travis-ci/travis-ci

A distributed build system for the open source community. | This chat is not actively monitored by Travis CI employees, for Travis support please email support@travis-ci.org.

Jay Wong

@xiaohk

My python program was terminated due to MemoryError. Is there anyway to request more memory to build my project?

Dariusz Rumiński

@keradus

if you pay for your travis account, contact support

ghking

@ghking1

hello, how can I create travis-ci server in local area network?

and does travis-ci server open source?

Paul S.

@pshirshov

Does anyone experience issues fetching artifacts from maven central on .com right now?..

Francisco Albert Albusac

@tatitati

Guys, I'm trying to use travis with docker-compose to run mysql. This container is publishing the port 3306 ,however I have this from travis:

Error starting userland proxy: listen tcp 0.0.0.0:3306: bind: address already in use
ERROR: Encountered errors while bringing up the project.
The command "docker-compose up" failed and exited with 1 during .

I was wondering if there is any work around for this.....

Francisco Albert Albusac

@tatitati

Also, can I set an environmental variable from a different place of "env"?, I would like to set a variable as result of a command

Alex

@cyberpunk1971

Hi, i'm trying to set up my react app to deploy from heroku but i'm having a couple of issues. First how does the deployment process work. I push from my local environment to git hub repo, then build starts, then deploys to heroku? Also, I am passing the test but fail deploy because it can't find that app. I have tried using the repo name and also the heroku auto generated name of the app but both throw the error. Has anyone had this happen before?

Alex

@cyberpunk1971

Ok, nvm it works now. I guess there is a bit of a delay between pushing up to the repo and travis.

Clayton Long

@Clayton7510

It doesn't look like Travis is bootstrapping Java 8 on new builds. I received the following error. Is there a new bug here? I am on GitHub. So, perhaps the outage had an impact on my builds? I just figured once builds started happening again then everything would just resume.

$ export JAVA_HOME=~/oraclejdk8
$ export PATH="$JAVA_HOME/bin:$PATH"
$ ~/bin/install-jdk.sh --target "/home/travis/oraclejdk8" --workspace "/home/travis/.cache/install-jdk" --feature "8" --license "BCL"
install-jdk.sh 2019-05-02
Expected feature release number in range of 9 to 13, but got: 8
The command "~/bin/install-jdk.sh --target "/home/travis/oraclejdk8" --workspace "/home/travis/.cache/install-jdk" --feature "8" --license "BCL"" failed and exited with 3 during .

config:

language: java
sudo: false
branches:
  only:
  - develop
  - /^release\/v.*$/
  - /^hotfix\/v.*$/
cache:
  directories:
    - $HOME/.m2
jdk:
- oraclejdk8

Clayton Long

@Clayton7510

Looks like it's just not working with TravisCI anymore. Man, I've been using TravisCI for years with GitHub. But no reported issues + an incredibly vague TravisCI issue and no word on updates. CircleCI's config is a little more cumbersome. But if it gets me building again...

Gonzalo Casas

@gnz_twitter

Hi folks, anyone knows if it's possible to switch to Linux containers when using docker in Windows builds?

Gonzalo Casas

@gnz_twitter

..it looks like it's not really possible at the moment.

Andrea Stevanato

@thymbahutymba

Hi guys, I have some problem to deploy after build. I used travis sshkey generate for generate the sshkey and then encrypt the private key and put as env variable under deploy: api_key: secure: $GITHUB_TOKEN. The repository is private and the error that i get is the following: /home/travis/.rvm/gems/ruby-2.4.5/gems/octokit-4.6.2/lib/octokit/response/raise_error.rb:16:in `on_complete': GET https://api.github.com/user: 401 - Bad credentials // See: https://developer.github.com/v3 (Octokit::Unauthorized). What I did not understand?

Győri Sándor

@AlexAegis

Hi there! Do matrix jobs work in the same workspace?

Győri Sándor

@AlexAegis

If so, I'd like to combine two separate .lcov coverage reports. both produced in separate jobs. I have the script that combines them commited into the repo. I just want to run it after both completes successfully and upload it. Is there a hook for this?

Dariusz Rumiński

@keradus

@AlexAegis , no, each job on given build run on separated machine

Tobiah

@TobiTenno

hi, i'm attempting to expose an environment variable during a node(js) script and then check it in a subsequent script step

here's my config


language: node_js
branches:
  only:
    - gh-pages
    - /.*/
node_js:
  - '8'
script:
  - node generateData.js
  - if [ "$TRAVIS_COMMIT_BUILD" = "true" ]; then chmod +x .travis/push.sh  ; else echo "do build? $TRAVIS_COMMIT_BUILD" ; fi
  - if [ "$TRAVIS_COMMIT_BUILD" = "true" ]; then ./.travis/push.sh  ; fi

so, in the node generateData.js line, i'm calling process.env.COMMIT_BUILD = "true", but the debug statement i added (echo "do build? $TRAVIS_COMMIT_BUILD") always comes up empty

anyone have an idea how to get this exposed, or would it be possible to set an env var from the exit code of the node script?

