New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
should output the log as the command goes when using travis_wait #4190
Comments
Sorry, that is how the function is implemented right now. If you have suggestions for improvement, you can look at the code here: https://github.com/travis-ci/travis-build/blob/68cdaaf2cfcad54671c3b5896cd867e21a63c737/lib/travis/build/templates/header.sh#L108-145 |
Hi @Da-Huang Thanks for your suggestion. As @BanzaiMan suggested, we are happy to take PRs to improve this. I'm going to close this issue for now as we can discuss any changes within the PR. Thanks a bundle Josh |
@Da-Huang enjoy the following solution: long-living-command & # send the long living command to background!
minutes=0
limit=30
while kill -0 $! >/dev/null 2>&1; do
echo -n -e " \b" # never leave evidences!
if [ $minutes == $limit ]; then
break;
fi
minutes=$((minutes+1))
sleep 60
done |
Wow, cool! Thanks, @evilaliv3 |
I replaced
|
This enables writing log before the command finishes. travis-ci/travis-ci#4190 (comment)
Thanks @abitrolly, that works perfectly. 👍 I ran into some issues with OSX jobs though, it looks like "minutes" / the |
This is done by having an echo on the background with 3x5min sleeps. The echo is killed if the rest of the travis_script takes less time to complete. Solution inspired from this issue comment while looking at travis_wait: travis-ci/travis-ci#4190 (comment)
This is done by having an echo on the background with 3x5min sleeps. The echo is killed if the rest of the travis_script takes less time to complete. Solution inspired from this issue comment while looking at travis_wait: travis-ci/travis-ci#4190 (comment)
…'t output anything for a few minutes. travis-ci/travis-ci#4190 (comment)
* First attempt to setup Travis integration * Set dist to bionic * Add R and graphiz * Add R and Python cache * Update apt-get before installs * sudo-install R packages * Use R 3.5.2 * Fix sources list * Add key for CRAN * Send artefacts back to GitHub * Fix version setup * Add tag utility * Ignore R version for now, R team should have kept 3.5 alone.. * Update setup to properly install R 3.5 * [9 debugging commits] * Fix R setup setup * Attempts to fix the problem with too long logs * Ignore deprecated * Try quiet installs * Add explicit install of readat dependency * More quiet arguments & debugs * Move travis_wait to setup.sh * Use workaround for travis_wait consuming the logs, see: travis-ci/travis-ci#4190 (comment) * More improvements to Travis build * Try to make better use of cache (manual, no sudo) * Add quiet to devtools::install_url * Local R lib setup attempts: * With .Renviron? * In home? * With mkdir and standard path * Make sure interactive is on * Try to install limma with Bioconductor * Comment-out unnecessary env change & cleanup * Use a single tag to force Travis to use cache * Upgrade pip packages for reproducibility builds & revert * Run nbpipeline from root * Add raw samples data * Enable Travis cache for notebooks * Do not attempt to display graph in Travis console * Remove spurious apt: true
and thus it does not work with `-v` for investigating hangs well. This seems to be by design: travis-ci/travis-ci#4190 Also I simplified a comment about `homebrew.update`. I had this in-flight change while editing e05f397 but forgot to ammend this.
* First attempt to setup Travis integration * Set dist to bionic * Add R and graphiz * Add R and Python cache * Update apt-get before installs * sudo-install R packages * Use R 3.5.2 * Fix sources list * Add key for CRAN * Send artefacts back to GitHub * Fix version setup * Add tag utility * Ignore R version for now, R team should have kept 3.5 alone.. * Update setup to properly install R 3.5 * [9 debugging commits] * Fix R setup setup * Attempts to fix the problem with too long logs * Ignore deprecated * Try quiet installs * Add explicit install of readat dependency * More quiet arguments & debugs * Move travis_wait to setup.sh * Use workaround for travis_wait consuming the logs, see: travis-ci/travis-ci#4190 (comment) * More improvements to Travis build * Try to make better use of cache (manual, no sudo) * Add quiet to devtools::install_url * Local R lib setup attempts: * With .Renviron? * In home? * With mkdir and standard path * Make sure interactive is on * Try to install limma with Bioconductor * Comment-out unnecessary env change & cleanup * Use a single tag to force Travis to use cache * Upgrade pip packages for reproducibility builds & revert * Run nbpipeline from root * Add raw samples data * Enable Travis cache for notebooks * Do not attempt to display graph in Travis console * Remove spurious apt: true
Use trick from travis-ci/travis-ci#4190 (comment) to make travis not to abort build if no output produced for over 10 minutes. Signed-off-by: Lev Stipakov <lev@openvpn.net>
Hi @david-gibbs-ig , maybe the following works... See travis-ci/travis-ci#4190 (comment)
When
travis_wait
is added as prefix, the log should be output as the command goes, instead of after it finishes.The text was updated successfully, but these errors were encountered: