issues.csv

Id, Title, Creator, Assignee, Milestone, State, Body Text
1836,Error while accessing my build details,rvdwijngaard, , ,OPEN,Hi,

I get an error if I try to access https://app.shippable.com/projects/557f0f35edd7f2c0521800b4

The error I get is: {"id":2100,"message":"No build found","methodName":"validateCaller|_checkAccessLevel"},
1835,Previous Build History Gone,hillbilly90, , ,OPEN,Over the past 2 months we have been using shippable to deploy to our Amazon server. As of today, I can't see any of the previous builds. It even says 'No builds have been triggered yet.'

Any ideas?,
1834,Errors when trying to force sync or trigger build by push,EHLOVader, , ,OPEN,I have some strange errors showing up on the top of the page, and recent pushes don't appear on shippable. 

```json
{"id":1005,"message":"Internal API error : getProjectById","methodName":"chkPermProjectId|_getProjectById"}
```

Force sync also produces a different error that appears across the top.

```json
{"id":3300,"message":"No Owner for subscriptionId::555d3a2eedd7f2c052f68fb8","methodName":"caller:555d3a6dfba96c0c005d4377|subscriptions|syncById:555d3a2eedd7f2c052f68fb8|_getValidOwner"}
```
,
1832,Source location not set to value specified within "Settings",leonfs, , ,OPEN,Hi,

I'm getting failures on the builds due to a problem on the working directory path not being set to the value specified within Settings.

If I run:
```
before_install:
  - pwd
````
I get /root

I had to run in order to have successful builds:
```
before_install:
  - cd $SHIPPABLE_REPO_DIR
```

Link to failed build
https://app.shippable.com/builds/55e63a79fe05460b00f9f989
,
1829,scala: console output format is borked,ahjohannessen, , ,OPEN,Recently the console output format from build log is borked. We get output like this:

```
[2K[info] Resolving org.scala-lang.modules#scala-parser-combinators_2.11;1.0.4 ...

[info] Resolving org.scala-lang.modules#scala-xml_2.11;1.0.4 ...

[info] Resolving com.google.protobuf#protobuf-java;2.6.1 ...

[info] Resolving com.trueaccord.scalapb#scalapb-runtime_2.11;0.4.15 ...

[info] Resolving com.trueaccord.lenses#lenses_2.11;0.3 ...

```
and
```
[info] No tests were executed.
[info] BehaviourTestingSpec:
[info] BehaviorTesting
[info] - should provide given-when-then with success
```
,
1828,Builds timeout after success,RealNobody, , ,OPEN,This is not something that happens every time, but it has been happening a reasonable amount recently.

Sometimes when we run a build, it times out.  I have not identified any reliable circumstances under which this happens.

When I look at the timed out builds, I cannot see anything in the console output that indicates why it timed out.  In the console output, I see the same basic output as I do for a successful run except that the last step that is shown is `uploading artifacts` and the `save_reports` step is the only missing step.

If we do get a timeout, running it again often gets a green build.

The test suite on average runs for about 30 minutes, well within the 2 hour or so timeout.

The only diagnostic information I may have is the following:

The time for a script does not seem to be based on how long a test actively runs.  The time starts as of the moment that the test build button is clicked rather than as of the time that the test actually starts doing active work.  I have seen at least one timeout where a test started and couldn't get a worker for the timeout period so it never even pulled the code from git or anything, it just timedout waiting.

I don't know in any of the timeout failures we have if this is a factor or not.  It could be in that if the timeout is 2 hours and the run sits for 1.9 hours before starting to pull and run, the timeout could be a false negative because the total run time was too long.

It seems weird to me though that in 5 or 6 timeouts that I can remember looking at they all failed at the same point - right after uploading artifacts.

From my perspective, it seems like one of the following may be happening:

* The timeout is occurring because shippable is doing some post processing work that is getting into some kind of infinite loop and timing out.
* The timeout is only checked at certain points which would allow the script to start running, run and then fail at the first checkpoint.

It would be nice if the timeout time was based on the run time after it gets a worker rather than from the time that the build is started.  That would make more sense to me.,
1826,Postgres and Ubuntu = random password,Tobur, , ,OPEN,The same yml, have different result sometimes success, sometimes failed:

  PGPASSWORD=shippable psql -U postgres -c "create USER dbuser WITH PASSWORD 'dbpassword' SUPERUSER;"
psql: FATAL:  password authentication failed for user "postgres"

Some RANDOM exist here.

I tried also, what you are say in DOC:
psql -U postgres -c "create USER dbuser WITH PASSWORD 'dbpassword' SUPERUSER;"
psql: FATAL:  password authentication failed for user "postgres"

And again sometimes this work, sometimes not.

Please help me to resolve this issue.

On the documentation:

 # Postgre binds to 127.0.0.1 by default and is started on boot. Default username is "postgres" with no password. Create a DB as part of before script to use it,
1825,UNSTABLE build still calls after_success,LuisMuniz, , ,OPEN,Hi,

I read from diverse sources that I should place my deployment instructions in after_success, but I found out today that my unstable build was deployed to my production environment, why is this? Should I put my deployment code somewhere else? Should I check the build status myself?,
1824,Email invites sent to long gone team members,sdlarsen, , ,OPEN,Hi. We're using shippable with bitbucket integration. For some reason we keep getting email invites sent to team members that are no longer with us, and not present in our company bitbucket account. Why is that and how do we stop it?

Br.
Søren
ÜberResearch,
1823,Feature Request: Validate Docker Hub Credentials on save,behemphi, , ,OPEN,When I save my Docker hub credentials, I don't know for sure I have them right.  This presents a problem later if I don't get the expected result because I don't necessarily know where the problem lies.  If I do know the problem is the credentials I have to navigate away to the right place to attempt a fix at which time I am still in the same situation: I don't _know_ I have them right.

Thanks for a cool product.,
1821,[Security] Builds and details still accessible when user is excluded from Bitbucket organization,bittner, , ,OPEN,When you're part of an organization and you create a Shippable account by logging in using e.g. Bitbucket you can see all public and private repositories.

When your account is removed from the Bitbucket organization then you can still see the builds, and you can still even inspect the details of all builds.

Is this behavior intended?,
1819,Can't trigger new builds,nigelwinterburn, , ,OPEN,I can't trigger a new build. At first, I thought maybe the Jira-generated super long branch name caused a problem, but now I can't get any branch to trigger a new build. I have reset the project and verified the new deploy keys were added in bitbucket. 

No new jobs appear after successful pushes.,
1818,Feature request: make it easier to identify skipped tests in test report,ixdy, , ,OPEN,I created a PR to use Shippable's test report parsing functionality in kubernetes/kubernetes/pull/13114.

A test report from an example run such as (https://app.shippable.com/builds/55de3d96f44cfb0c00153ccf) shows 1886 Passing and 9 Skipped tests, but there is no easy way to identify which tests were skipped.

Additionally, the GitHub commit status shows "Shippable builds completed. 3772/3790 tests passed", not making it clear that there were skipped tests. (Instead, one might wonder why some tests didn't pass, wondering if there were actually failures.),
1817,Code coverage report not useful with Go projects,ixdy, , ,OPEN,I tried adding Shippable code coverage support to the Kubernetes unit tests in kubernetes/kubernetes#13171. Shippable parses the coverage XML file without issue, but it reports 0% branch and sequence coverage across the entire project (e.g. https://app.shippable.com/builds/55de3e23f44cfb0c0015a260).

I'm guessing this is because the Go coverage tools seem to only report statement coverage. As it stands, this means that the coverage utility in Shippable isn't really useful for us, however.,
1813,Artifacts are not being uploaded sometimes,RealNobody, , ,OPEN,When a build succeeds or if ONLY RSpec fails, then artifacts are uploaded.

However, (so far) whenever the Cucumber tests fail, artifacts are not uploaded.

