2020 à 12:45:20 UTC+1, seb@gmail.com a écrit :
> It seems that there is still (or again) something broken. For the images
> of the beta releases since the 8th of December Sage crashes if I try
I don't know anything about docker images, but "illegal instruction" is
usually an indication that the binary was made for a CPU that accepts more
commands than the one you use.
On Mon, 28 Dec 2020, 11:45 seb@gmail.com, wrote:
> It seems that there is still (or again) something broken. For t
It seems that there is still (or again) something broken. For the images of
the beta releases since the 8th of December Sage crashes if I try to start
it:
Sage build/upgrade complete!
make[1]: Leaving directory '/home/sage/sage'
sage@8711c66abafe:~/sage$ ./sage
┌───
On Tue, Dec 8, 2020 at 5:17 PM Matthias Koeppe wrote:
>
> On Tuesday, December 8, 2020 at 8:16:53 AM UTC-8 Dima Pasechnik wrote:
>>
>> On Tue, Dec 8, 2020 at 3:56 PM Eric Gourgoulhon wrote:
>> > On the page
>> > https://hub.docker.com/r/sagemath/sagemath/tags?page=1&ordering=last_updated
>> > we
On Tuesday, December 8, 2020 at 6:17:52 PM UTC+1 Matthias Koeppe wrote:
> User packages' CI scripts may be referring to this tag.
>
>
I would remove the `-py3` suffix or otherwise, we think that the
`sage9.3.beta3` tag is not using Python 3.
--
You received this message because you are subsc
On Tuesday, December 8, 2020 at 8:16:53 AM UTC-8 Dima Pasechnik wrote:
> On Tue, Dec 8, 2020 at 3:56 PM Eric Gourgoulhon
> wrote:
> > On the page
> >
> https://hub.docker.com/r/sagemath/sagemath/tags?page=1&ordering=last_updated
> > we see the Docker images develop and develop-py3, as well a
On Tue, Dec 8, 2020 at 8:33 AM Frédéric Chapoton wrote:
>
> the build for 9.3.beta3 has passed successfully, so there should be a docker
> for that, and a docker for develop that should point to the same.
>
> https://gitlab.com/sagemath/sage/-/pipelines
>
Can we get this documented in the Develo
On Tue, Dec 8, 2020 at 3:56 PM Eric Gourgoulhon wrote:
>
>
>
> Le mardi 8 décembre 2020 à 09:33:03 UTC+1, Frédéric Chapoton a écrit :
>>
>> the build for 9.3.beta3 has passed successfully, so there should be a docker
>> for that, and a docker for develop that should point to the same.
>
>
> Cool,
Le mardi 8 décembre 2020 à 09:33:03 UTC+1, Frédéric Chapoton a écrit :
> the build for 9.3.beta3 has passed successfully, so there should be a
> docker for that, and a docker for develop that should point to the same.
Cool, thank you Frédéric!
On the page
https://hub.docker.com/r/sagemath/sa
the build for 9.3.beta3 has passed successfully, so there should be a
docker for that, and a docker for develop that should point to the same.
https://gitlab.com/sagemath/sage/-/pipelines
Frédéric
Le jeudi 15 octobre 2020 à 12:09:15 UTC+2, Sébastien Labbé a écrit :
> Bonjour,
>
> Thanks to Vin
A .gitattributes file was introduced
in https://trac.sagemath.org/ticket/29733 that enforces this CRLF to LF
conversion
Le dimanche 8 novembre 2020 à 14:17:41 UTC+1, Frédéric Chapoton a écrit :
> and this seems to be due to automatic replacement of CRLF by LF
>
> warning: CRLF will be replaced
and this seems to be due to automatic replacement of CRLF by LF
warning: CRLF will be replaced by LF in
build/pkgs/planarity/patches/extern.patch.
The file will have its original line endings in your working directory.
Le dimanche 8 novembre 2020 à 14:09:28 UTC+1, Frédéric Chapoton a écrit :
>
artifact complains about our patches not applying:
[planarity-3.0.0.5.p0] patching file c/planarity.h
[planarity-3.0.0.5.p0] Hunk #1 FAILED at 77 (different line endings).
[planarity-3.0.0.5.p0] Hunk #2 FAILED at 88 (different line endings).
[planarity-3.0.0.5.p0] 2 out of 2 hunks FAILED -- saving
The 9.3.beta1 build on gitlab is blocked by a failure with "planarity"
see
https://gitlab.com/sagemath/dev/trac/-/pipelines/213154052
Le jeudi 5 novembre 2020 à 11:06:28 UTC+1, Sébastien Labbé a écrit :
> At
> https://hub.docker.com/r/sagemath/sagemath-dev/tags
> tags are still 4 months old.
>
>
At
https://hub.docker.com/r/sagemath/sagemath-dev/tags
tags are still 4 months old.
I notice that #30736 is in 9.3.beta0 but not #30820 yet. This should be
fixed automatically if #30820 is in the next beta?
On Sunday, October 25, 2020 at 8:25:46 PM UTC+1 Frédéric Chapoton wrote:
> Could some p
Could some please review https://trac.sagemath.org/ticket/30820 ?
Maybe also relevant : https://trac.sagemath.org/ticket/30736 (already
positive review)
Le samedi 24 octobre 2020 à 09:59:00 UTC+2, Frédéric Chapoton a écrit :
> I have made https://trac.sagemath.org/ticket/30820 for some small fi
Build failure comes from not being able to pull something :
+ docker pull source-clean
Error response from daemon: pull access denied for source-clean, repository
does not exist or may require 'docker login': denied: requested access to
the resource is denied
Le vendredi 23 octobre 2020 à 20:
I have made https://trac.sagemath.org/ticket/30820 for some small fixes.
Le samedi 24 octobre 2020 à 09:20:17 UTC+2, Frédéric Chapoton a écrit :
> I can see recent tentatives from Orsay, but they failed. Maybe Erik did
> launch a runner there ?
>
> Our .ci setting needs to be adapted to use pyth
I can see recent tentatives from Orsay, but they failed. Maybe Erik did
launch a runner there ?
Our .ci setting needs to be adapted to use python3 by default. And there is
a complaint about
.ci/build-docker.sh: line 60: syntax error: unterminated quoted string
but the script looks ok to me.
L
It seems that all pipelines are failing since 9.2.beta4. Could someone
please have a look ?
https://gitlab.com/sagemath/sage/-/pipelines
Le jeudi 15 octobre 2020 à 12:09:15 UTC+2, Sébastien Labbé a écrit :
> Bonjour,
>
> Thanks to Vincent, I learned recently how to use continuous integration in
25 matches
Mail list logo