Tobiah

@TobiTenno

I decided to just write out a temp file that's ignored by git and cat that into the env var

Turing Zhu

@Turing-Chu

I encountered the following error. Who can tell me what happened and how can I fix it?

Deploying application
/Users/travis/.rvm/gems/ruby-2.4.3/gems/dpl-pages-1.10.11/lib/dpl/provider/pages.rb:141:in `mkdir': File exists @ dir_s_mkdir - /var/folders/nz/vv4_9tw56nv9k3tkvyszvwg80000gn/T/d20190622-3759-1lk0umr/work (Errno::EEXIST)
    from /Users/travis/.rvm/gems/ruby-2.4.3/gems/dpl-pages-1.10.11/lib/dpl/provider/pages.rb:141:in `github_pull_or_init'
    from /Users/travis/.rvm/gems/ruby-2.4.3/gems/dpl-pages-1.10.11/lib/dpl/provider/pages.rb:195:in `block in push_app'
    from /Users/travis/.rvm/rubies/ruby-2.4.3/lib/ruby/2.4.0/tmpdir.rb:89:in `mktmpdir'
    from /Users/travis/.rvm/gems/ruby-2.4.3/gems/dpl-pages-1.10.11/lib/dpl/provider/pages.rb:190:in `push_app'
    from /Users/travis/.rvm/gems/ruby-2.4.3/gems/dpl-1.10.11/lib/dpl/provider.rb:199:in `block in deploy'
    from /Users/travis/.rvm/gems/ruby-2.4.3/gems/dpl-1.10.11/lib/dpl/cli.rb:41:in `fold'
    from /Users/travis/.rvm/gems/ruby-2.4.3/gems/dpl-1.10.11/lib/dpl/provider.rb:199:in `deploy'
    from /Users/travis/.rvm/gems/ruby-2.4.3/gems/dpl-1.10.11/lib/dpl/cli.rb:32:in `run'
    from /Users/travis/.rvm/gems/ruby-2.4.3/gems/dpl-1.10.11/lib/dpl/cli.rb:7:in `run'
    from /Users/travis/.rvm/gems/ruby-2.4.3/gems/dpl-1.10.11/bin/dpl:5:in `<top (required)>'
    from /Users/travis/.rvm/gems/ruby-2.4.3/bin/dpl:23:in `load'
    from /Users/travis/.rvm/gems/ruby-2.4.3/bin/dpl:23:in `<main>'
failed to deploy

L!εsAμεr

@liesauer

hey guys, is this way possible?

or there is another right way

Anuoluwapo Akinseye

@an-apluss

hi everyone, I got this error message while travis-ci is running my test.
snapshot of error
I need guide(s) on how to get it fix. thanks

Gytis Šk.

@gskema

Hello, I am running build on php7.4snapshot, but it's the only version that's failing due to ParseError (which is supposed to be caught by catch block). Anyone dealt with similar issues ?

John Zabroski

@jzabroski

Is there a way to benchmark a Travis-CI build, similar to how TeamCity works? I joined a project, FluentMigrator, as a co-maintainer, and am trying to wrap my head around Travis CI for the first time.... it doesnt seem to have the nice performance benchmarking feature of TeamCity where I can see how long each sub-step took.

I can kind of see with my own eyeballs where the problem is right now, but the build timed out after 50 minutes, so i am assuming there may be multiple inefficiencies.

Setting up mono-jay (5.20.1.19-0xamarin2+ubuntu1604b1) ...
Setting up mono-csharp-shell (5.20.1.19-0xamarin2+ubuntu1604b1) ...
Setting up monodoc-base (5.20.1.19-0xamarin2+ubuntu1604b1) ...
Setting up referenceassemblies-pcl (2014.04.14-1xamarin5+ubuntu1604b1) ...
Setting up mono-roslyn (5.20.1.19-0xamarin2+ubuntu1604b1) ...
Mono precompiling /usr/lib/mono/4.5/csc.exe for amd64 (trying with LLVM, this might take a few minutes)...
Mono precompiling /usr/lib/mono/4.5/vbc.exe for amd64 (trying with LLVM, this might take a few minutes)...
Mono precompiling /usr/lib/mono/4.5/VBCSCompiler.exe for amd64 (trying with LLVM, this might take a few minutes)...
Mono precompiling /usr/lib/mono/4.5/Microsoft.CodeAnalysis.CSharp.dll for amd64 (trying with LLVM, this might take a few minutes)...

John Zabroski

@jzabroski

I also now see the following :

C# support for Travis-CI is community maintained.
Please open any issues at https://travis-ci.community/c/languages/37-category and cc @joshua-anderson @akoeplinger @nterry

Is it possible the C# toolchain has some regression causing issues?

I Found this reply by @joshua-anderson https://travis-ci.community/t/net-core-sdk-installed-on-every-build/2836/3

Travis-CI spins up a fresh virtual machine on every build, so we have to re-install .Net Core every time.
Short of pre-installing C# on the virtual machine image, which isn’t possible right now (Travis-CI tries to minimize the amount of software preinstalled on the image), I don’t think there’s much we can do.