Here is a sample Cucumber failure:  https://app.shippable.com/builds/55d78ac1831cd10b00df2f5b
An RSpec failure (with artifacts): https://app.shippable.com/builds/55d7640032a7bd0c00a467a9
And a success: https://app.shippable.com/builds/55d7a2fd1f61b10c00280272

I can't find anything in the console output that tells me why the artifacts aren't being uploaded.  All I see is:

```
script
  <output of running the script>

uploading artifacts to s3
  echo 'skipping upload'
```

What is interesting is that I do NOT see the after script:
```
script:
 - rspec -r yarjuf -f JUnit -o shippable/testresults/results.xml
 - cd dashboard && node_modules/.bin/ember test ci --silent -c testem-ci.json > ../shippable/testresults/emberresults.xml && cd ..
 - DISPLAY=localhost:1.0 xvfb-run bundle exec cucumber --tags ~@skip-ci --format junit --out shippable/testresults

after_script:
 - cp -r cornucopia_report shippable/testresults
```

What I am wondering is why does the after_script not run if Cucumber fails.  I really need the artifacts because they contain diagnostic information to tell me why Cucumber is failing.

Thanks,
1808,Including global environment variables in matrix,ixdy, , ,OPEN,I'm trying to include a number of environment variables across all variations of our matrix. In particular, the configuration I'm trying is [here](https://github.com/ixdy/kubernetes/blob/coveralls/shippable.yml).

However, based on the [Shippable build results](https://app.shippable.com/builds/55d5188f4e6e0d0c001d6dba), it seems like these variables aren't being exported. They don't appear in the script tab, and an echo debug command which prints `"Reporting coverage results to Coveralls for service ${CI_NAME:-}"` isn't showing the variable defined, either.

I've tried reading through the docs at http://docs.shippable.com/yml_reference/, but it isn't very clear what I'm doing wrong.

My issue seems similar to #1223, but there wasn't an applicable resolution there.,
1805,Cannot use Android SDK 21,issue-reporter, , ,OPEN,I am trying to build a project with the [Google Maps Android API](https://developers.google.com/maps/documentation/android/), and the dependencies for the `build.gradle` look like this:

```
dependencies {
    repositories {
        mavenCentral()
    }
    compile fileTree(dir: 'libs', include: '*.jar')
    compile 'com.google.android.gms:play-services:7.8.0'
}
```

However, when building in Shippable, it fails with:

```
> Could not resolve all dependencies for configuration ':_debugCompile'.
> Could not find com.android.support:support-v4:22.2.0.
Searched in the following locations:
https://repo1.maven.org/maven2/com/android/support/support-v4/22.2.0/support-v4-22.2.0.pom
https://repo1.maven.org/maven2/com/android/support/support-v4/22.2.0/support-v4-22.2.0.aar
file:/root/.android-sdk/extras/android/m2repository/com/android/support/support-v4/22.2.0/support-v4-22.2.0.pom
file:/root/.android-sdk/extras/android/m2repository/com/android/support/support-v4/22.2.0/support-v4-22.2.0.aar
file:/root/.android-sdk/extras/google/m2repository/com/android/support/support-v4/22.2.0/support-v4-22.2.0.pom
file:/root/.android-sdk/extras/google/m2repository/com/android/support/support-v4/22.2.0/support-v4-22.2.0.aar
Required by:
:project:unspecified > com.google.android.gms:play-services:7.8.0 > com.google.android.gms:play-services-base:7.8.0
> Could not find com.android.support:mediarouter-v7:22.2.0.
Searched in the following locations:
https://repo1.maven.org/maven2/com/android/support/mediarouter-v7/22.2.0/mediarouter-v7-22.2.0.pom
https://repo1.maven.org/maven2/com/android/support/mediarouter-v7/22.2.0/mediarouter-v7-22.2.0.aar
file:/root/.android-sdk/extras/android/m2repository/com/android/support/mediarouter-v7/22.2.0/mediarouter-v7-22.2.0.pom
file:/root/.android-sdk/extras/android/m2repository/com/android/support/mediarouter-v7/22.2.0/mediarouter-v7-22.2.0.aar
file:/root/.android-sdk/extras/google/m2repository/com/android/support/mediarouter-v7/22.2.0/mediarouter-v7-22.2.0.pom
file:/root/.android-sdk/extras/google/m2repository/com/android/support/mediarouter-v7/22.2.0/mediarouter-v7-22.2.0.aar
Required by:
:project:unspecified > com.google.android.gms:play-services:7.8.0 > com.google.android.gms:play-services-cast:7.8.0
```,
1803,Build times out,kkarski, , ,OPEN,Have two builds timing out despite having 5 minions:

https://app.shippable.com/projects/5521ef255ab6cc1352b8b4eb
https://app.shippable.com/projects/5521ef255ab6cc1352b8b4ee

No error messages, logs or other information available.,
1799,Is there any plan to integrate with Atlasssian JIRA?,shamasis, , ,OPEN,Hello there,

Is there any plan to allow integration with Project Management systems? Especially JIRA?,
1783,when we "go get" a binary, it does not seem to be in the $PATH,sebest, , ,OPEN,see https://app.shippable.com/builds/55cc5955774b820c00800bb9

the same build is working on travis, so i guess it is just $PATH missing the path to go binaries,
1779,Golang build is too slow,caarlos0, , ,OPEN,Every build, Shippable downloads and install the Go version using `gvm`... I believe this could be optimized...,
1777,Cloud Foundry Support,thesynapses, , ,OPEN,Hi Support
Do you have any plans of supporting Cloud foundry builds ? 

Regards,
Sureel,
1764,Stop trimming console logs,pquentin, , ,OPEN,When downloading console logs, all spaces are preserved.

However, the console tab of any build page displays console logs without spaces (app.min.js appears to call `trim()` on every `output` in the `console` array). This is destructive and makes full tracebacks very hard to read. Can you please stop trimming those spaces? Thanks!,
1756,Google App Engine deployments aren't sustainable long-term,aaronwest, , ,OPEN,Google App Engine deployments, using the appcfg.py Python toolkit, do not work long-term. I'm using GAE SDK v1.9.24 and following the Shippable documentation listed [here](http://docs.shippable.com/appengine/#deployment-of-a-python-application) and am not able to get GAE deployments to work long-term.

I am able to get deployments to work, under certain conditions, for an hour until the GAE tokens expire. More on this below.

The Shippable documentation linked above outlines two scenarios for GAE deployments: password authentication or OAuth2 authentication. I've tried both and have been partially successful with OAuth2 authentication.

#### Password authentication

The Shippable docs instruct users to global env vars, one for your Google email and one for your Google password. With of course the password being encrypted using the project settings on Shippable. Then, the docs say to use the following command:

>- echo "$GAE_PASSWORD" | $GAE_DIR/google_appengine/appcfg.py -e "$EMAIL" --passin update .

The `--passin` flag is not available in the GAE SDK v1.9.24. I'm not sure when it was removed but it isn't available in the latest version of the SDK and there doesn't appear to be a replacement. Nevertheless, I attempted to get password-based authentication working by searching and seeing how others were doing it. I also consulted the Google docs. I was never able to get it working.

#### OAuth2 authentication

After failing with password-based authentication I tried OAuth2. I followed the Shippable instructions [here](http://docs.shippable.com/appengine/#deployment-of-a-python-application) which ask users to call a method of appcfg.py (`list_versions`) which lists application versions (Google supports up to 10 versions) for the app you have deployed on GAE. A byproduct of the calling this command - or any other command - of appcfg.py is the storage of an app token in the user's home directory. The token is called `.appcfg_oauth2_tokens` and is a basic JSON file which includes several pieces of information. Notable data are the app token string and the token expiry. The token expiry is a date in UTC (more specifically it is stored as zulu time in JSON) and is set to +1 hours from the time the `.appcfg_oauth2_tokens` was written to disk.

NOTE: The token is only written to disk if the user accepts the OAuth2 session in the browser.

The Shippable instructions go on to ask the user to encrypt the access token using a Shippable secure variable. No specific instructions are given which is unfortunate. But, the basic workflow is for the user to copy only the string in the `access_token` portion of the `.appcfg_oauth2_tokens` file. Then, copy that value, plus a corresponding name you make up, into *Encrypt Env Vars* in your Shippable project settings. This will then give you a value you can define as a secret in your shippable.yml file and use as a `$variable` in the appcfg.py command.

Once I figured all of this out I pushed a feature branch to GitHub and Shippable walked through the build process. And it worked. I was super happy.

