Repository navigation
run bundler with --retry to avoid network dodginess - #21
pivotalfish wants to merge 1 commit into
Conversation
|
Hey pivotalfish! Thanks for submitting this pull request! All pull request authors must have a Contributor License Agreement (CLA) on-file with us. Please sign the appropriate CLA (individual or corporate). When sending signed CLA please provide your github username in case of individual CLA or the list of github usernames that can make pull requests on behalf of your organization. |
|
We have created an issue in Pivotal Tracker to manage this. You can view the current status of your issue at: https://www.pivotaltracker.com/story/show/78643382. |
|
Hi PivotalFish! Thanks for the PR. Firstly, I am reticent to take this pull request. It's a patch for a deeper issue, which is, we'd like this to not be necessary. Can you attach logs of the failure mode for us? Anything that could help us identify what kind of flakiness is occurring. I'll get @chwu-cf at PWS to look into this. |
|
We deploy a large Rails apps which depends on 237 gems in total (this includes the development and test groups) to PWS. The error we see is that staging the app fails - we upload the app and successfully start a bundler run, which then terminates abruptly. We don't get much useful error output - the output from CF_TRACE shows the cf cli client polling the API endpoint and getting 400s (an odd HTTP code for I'm busy), and then giving up at some point. The deeper issue is that requests to endpoints over the internet may arbitrary fail and rubygems.org may not successfully respond to every HTTP ever - if you can think of a good workaround for An alternate approach is to run |
|
Per discussion this morning, we're recommending vendoring as the best workaround when this issue occurs, as well as kicking the question of connectivity up to our mates in PWS. |
We used to run
bundle packageprior to uploading our app to PWS - this seems to have stopped working and we've been battling some flakiness with our builds as a result.A handy workaround is to make bundler retry if fetching fails - our app uses quite a few gems, and this seems to fix sporadic flakiness we experience when pushing.