John Zabroski

@jzabroski

Updated that thread with a quesiton as to whether tunneling my build through docker would speed things up https://travis-ci.community/t/net-core-sdk-installed-on-every-build/2836/11?u=jzabroski

John Zabroski

@jzabroski

Would anyone be able to explain to me why some of my builds take 6 minutes on Travis CI and others time out at 50 minutes? Is it just "noisy neighbors"?

I read https://docs.travis-ci.com/user/common-build-problems/#timeouts-installing-dependencies but the fact the timeout occurs at 50 minutes is just strange to me.

https://docs.travis-ci.com/user/common-build-problems/#build-times-out-because-no-output-was-received discusses using travis_wait, but what is frustrating for me is it is not clear WHY the build timed out.

Separately, the 50 minute thing is mentioned as a "common problem" here: https://docs.travis-ci.com/user/customizing-the-build#build-timeouts but it's not mentioned as a common build problem in https://docs.travis-ci.com/user/common-build-problems/

Noah

@coolreader18

Hi! Can anyone tell me why this build didn't deploy to GitHub pages? I think it might be something about on.branch, but that doesn't seem like it would be necessary here. https://travis-ci.org/RustPython/RustPython/jobs/564569924

Jonathan Leitschuh

@JLLeitschuh

Yes, but only if you use encrypted variables. See the Travis CI documentation for more details

Will Carhart

@wcarhart

Hi everyone, new to Travis. I'm trying to automate testing/deployment via PRs to my repo. Can anyone answer this question: https://stackoverflow.com/questions/57387227/how-to-run-tests-on-pr-deploy-on-merge-with-travis-ci

Keenan Brock

@kbrock

Hi all,
I'm looking for a way to reduce build resources:

When testing a rails gem against pg, sqlite, and mysql, people typically setup an env switch:

env:
  - DB=sqlite
  - DB=mysql
  - DB=postgresql

services:
  - mysql
  - postgresql

Is there a way to limit a service to an environment?
e.g.: only run mysql when DB=mysql and only run postgres when DB=postgresql?

thanks all

Keenan Brock

@kbrock

oh well, if I figure it out, I'll try and update the online docs.
have a great day all

Thibault Brocherieux

@zephylac

Hey, Im' having issue with travis, I'm trying to deploy angular project on Google Flex App Engine.

I'm building prod build using angular ng build directive, then deploying, at this point the Dockerfile used to deploy on gcloud doesn't work because it can't find prod build, I've used debug prompt, production build exist and is where it should be. I'm currently asking myself if this comes from travis.

dist: trusty
sudo: false
language: node_js
node_js:
- '8'
cache:
  directories:
  - "./my-app/node_modules"

install:
- cd my-app
- npm install

script:
- npm run lint
- ng build --prod --output-path ../nginx/www
- cd nginx

deploy:
  provider: gae
  keyfile: "../client-secret.json"
  project: "my-app"
  config: "app.yml"
  on:
    branch: master

Docker is exiting here because it can't find www folder

ADD www /usr/share/nginx/www/

Thibault Brocherieux

@zephylac

Capture d’écran 2019-09-23 à 14.56.35.png

Thibault Brocherieux

@zephylac

To give some more insights :

Below is the output given from commands run in travis.yml script

$ cd nginx
The command "cd nginx" exited with 0.
0.01s$ ls
app.yml
Dockerfile
nginx.conf
www
The command "ls" exited with 0.

It would seem that www folder exist.

Now inside Dockerfile, I'm copying current build context

COPY . /tmp/build/

RUN ls /tmp/build/

The problem is that current context doesn't include

Step 6/9 : COPY . /tmp/build/
Step 7/9 : RUN ls /tmp/build/
Dockerfile
app.yml
nginx.conf
source-context.json

Thibault Brocherieux

@zephylac

Woud it be originated that file context given to deploy isn't updated or something? I can't find doc page explaining it.

Thibault Brocherieux

@zephylac

FYI : I found the issue. By default travis stash all the changes -> https://docs.travis-ci.com/user/deployment#uploading-files-and-skip_cleanup . Just needed to add inside the deploy

deploy:
  skip_cleanup: true

Dmitry

@xotonic

How can I skip after_success stage in matrix builds but one?
Basically, I want to collect test coverage data for only one build in matrix

language: java
jdk:
  - openjdk11
  - openjdk-ea
matrix:
  allow_failures:
    - jdk: openjdk-ea
before_script:
  - chmod +x gradlew
script:
  - ./gradlew check --info
after_success:
  bash <(curl -s https://codecov.io/bash) -cF server

davide-romanini

@davide-romanini

someone know if the PyPI deployment is available on Windows build environment?

Joan Massich

@massich

is there a way to force homebrew addon to use sources?
in other words i would like to replace these 2 lines with the addon

    brew install hdf5 libmatio;
    brew reinstall -s libmatio;

Daniel Ruf

@DanielRuf

Hi, it seems macOS builds are not running, they are not even scheduled.

Where communities thrive

People

Repo info

Activity