An little while later I executed the same process and the Shippable build failed with error message:

>google.appengine.tools.appengine_rpc_httplib2.AuthPermanentFail: Access token is invalid.

So I spent a few hours searching, reading GAE documentation, and trying to get builds to work. I discovered the invalid token was due to GAE expiring ALL OAuth2 tokens after 1 hour. On a local machine such as my Mac this isn't a problem. The `.appcfg_oauth2_tokens` file exists on my machine and when I execute appcfg.py GAE simply updates the `access_token` value and the `token_expiry`. When executing the same command in a Shippable build it fails. This is because the `.appcfg_oauth2_tokens` file doesn't exist in the container AND the token being used as the secret (encrypted env var) is now invalid.

After more searching I tried the `--oauth2_refresh_token` switch in appcfg.py instead of the `--oauth2_access_token` switch in the Shippable documentation. As long as the token I was using as the secret env var was not expired, this command worked.

In summary, there doesn't appear to be a way to get long-term Shippable builds working with GAE. I couldn't get password authentication working and even if I could I don't want the deployment process tied to my Google Apps email. Also, the OAuth2 based authentication, while it can be engineered to work, doesn't work beyond 1 hour.

Here's some documentation pages I read while I was trying to get all this working. Hopefully, someone else can make sense of this process and we can get things working. In the meantime, my team will build and deploy to GAE using appcfg.py locally instead of through Shippable webhooks to GitHub. That's unfortunate but we need a predictable/working solution.

http://docs.shippable.com/appengine/
https://developers.google.com/console/help/new/#serviceaccounts
https://developers.google.com/identity/protocols/OAuth2ServiceAccount
https://cloud.google.com/appengine/docs/python/gettingstartedpython27/uploading,
1755,Strange fail,luckyraul, , ,OPEN,https://app.shippable.com/builds/55c1efd99f6e5e0b000974d8,
1752,minor: allow ANSI escape code in web build output log,pmoust, , ,OPEN,,
1746,It takes over 30s to load the project page,d3ming, , ,OPEN,Repro page: https://app.shippable.com/projects/55b81febedd7f2c0528a1c54

Takes over 30s to load the branches/ jobs for my project, and generally page load is slow.  Are there any performance issues being investigated now or am I a special case?  This is making it nearly impossible to use shippable for me.,
1745,Initial view after signing up is very unhelpful ...,grosser, , ,OPEN,Some inline documentation on how to get started would be nice instead of just a blank page

<img width="1075" alt="screen shot 2015-07-30 at 9 33 47 pm" src="https://cloud.githubusercontent.com/assets/11367/9000831/cb02972a-3702-11e5-9b96-b40d65cbe04a.png">
,
1743,Build failures with no errors and no test reports,vkarpov15, , ,OPEN,Heya,

I'm pretty new with shippable, but looks like all of my builds today are breaking despite the fact that the scripts all succeed and the tests all pass. Can't figure out why shippable is reporting failures. If you need more data (specific build ids, etc.) can you shoot me an email?

Thanks,
Valeri,
1740,Older build for the same pull request does not die on new commit,stekycz, , ,OPEN,We have found an issue that creates a lot of builds in the queue which makes a lot of delays. It can be done by forgotten change which should be in the pull request. I know that it can be killed manually but we usually want to kill previous build.

Is it possible to make it? Or is it possible to make it configurable per project?,
1723,Can't build Docker image with FROM poiting to private GCR image,rvrignaud, , ,OPEN,Hello,

We use GCR integration for our Docker builds (post-CI with image push).
For images with a FROM (parent image) that point to a public image (like ubuntu:14.04) it works fine.
Whoever if we try to use a FROM with a private GCR image we get a 403 Access Denied.
(like this build https://app.shippable.com/builds/55b6231f67a1d80b00778f78).


,
1720,Can't access my formation service,magnus-larsson, , ,OPEN,I can no longer access my service in my formation:
(worked fine a couple of days ago)

<img width="575" alt="screen shot 2015-07-25 at 11 50 13" src="https://cloud.githubusercontent.com/assets/1313727/8888945/9866a268-32c3-11e5-9241-14ba8022e0ec.png">

<img width="819" alt="screen shot 2015-07-25 at 11 50 49" src="https://cloud.githubusercontent.com/assets/1313727/8888942/80a1f9e8-32c3-11e5-8624-d82e7cff4c49.png">

(I don't know if it is related but the console says I have 3 containers up and running, while I only have started one service in my formation?)

I just restarted the service and Tomcat reports no problems to startup:

<img width="812" alt="screen shot 2015-07-25 at 11 53 31" src="https://cloud.githubusercontent.com/assets/1313727/8888955/f3542ff6-32c3-11e5-8d41-087430d4dc94.png">

But when I try to access it with curl or just connect using telnet I get connection refused errors:

    $ telnet mt-fo-lb-168423950.us-east-1.elb.amazonaws.com 30507
    Trying 54.175.216.126...
    telnet: connect to address 54.175.216.126: Connection refused
    Trying 54.152.137.164...
    telnet: connect to address 54.152.137.164: Connection refused
    telnet: Unable to connect to remote host

Hope you can help!

Regards,
Magnus.,
1718,Running parallel processes hangs a build,rudism, , ,OPEN,The latest build I have seen this in: https://app.shippable.com/builds/55b2b63e432f1f0c00fd0b32

In my build I'm trying to call a shell script that runs several test suites in parallel. I've used a few different methods, including spawning the processes with `&` and using the `wait` command, using GNU Parallel, and using `xargs -P`. In every case, after the command finishes, the build hangs and eventually times out. For example, calling a script that looks something like any of these seems to hang the build:

    #!/bin/sh
    testrunner --class class1 &
    testrunner --class class2 &
    wait
or

    #!/bin/sh
    parallel testrunner --class {} ::: class1 class2
or

    #!/bin/sh
    echo class1 class2 | xargs -n1 -P2 testrunner --class

I see the output from each of the commands in every case up to the completion of the final process, but then all output stops and the build does not seem to progress any farther.

The strange thing is that I have a notifier that runs in the `after_success` and `after_failure` blocks, and the jobs that hang will notify me that they have succeeded (meaning the `after_success` block is being run, even though the job is still listed as processing and eventually times out).,
1716,Issue with docker image pull from quay.io,glachesis, , ,OPEN,I'm using the Pre-CI docker integration with a base image pull from quay.io.

The pull only works every other time and often breaks with a 403 error.

From the logs  :

Sending build context to Docker daemon
Step 0 : FROM quay.io/xxx/xxxx
Pulling repository quay.io/xxx/xxxx
Error: Status 403 trying to pull repository xxx/xxxx: "{\"error\": \"Permission Denied\"}"

Any advice ?,
1713,HHVM Seriously Out of Date,jwriteclub, , ,OPEN,Running the following in before_script

`- php --version`

During a run with hhvm selected indicates that the current hhvm version is

```
HipHop VM 3.3.1 (rel)
Compiler: tags/HHVM-3.3.1-0-ge0c98e21167b425dddf1fc9efe78c9f7a36db268
Repo schema: f53e1743f31a2dee70c46aa66fb1b321c3e97c10
Extension API: 20140829
```

Whereas the most recent version of HHVM is 3.8.0. There's been a ton of changes in the meantime, and a number of projects which pass on recentish HHVM (3.6+ tested on an amazon VM) fail with such an old version.

In addition, PHP 5.6 is currently at 5.6.0, while 5.6.11 is available.

What are the chances of getting some updates to these runtimes?,
1712,Can't have node_js 0.12 alongside xvfb without waiting 10+ minutes per build?,posita, , ,OPEN,Originally, our `shippable.yml` file was this:

```yml
language: node_js

node_js:
  - "0.12"

addons:
  firefox: "29.0"

services:

before_script:
  - "git clone --depth 1 https://github.com/laurentj/slimerjs.git"
  - "export SLIMERJS_BIN=`pwd`/slimerjs/src/slimerjs"
  - "export SLIMERJSLAUNCHER=firefox-bin"
  - "export DISPLAY=:99.0"
  - "/etc/init.d/xvfb start"
```

But that took 10+ minutes to set up before even getting to the tests. So we added:

```yml
cache: true
```

But that didn't seem to do anything, plus it's [going away](http://blog.shippable.com/shippable-3.0-breaking-changes#BuildImage). So we added:

```yml
build_image: shippableimages/ubuntu1204_nodejs
```
Which sped up the time it took to get to the `before_script` step quite a bit, but then we failed on `/etc/inid./xvfb start`:

```
/root/[...].sh: line 87: /etc/init.d/xvfb: No such file or directory
```

Waiting 10+ minutes for a build is unusable for us. How do we use node_js 0.12 alongside xvfb?,
1709,CPU leak during archiving step,xherriot, , ,OPEN,Hello,

During the build, the console works fine but when arrived to the archiving step, my CPU goes to 100% (and ram x2), (Firefox does not respond) until the end of the step.

Before & after archiving:
<img width="547" alt="screen shot 2015-07-22 at 12 35 00" src="https://cloud.githubusercontent.com/assets/5094797/8823568/2ca0b476-306e-11e5-8db2-151f14325ac3.png">

During archiving:
<img width="555" alt="screen shot 2015-07-22 at 11 24 33" src="https://cloud.githubusercontent.com/assets/5094797/8823459/463f6bb2-306d-11e5-9129-4add6ddab19c.png">

Firefox 39 / Mac 10.9.5

,
1708,Pushing to GCR causes 403 "Access Denied",donmasakayan, , ,OPEN,In the final step of the build process when we're about to push the container to GCR, I'm getting either this error ([Build] (https://app.shippable.com/builds/55af5ca285656f0b006cf3ca)):
```
Sending image list
Error: Status 403 trying to push repository friendly-neighbors/api: "Access denied."
```

OR this error ([Build] (https://app.shippable.com/builds/55af64ec85656f0b007cbece)):
```
Sending image list
Please login prior to push:
Username: EOF
```


In GCR's documentation, [they do mention something about this here](https://cloud.google.com/tools/container-registry/?hl=en_US&_ga=1.128611918.357180263.1429696607#access_denied). Something about docker 1.7.x breaking the way docker writes credentials.

Other than that, I've set up GCR integration correctly.

Please advise. Thanks.,
1705,bundle install --local ignored,sleverbor, , ,OPEN,seems to install from rubygems anyway,
1698,MultiOS Tests,AbdealiJK, , ,OPEN,HI,

I was trying to setup tests for windows, OSX and Linux on shippable - and I wasn't able to find any yaml configuration for making a build matrix with all three.

Is multi OS tests possible ?

Is there any example yml file to test all 3 OSes with python I could refer to ?,
1697,Long-running mono process segfaults,rudism, , ,OPEN,I am having an issue with a shippable build that installs mono, compiles a project, and then runs some integration tests.

Here is an example of a failed build:
https://app.shippable.com/builds/55ad582a696c7f0b00588662

The step that fails is using mono to run the xunit console runner against our integration tests DLL. The integration tests are probably the most resource intensive step of our build--it downloads and restores a postgresql database from S3, then spins up local copies of our application (which get linked to a copy of the restored database, and tested by making local web requests against). We have a number of different test suites that run which each get their own copy of the database (to prevent each suite from affecting the data that the others have access to). This takes quite a long time to complete, and has successfully completed in some builds, but more often than not I get a segmentation fault in the mono executable.

When I load up ubuntu1404_base into a local docker container on my machine and run all of the commands from my shippable yml file locally, everything works fine every time (no seg faults), which leads me to believe there is some difference or perhaps some limitation being imposed on the process when Shippable runs it, but I am at a loss with trying to figure out how to debug or track this problem down.

Any help, suggestions, or guidance would be appreciated.
,
1695,GO Versions,Nalum, , ,OPEN,I'm trying to run tests on a go project and have added the following versions:
```
language: go

go:
  - 1.4
  - release 
  - tip
```

The problem I'm encountering is the following for both `release` and `tip`:

![shippable-error-01](https://cloud.githubusercontent.com/assets/384180/8773753/99119d10-2ece-11e5-814e-9fd0c2d82a5b.png)
,
1692,Extract file from container; publish as binary asset to GitHub release,garthk, , ,OPEN,I need to run a task in a Docker container and extract the result, e.g.:

```
set -x
docker build -t $TAG .
CID=`docker run -d $TAG`
docker logs -f $CID
docker cp $CID:$OUTPUTFILE .
```

How do I do this with a pre- or post-CI Docker build? The documentation doesn't make it clear whether Shippable can conclude a build by extracting build artifacts and [post it to a GitHub release](https://github.com/quantum/shippable-utils/blob/master/github-release.sh).,
1691,Automatically bring up new EC2 instance before running tests?,calebwherry, , ,OPEN,Instead of paying Amazon to keep an EC2 instance running 24/7, is there a way to have shippable spin one up automatically when tests are triggered?

Thanks!,
1688,Build process is totally hang,mustela, , ,OPEN,A built has been triggered an hour ago and I only have: 

`Waiting for logs...` 

https://app.shippable.com/builds/55a9184d2434f00c00d16ab1





,
1683,Coverage decrease alert,antonienko, , ,OPEN,Would it be possible to implement an alert when the code coverage decreases significantly?

I want to be able to know if some developer pushed code without proper test coverage.,
1681,Shippable won't build Centos 7 containers,buley, , ,OPEN,I've found the following at the top of a `Dockerfile` will cause my builds to fail on Shippable:

	FROM    centos:latest
	RUN yum -y install epel-release \
	  && yum -y update \
	  && yum -y upgrade
    ...

I've tried a number of different pulled images to build from, but it seems related to (apparently) some usage of AUFS somewhere.  https://github.com/docker/docker/issues/6980 Using the currently-supported `centos6.6` was my fix.

e.g. https://app.shippable.com/builds/55a72e96b617690b00155cd9

The key error:

> 	error: unpacking of archive failed on file /usr/bin/ping: cpio: cap_set_file


	Step 0 : FROM centos:latest
	latest: Pulling from centos
	f1b10cd84249: Already exists
	c852f6d61e65: Already exists
	7322fbe74aa5: Already exists
	7322fbe74aa5: Already exists
	Digest: sha256:57554136c655abb33ecb7bb790b1db0279668d3763c3b81f31bc6c4e60e4a1f3
	Status: Image is up to date for centos:latest
	---> 7322fbe74aa5
	Step 1 : MAINTAINER Taylor Buley (taylor.buley@mcnaughton.media)
	---> Running in edbbac70d137
	---> f00936843cd7
	Removing intermediate container edbbac70d137
	Step 2 : RUN groupadd -g 1855 application && /usr/sbin/useradd --uid 1855 --create-home --home-dir /usr/local/application --shell /bin/bash -g application application && /usr/sbin/usermod -g application root && chown -R application:application /usr/local/ && chown -R application:application /usr/src/ && chown -R application:application /usr/lib/ && chown -R application:application /usr/bin/ && chown -R root:application /var/log && chown -R application:application /srv && chmod 765 /srv && chmod +s /srv
	---> Running in e2cbae8418f3
	---> ee93c72e5f85
	Removing intermediate container e2cbae8418f3
	Step 3 : RUN yum -y install epel-release && yum -y update && yum -y upgrade
	---> Running in f3a7f6ed5f26
	Loaded plugins: fastestmirror
	Determining fastest mirrors
	* base: mirror.umd.edu
	* extras: mirror.vtti.vt.edu
	* updates: mirror.cc.columbia.edu
	Resolving Dependencies
	--> Running transaction check
	---> Package epel-release.noarch 0:7-5 will be installed
	--> Finished Dependency Resolution
	Dependencies Resolved
	================================================================================
	Package Arch Version Repository Size
	================================================================================
	Installing:
	epel-release noarch 7-5 extras 14 k
	Transaction Summary
	================================================================================
	Install 1 Package
	Total download size: 14 k
	Installed size: 24 k
	Downloading packages:
	warning: /var/cache/yum/x86_64/7/extras/packages/epel-release-7-5.noarch.rpm: Header V3 RSA/SHA256 Signature, key ID f4a80eb5: NOKEY
	Public key for epel-release-7-5.noarch.rpm is not installed
	Retrieving key from file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7
	Importing GPG key 0xF4A80EB5:
	Userid : "CentOS-7 Key (CentOS 7 Official Signing Key) <security@centos.org>"
	Fingerprint: 6341 ab27 53d7 8a78 a7c2 7bb1 24c6 a8a7 f4a8 0eb5
	Package : centos-release-7-1.1503.el7.centos.2.8.x86_64 (@CentOS)
	From : /etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7
	Running transaction check
	Running transaction test
	Transaction test succeeded
	Running transaction
	Installing : epel-release-7-5.noarch 1/1
	Verifying : epel-release-7-5.noarch 1/1
	Installed:
	epel-release.noarch 0:7-5
	Complete!
	Loaded plugins: fastestmirror
	http://mirror.es.its.nyu.edu/epel/7/x86_64/repodata/177d6316481d72b37adfa268bc016669c6c8092a815b20c3c962ecb258b84cc9-primary.sqlite.xz: [Errno 14] HTTP Error 404 - Not Found
	Trying other mirror.
	http://mirror.cc.columbia.edu/pub/linux/epel/7/x86_64/repodata/177d6316481d72b37adfa268bc016669c6c8092a815b20c3c962ecb258b84cc9-primary.sqlite.xz: [Errno 14] HTTP Error 404 - Not Found
	Trying other mirror.
	Loading mirror speeds from cached hostfile
	* base: mirror.umd.edu
	* epel: mirror.symnds.com
	* extras: mirror.vtti.vt.edu
	* updates: mirror.cc.columbia.edu
	Resolving Dependencies
	--> Running transaction check
	---> Package iputils.x86_64 0:20121221-6.el7 will be updated
	---> Package iputils.x86_64 0:20121221-6.el7_1.1 will be an update
	---> Package nss.x86_64 0:3.18.0-2.2.el7_1 will be updated
	---> Package nss.x86_64 0:3.19.1-3.el7_1 will be an update
	---> Package nss-sysinit.x86_64 0:3.18.0-2.2.el7_1 will be updated
	---> Package nss-sysinit.x86_64 0:3.19.1-3.el7_1 will be an update
	---> Package nss-tools.x86_64 0:3.18.0-2.2.el7_1 will be updated
	---> Package nss-tools.x86_64 0:3.19.1-3.el7_1 will be an update
	---> Package nss-util.x86_64 0:3.18.0-1.el7_1 will be updated
	---> Package nss-util.x86_64 0:3.19.1-1.el7_1 will be an update
	---> Package openssl-libs.x86_64 1:1.0.1e-42.el7.8 will be updated
	---> Package openssl-libs.x86_64 1:1.0.1e-42.el7.9 will be an update
	---> Package python.x86_64 0:2.7.5-16.el7 will be updated
	---> Package python.x86_64 0:2.7.5-18.el7_1.1 will be an update
	---> Package python-libs.x86_64 0:2.7.5-16.el7 will be updated
	---> Package python-libs.x86_64 0:2.7.5-18.el7_1.1 will be an update
	---> Package tzdata.noarch 0:2015d-1.el7 will be updated
	---> Package tzdata.noarch 0:2015e-1.el7 will be an update
	--> Finished Dependency Resolution
	Dependencies Resolved
	================================================================================
	Package Arch Version Repository Size
	================================================================================
	Updating:
	iputils x86_64 20121221-6.el7_1.1 updates 152 k
	nss x86_64 3.19.1-3.el7_1 updates 846 k
	nss-sysinit x86_64 3.19.1-3.el7_1 updates 51 k
	nss-tools x86_64 3.19.1-3.el7_1 updates 478 k
	nss-util x86_64 3.19.1-1.el7_1 updates 70 k
	openssl-libs x86_64 1:1.0.1e-42.el7.9 updates 949 k
	python x86_64 2.7.5-18.el7_1.1 updates 86 k
	python-libs x86_64 2.7.5-18.el7_1.1 updates 5.6 M
	tzdata noarch 2015e-1.el7 updates 432 k
	Transaction Summary
	================================================================================
	Upgrade 9 Packages
	Total download size: 8.6 M
	Downloading packages:
	Delta RPMs disabled because /usr/bin/applydeltarpm not installed.
	--------------------------------------------------------------------------------
	Total 3.1 MB/s | 8.6 MB 00:02
	Running transaction check
	Running transaction test
	Transaction test succeeded
	Running transaction
	Updating : nss-util-3.19.1-1.el7_1.x86_64 1/18
	Updating : nss-sysinit-3.19.1-3.el7_1.x86_64 2/18
	Updating : nss-3.19.1-3.el7_1.x86_64 3/18
	Updating : 1:openssl-libs-1.0.1e-42.el7.9.x86_64 4/18
	Updating : python-libs-2.7.5-18.el7_1.1.x86_64 5/18
	Updating : python-2.7.5-18.el7_1.1.x86_64 6/18
	Updating : iputils-20121221-6.el7_1.1.x86_64 7/18Error unpacking rpm package iputils-20121221-6.el7_1.1.x86_64
	
	error: unpacking of archive failed on file /usr/bin/ping: cpio: cap_set_file
	Updating : nss-tools-3.19.1-3.el7_1.x86_64 8/18
	error: iputils-20121221-6.el7_1.1.x86_64: install failed
	Updating : tzdata-2015e-1.el7.noarch 9/18
	Cleanup : nss-tools-3.18.0-2.2.el7_1.x86_64 10/18
	Cleanup : nss-sysinit-3.18.0-2.2.el7_1.x86_64 11/18
	Cleanup : nss-3.18.0-2.2.el7_1.x86_64 12/18
	Cleanup : python-2.7.5-16.el7.x86_64 13/18
	error: iputils-20121221-6.el7.x86_64: erase skipped
	Cleanup : tzdata-2015d-1.el7.noarch 14/18
	Cleanup : python-libs-2.7.5-16.el7.x86_64 15/18
	Cleanup : 1:openssl-libs-1.0.1e-42.el7.8.x86_64 16/18
	Cleanup : nss-util-3.18.0-1.el7_1.x86_64 17/18
	Verifying : 1:openssl-libs-1.0.1e-42.el7.9.x86_64 1/18
	Verifying : nss-3.19.1-3.el7_1.x86_64 2/18
	Verifying : python-libs-2.7.5-18.el7_1.1.x86_64 3/18
	Verifying : nss-sysinit-3.19.1-3.el7_1.x86_64 4/18
	Verifying : python-2.7.5-18.el7_1.1.x86_64 5/18
	Verifying : nss-tools-3.19.1-3.el7_1.x86_64 6/18
	Verifying : nss-util-3.19.1-1.el7_1.x86_64 7/18
	Verifying : tzdata-2015e-1.el7.noarch 8/18
	Verifying : python-2.7.5-16.el7.x86_64 9/18
	Verifying : nss-3.18.0-2.2.el7_1.x86_64 10/18
	Verifying : python-libs-2.7.5-16.el7.x86_64 11/18
	Verifying : iputils-20121221-6.el7_1.1.x86_64 12/18
	Verifying : nss-tools-3.18.0-2.2.el7_1.x86_64 13/18
	Verifying : tzdata-2015d-1.el7.noarch 14/18iputils-20121221-6.el7.x86_64 was supposed to be removed but is not!
	
	Verifying : iputils-20121221-6.el7.x86_64 15/18
	Verifying : nss-sysinit-3.18.0-2.2.el7_1.x86_64 16/18
	Verifying : 1:openssl-libs-1.0.1e-42.el7.8.x86_64 17/18
	Verifying : nss-util-3.18.0-1.el7_1.x86_64 18/18
	Updated:
	nss.x86_64 0:3.19.1-3.el7_1 nss-sysinit.x86_64 0:3.19.1-3.el7_1
	nss-tools.x86_64 0:3.19.1-3.el7_1 nss-util.x86_64 0:3.19.1-1.el7_1
	openssl-libs.x86_64 1:1.0.1e-42.el7.9 python.x86_64 0:2.7.5-18.el7_1.1
	python-libs.x86_64 0:2.7.5-18.el7_1.1 tzdata.noarch 0:2015e-1.el7
	Failed:
	iputils.x86_64 0:20121221-6.el7 iputils.x86_64 0:20121221-6.el7_1.1
	Complete!
	Removing intermediate container f3a7f6ed5f26
	The command '/bin/sh -c yum -y install epel-release && yum -y update && yum -y upgrade' returned a non-zero code: 1

(Interestingly, if I used `centos:centos7` it said the image was up to date and hung forever. Seems to be OK with `centos6`),
1673,Status badges require a none standard font be installed.,PJnes, , ,OPEN,The build status badges look terrible unless the user has DejaVuSans installed, as they'll fallback to the browser default.

![screenshot](https://cloud.githubusercontent.com/assets/5631/8663463/141f6d18-29c3-11e5-84d4-5aae0c9c9b38.png)

Any chance of adding some fallbacks alongside DejaVuSans?
,
1670,java.lang.IllegalStateException: endPosTable already set during a Process Annotation stage in JVM,patricioe, , ,OPEN,Hi support team, I notice this error when running my build. I seems to be a JVM reported issue which does not fail locally.

This is the link to my build: https://app.shippable.com/builds/559f1cac90252e0c006801ce

This is my dev version of the JVM:

```
java version "1.8.0_25"
Java(TM) SE Runtime Environment (build 1.8.0_25-b17)
Java HotSpot(TM) 64-Bit Server VM (build 25.25-b02, mixed mode)
```

```
[INFO] --- maven-compiler-plugin:3.3:compile (default-compile) @ dtos ---
[INFO] Changes detected - recompiling the module!
[INFO] Compiling 25 source files to /home/shippable/workspace/src/bitbucket.org/codigodevs/codigo/dtos/target/classes
An exception has occurred in the compiler (1.8.0_25). Please file a bug at the Java Developer Connection (http://java.sun.com/webapps/bugreport) after checking the Bug Parade for duplicates. Include your program and the following diagnostic in your report. Thank you.
java.lang.IllegalStateException: endPosTable already set
at com.sun.tools.javac.util.DiagnosticSource.setEndPosTable(DiagnosticSource.java:136)
at com.sun.tools.javac.util.Log.setEndPosTable(Log.java:350)
at com.sun.tools.javac.main.JavaCompiler.parse(JavaCompiler.java:667)
at com.sun.tools.javac.main.JavaCompiler.parseFiles(JavaCompiler.java:950)
at com.sun.tools.javac.processing.JavacProcessingEnvironment$Round.<init>(JavacProcessingEnvironment.java:892)
at com.sun.tools.javac.processing.JavacProcessingEnvironment$Round.next(JavacProcessingEnvironment.java:921)
at com.sun.tools.javac.processing.JavacProcessingEnvironment.doProcessing(JavacProcessingEnvironment.java:1187)
at com.sun.tools.javac.main.JavaCompiler.processAnnotations(JavaCompiler.java:1170)
at com.sun.tools.javac.main.JavaCompiler.compile(JavaCompiler.java:856)
at com.sun.tools.javac.main.Main.compile(Main.java:523)
at com.sun.tools.javac.api.JavacTaskImpl.doCall(JavacTaskImpl.java:129)
at com.sun.tools.javac.api.JavacTaskImpl.call(JavacTaskImpl.java:138)
at org.codehaus.plexus.compiler.javac.JavaxToolsCompiler.compileInProcess(JavaxToolsCompiler.java:125)
at org.codehaus.plexus.compiler.javac.JavacCompiler.performCompile(JavacCompiler.java:169)
at org.apache.maven.plugin.compiler.AbstractCompilerMojo.execute(AbstractCompilerMojo.java:825)
at org.apache.maven.plugin.compiler.CompilerMojo.execute(CompilerMojo.java:129)
at org.apache.maven.plugin.DefaultBuildPluginManager.executeMojo(DefaultBuildPluginManager.java:133)
at org.apache.maven.lifecycle.internal.MojoExecutor.execute(MojoExecutor.java:208)
at org.apache.maven.lifecycle.internal.MojoExecutor.execute(MojoExecutor.java:153)
at org.apache.maven.lifecycle.internal.MojoExecutor.execute(MojoExecutor.java:145)
at org.apache.maven.lifecycle.internal.LifecycleModuleBuilder.buildProject(LifecycleModuleBuilder.java:108)
at org.apache.maven.lifecycle.internal.LifecycleModuleBuilder.buildProject(LifecycleModuleBuilder.java:76)
at org.apache.maven.lifecycle.internal.builder.singlethreaded.SingleThreadedBuilder.build(SingleThreadedBuilder.java:51)
at org.apache.maven.lifecycle.internal.LifecycleStarter.execute(LifecycleStarter.java:116)
at org.apache.maven.DefaultMaven.doExecute(DefaultMaven.java:361)
at org.apache.maven.DefaultMaven.execute(DefaultMaven.java:155)
at org.apache.maven.cli.MavenCli.execute(MavenCli.java:584)
at org.apache.maven.cli.MavenCli.doMain(MavenCli.java:213)
at org.apache.maven.cli.MavenCli.main(MavenCli.java:157)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:483)
at org.codehaus.plexus.classworlds.launcher.Launcher.launchEnhanced(Launcher.java:289)
at org.codehaus.plexus.classworlds.launcher.Launcher.launch(Launcher.java:229)
at org.codehaus.plexus.classworlds.launcher.Launcher.mainWithExitCode(Launcher.java:415)
at org.codehaus.plexus.classworlds.launcher.Launcher.main(Launcher.java:356)
```,
1668,ERROR: Database server does not support the InnoDB storage engine. ,luckyraul, , ,OPEN,build 559ee5bc72c8990c0031cc12,
1667,docker image tests after build image / before push container,trentmswanson, , ,OPEN,I would like to run tests on the image output that was built before it's pushed to docker hub.

Is there something like "after_buildimage" or "before_pushcontainer" events where I can execute some scripts to run and test?

How are folks doing  this today?  Am I missing something?,
1665,/subscriptions/... page consistently takes more than 10 seconds to load,pquentin, , ,OPEN,Shippable has always been slow since we started using it in December 2014. We had hoped speed would have become a priority, but it does not look like it. Our network is not fast, but Shippable still is the exception. The slowest page on shippable for us it the subscriptions page, that list the status of our 39 shippable projects.

 * On load, Firefox reports 151 requests and 2MB page size, for a page load time between 10.71s and 11.24s for a sample size of 5 pages loads.
 * On reload, Firefox reports 159 requests and 2.5MB page size, for a page load time between 12.10s and 13.30s.

The page load seems to be divided in three main parts:

 1. Calls to app.shippable.com. They're *fast*, between 50ms and 500ms and well distributed, so they end after 1.5s at most.
 2. 30 requests to api.shippable.com. They're also fast, but seem to happen sequentially. They end after 8s.
 3. Many calls to api.shippable.com/projects/.../searchBuilds?status=complete&limit=1&isPR=false. They seem to be distributed, but seem to hit what looks like a concurrent requests Firefox limit. They spend more and more time waiting, and the last requests take 3s.

I'd be happy to share anything else that could help debug the issue. I'll send the organization name by email, but continuing the discussion here would be nice. Thank you for shippable!,
1661,Build fails despite 100% test pass and console showing no errors,antonbass, , ,OPEN,This has happened a few times now

user: max_power (via bitbucket)

<img width="1282" alt="screen shot 2015-07-08 at 11 17 03" src="https://cloud.githubusercontent.com/assets/5188810/8567105/ec10a508-2562-11e5-9a86-7c1b832be046.png">

<img width="1264" alt="screen shot 2015-07-08 at 11 18 34" src="https://cloud.githubusercontent.com/assets/5188810/8567118/09538964-2563-11e5-877c-489ff01f7481.png">

,
1660,100% Test passing, 0 Failures, but Build Fails,StephenGrider, , ,OPEN,I'm having an issue with this merge build: https://app.shippable.com/builds/559c634ac8ded311002ab0e5
There are no errors being reported by the UI or the console, yet the build is still being marked as failing.  

In addition, the commit version of the build is getting a `fail to start chrome` error: https://app.shippable.com/builds/559c634bc8ded311002ab0e8

If any other info is required for troubleshooting just let me know :)

Here is my `shippable.yaml`:

```
language: node_js

node_js:
  - 0.12

before_install:
  - source ~/.nvm/nvm.sh && nvm install $SHIPPABLE_NODE_VERSION
  - npm install -g karma-cli gulp-cli coffee-script
  - apt-get update
  - apt-get -y install xvfb chromium-browser
  - test -e /tmp/.X99-lock || /usr/bin/Xvfb :99 &
  - export DISPLAY=:99.0
  - export CHROME_BIN=/usr/bin/chromium-browser
  - coffee -c karma.conf.coffee

install:
  - npm install
  - gulp shippable

script:
  - karma start --browsers Chrome --singleRun true --reporters progress,junit
```,
1654,docker-compose support from shippable with elastic beanstalk,mozinrat, , ,OPEN,Sir, I would like to suggest support for deployment on ELB from a docker-compose.yml,
1653,Network issue ?,xherriot, , ,OPEN,Hello,

• Since we use Shippable, we have some errors and when we re-launch these builds, it works:
<img width="1150" alt="screen shot 2015-07-07 at 09 34 53" src="https://cloud.githubusercontent.com/assets/5094797/8541063/85283784-248b-11e5-9a90-f569e8be0486.png">
<img width="683" alt="screen shot 2015-07-07 at 09 25 49" src="https://cloud.githubusercontent.com/assets/5094797/8541068/8914f6fc-248b-11e5-8772-1a5d0c06351b.png">
<img width="303" alt="screen shot 2015-07-07 at 09 27 29" src="https://cloud.githubusercontent.com/assets/5094797/8541065/86fa961a-248b-11e5-814f-3e15461f5529.png">

• Your UI takes a while to load (~15 seconds on the project page).

• Until "after script", the process takes ~15/20 minutes, it's ok, but then it takes more an hour to finish the build (pull/build/push docker image).

Daily, it really is annoying :(,
1652,UI enhancements,cemo, , ,OPEN,1. After permitting to private repositories I had to click refresh button in UI. This is horrible because I lost a couple minutes to figure out why my private repos could not be displayed.

2. The red error message is displayed on main menu. This is pretty bad because I thought that there was a system error. I could not find add project button anywhere in UI.,
1651,Non-deterministic error messages in tests,hcarver, , ,OPEN,I'm getting random error messages in my tests about end-of-files and broken pipes. When the error occurs, there'll be a whole bunch of subsequent tests that fail (maybe 20% of the total). They often seem to start with integration tests that use Capybara, though they also seem to then affect database access from functional tests that don't use Capybara.

They almost always take the form of
`EOFError: end of file reached` or
`Errno::EPIPE: Broken pipe`.

It's very annoying. It's happening in about 25% of our builds, and means we have to manually re-run builds a lot of the time, which is a pain for our whole CD process.,
1649,Building JavaFX applications = cannot find symbol -> wrong Java 8 version?,stevenschwenke, , ,OPEN,My Maven project located at https://github.com/stevenschwenke/Java8Workshop fails to build because
<code>
[ERROR] COMPILATION ERROR : [INFO] ------------------------------------------------------------- [ERROR] /home/shippable/workspace/src/github.com/stevenschwenke/Java8Workshop/src/test/java/de/stevenschwenke/java/java8workshop/C_09_JavaFX.java:[193,13] cannot find symbol symbol: class Alert location: class de.stevenschwenke.java.java8workshop.C_09_JavaFX [ERROR] /home/shippable/workspace/src/github.com/stevenschwenke/Java8Workshop/src/test/java/de/stevenschwenke/java/java8workshop/C_09_JavaFX.java:[193,31] cannot find symbol symbol: class Alert location: class de.stevenschwenke.java.java8workshop.C_09_JavaFX [ERROR] /home/shippable/workspace/src/github.com/stevenschwenke/Java8Workshop/src/test/java/de/stevenschwenke/java/java8workshop/C_09_JavaFX.java:[193,42] package Alert does not exist [INFO] 3 errors [INFO] ------------------------------------------------------------- [INFO] ------------------------------------------------------------------------ [INFO] BUILD FAILURE 
</code>

In my yml-file, I set language: java and jdk: oraclejdk8.

I think the issue is that a JDK < Java 1.8u40 is used for this build. If so, using a newer version probably solves this issue. Local build with 1.8u45 is fine. How can I set this version in the yml file?

Thanks for your work and keep up the good stuff! :),
1648,docker build success, but no push,luckyraul, , ,OPEN,build id 559a582cce828c0b00372182,
1643,New BitBucket Web Hooks,EHLOVader, , ,OPEN,While looking for causes to #1642 on Bitbuckets end I noticed they recently updated the webhooks.

I know the webhook is added through the API  (at least that is how I remember it happening) but is there anything in the new webhooks which would benefit the process? 

Anyway I thought I would notify in case it helped.

__Documentation for new webhooks__
https://confluence.atlassian.com/display/BITBUCKET/Manage+Webhooks;jsessionid=08DF3E6B619EC3A9815170E2C470BE26.node3
__Blog Post for new webhooks__
https://blog.bitbucket.org/2015/06/24/the-new-bitbucket-webhooks/,
1640,Some PR Commits are being ignored,caarlos0, , ,OPEN,Examples:

- https://github.com/ContaAzul/ContaAzul/pull/8063
- https://github.com/ContaAzul/ContaAzul/pull/8066

The last commit of each of them is not being built and they don't appear in the "Pull Requests" view...,
1629,Clicking github link for failed build logs me into shippable but not the build in question,thockin, , ,OPEN,Clicking the link a second time works better.,
1620,Push tag to docker hub fails,jazmit, , ,OPEN,Using the build number or commit sha as a tag causes the build to fail with 'Tag does not exist':
````
sudo docker push schoolshape/web:master.63 2>/tmp/push-error.txt
The push refers to a repository [schoolshape/web] (len: 1)
sudo: unable to resolve host ip-20-0-100-153
time="2015-06-28T15:21:04Z" level=fatal msg="Error pushing to registry: Tag does not exist: master.63"
````
Reference to a failed build: https://app.shippable.com/builds/55901014cf4aed0c003f8bfe

The push works fine when I specify an existing tag such as 'latest'.,
1618,Any failure in script will not run after_script if the working directory is changed,tanin47, , ,OPEN,Please see: https://app.shippable.com/builds/558f4c03e3b3480d00d90d3f

I have this part:

```
script:
  - cd test
  - errgergerherhhr
```

It produces an error `/home/shippable/cab1dd50-c254-4655-99c5-3ef22e6408d5.sh: line 90: errgergerherhhr: command not found`

The `after_script` is never run.

Is this an expected behaviour?

---------

Ok, I have run a small test on this. It is because I change the working directory to somewhere else. I'm not sure why though.

--------

I've found a workaround:

```
script:
  - (cd test/ && exec some_program_that_errors)
```

In this way, the `after_script` is run properly.

The other combination won't run `after_script`, for example, `(cd test/ && some_program_that_errors)`.

Please let me know if you cannot reproduce it.,
1608,Dynamically change/cancel docker image to build ,xherriot, , ,OPEN,Hello,

I would like to do:
• Commit with tag *-alpha, build my/app:alpha
• Commit with tag *-beta, build my/app:beta
• Commit with tag *, build my/app:latest
• Commit without tag, CI without build a docker image.

To this, I have a bash script executed "after script":
```yml
after_script:
  - 'bash shippable-deploy.sh'
```

Content of this file:
```bash
#!/bin/bash

hash=$(git log -n 1 --pretty=format:"%H")
tag=$(git tag --points-at $hash)

if [[ -n "$tag" ]]
then
    echo "Tag: $tag"
    dir=./shippable/buildoutput/"$tag"/
    echo "Dir: $dir"
    mkdir "$dir"
    cp -R ./dist/* "$dir"
    ln -s "$dir"/../.common/ "$dir"/.common/
    echo "$tag" > ./shippable/buildoutput/"$tag"/version.txt
else
    echo "No tag related, we don't add this to the Docker image."
    echo "Hack for CI without docker build." > ./shippable/buildoutput/shippable.txt
fi
```

Then, the build is processed with my Dockerfile:
```docker
FROM my/app
ADD ./buildoutput/ /var/www/app/
```

Each build tagged add its own directory (named with the tag) to the docker image.
For now, for a commit without tag, I have to do a trick to build the Docker image without send my real build (if no file is added, shippable fail).
And for now, all my builds tagged are send to my/app:latest image (then, its a big image).
(For some reasons, I have to keep each tagged build a certain time)

It would be nice if inside the shippable.yml file, we could do something like that:
```yml
tags:
    *-alpha*:
        docker:
            build: Dockerfile.alpha
            to: my/app:alpha
    *-beta*:
        docker:
            build: Dockerfile.beta
            to: my/app:beta
    *:
        docker:
            build: Dockerfile
            to: my/app:latest
    none:
        docker: 
            build: false
```
Then check if the commit we test is related to a tag.

(Launch a CI when a new tag is send to the repository (without any commit) would be nice too) ,
1601,Guage resource usage on shippable,bprashanth, , ,OPEN,There are times when I've observed what looks like high n/w latencies in my test logs (but could also be eg: cpu starvation of a goroutine pulling stuff off of the network). We run tests in parallel and sometimes things might get loaded. How does one figure out where the resource boundry is? 

I don't mean how many cores I'm supposed to get, I mean how much cpu etc my tests are *actually* using. If i was on a linux box I would use top/ps/sar etc to profile it. I suppose the promise of virtualization is that this should not be necessary, but do you export any time-series machine stats? 

I only want a way to increase the confidence that a flake that the logs say is because of high latency, correlates with spikes of machine stats. I've not had much luck reproducing these spikes locally on n1-standard-1 gce machines.,
1600,Cancelled build = success ,david-anyvan, , ,OPEN,At the moment if a build is started and then cancelled it is treated as a success and triggers after_success. It should be treated as an error or neither,
1595,Build not found.,chechoacosta, , ,OPEN,Since yesterday I'm trying to manually run a build for a branch of my projects and Shippable return me this error message

`{"id":3000,"message":"No access or object not found for :id 558972cdbd5b2a0c00e0cc08","methodName":"chkPermForId|_getBuildById"}`
Build not found.,
1592,Trigger action only if all matrix builds succeeded,rvrignaud, , ,OPEN,Hello,

I would like to be able to trigger an action only if all build from a matrix succeed. Is there a way to do that ?

Thanks,
,
1590,Internal API error : getPermissionByProjectId,timstclair, , ,OPEN,I'm unable to view details for a failed shippable run on https://github.com/GoogleCloudPlatform/kubernetes/pull/10077
(https://app.shippable.com/builds/558376f415e8400d00004fe6)

I get the error:
```
{"id":1005,"message":"Internal API error : getPermissionByProjectId","methodName":"chkPermForId|_getSByProjectId"}
```

This looks like a permissions error (because I'm not a collaborator on the project?), but we would like the results to be public for anyone submitting PRs to the project.

@davidopp
@ixdy,
1587,enable code coverage for golang,rvdwijngaard, , ,OPEN,I have a build configured for go. My configuration is pretty much the same as the sample in your docs. After completing a build successfully, however, there is no code coverage. I added some logging statements to my shippable.yml and at the after-script hook there is a file named ./shippable/codecoverage/coverage.xml. 

The code coverage info of my build does not reflect the results from this file.

What else do I need to configure in order to get this working? ,
1584,Rabbitmq connection refused on port 5672,mcbridejc, , ,OPEN,I've enabled rabbitmq service, but it doesn't seem to be listening on the expected port. This used to work, but is now failing on multiple projects. 

In my shippable.yml: 

```
build_image: shippableimages/ubuntu1404_python
language: python
cache: true
archive: true
python:
  - "2.7"
services:
  - rabbitmq 
```

I'm using pika to connect to rabbitMQ service, with the following connection parameters: 

```
self._connection_parameters = pika.ConnectionParameters(
            host='localhost',
            port=5672,
            credentials=pika.PlainCredentials('guest', 'guest'),
            heartbeat_interval=heartbeat_interval
        )
```

But, it throws this exception: `AMQPConnectionError: Connection to 127.0.0.1:5672 failed: [Errno 111] Connection refused`. 

Seems like something changed with the rabbitmq service? ,
1583,API support to create and destroy Formations,pmoust, , ,OPEN,Great release guys, I believe we will be moving to Formations soon.

Your API needs some (or a lot) love first though.,
1582,Adding system-out and system-err junit xml tags to the Tests page,koszti, , ,OPEN,The test results page displays only the stats and the failure tags from the junit xmls.

Would it be possible to add the system-out and system-err tags too into the report? It would be very helpful to debug some failed tests.

![screen shot 2015-06-21 at 00 32 58](https://cloud.githubusercontent.com/assets/643687/8269728/3c7c0488-17ad-11e5-9f49-b638319ea485.png)
,
1566,Build time,pelegb, , ,OPEN,There seems to be a discrepancy between the build time displayed and summing the time all the build steps took. For example in the following build:
https://app.shippable.com/builds/5581e3033d07800f00060ca9

Your UI says the build time is 7 minutes while summing all the steps times you receive 110 + 4  secs = less than 2 minutes.,
1564,Device out of space,kkarski, , ,OPEN,Here is what I see in my logs:

134a5dcb8ebe: Error downloading dependent layers
63655319ed5c: Error pulling image (latest) from shippable/minv2, endpoint: https://registry-1.docker.io/v1/, ApplyLayer exit status 1 write /home/shippable/.gvm/gos/go1.2/src/pkg/go/token/position_test.go: no space left on device
63655319ed5c: Error pulling image (latest) from shippable/minv2, ApplyLayer exit status 1 write /home/shippable/.gvm/gos/go1.2/src/pkg/go/token/position_test.go: no space left on device
time="2015-06-17T20:20:01Z" level="fatal" msg="Error pulling image (latest) from shippable/minv2, ApplyLayer exit status 1 write /home/shippable/.gvm/gos/go1.2/src/pkg/go/token/position_test.go: no space left on device",
1563,Python 3.4 failure,pmacosta, , ,OPEN,I have a repo with a Python package that is passing all tests with Python 2.7 but it is failing with Python 3.4. The failure is not on the tests, the process does not even get there. As part of the test process I build the package with "python setup.py sdist". With Python 3.4 I get

ImportError: No module named 'setuptools' 

Build information: https://app.shippable.com/builds/5581b5c93d07800f0005bc56

I poked around and there doesn't seem to be a proper Python 3.4 tool chain (pip, py.test, etc.) in the minion. Do I have to activate a virtual environment or something like that? What is the best practice for Python 3.4?

Thanks!

Pablo,
1562,New version problems,caarlos0, , ,OPEN,- cannot stop/restart from the build view
- cannot download build log
- log view doesn't auto-update
- seems like the build distribution across custom nodes is broken (feels like it send almost all builds to a single machine)
- cannot see initialization logs on custom nodes view
- some pull requests simply don't appear in the PR list,
1557,Shippable does not display phpunit test results...,pennyarcade, , ,OPEN,Hello Team,

I followed the documentation very closely to set up phpunit tests (among others) with shippable using the '--log-junit' option and the '--coverage.xml' option. 
I am using the standard image for now.
My coverage results are displayed correctly and the phpunit output looks ok, too. (Failing as expected)
Sadly I can't seem to get the phpunit results displayed in the testresults tab. I tried different filenames and triplechecked output directory and format. Please help.

Also my composer update fails when I turn caching on. It gets a permission denied on the cache directory. File permissions are OK, there might be an Issue of files already being open that I can't reproduce locally. I am trying to test a small Symfony application. The intendet result should be saving build time by caching but keeping dependencies up to date.
Update: This only happens for the buildin global composer. I have no errors when using my own composer.phar in the project directory.

Third and least important, I'd like to also display my code sniffer results of PHPMD and PHPCS in the testresults tab. The output is set to xml, too. 

Here is an example build with the above problems : https://app.shippable.com/builds/55812bf06cde2e0c0077a5b7
(Build 43)

Here is my shippable.yml for reference:

```yaml
cache: true

# language setting
language: php

# php tag
php:
    # only test one php version until problems are solved.
    #- 5.3
    #- 5.4
    #- 5.5
    - 5.6
    #- hhvm

before_install:
    - cd imgboard
    # make sure the directories are writable (this shouldn't be nessessary and doesn't solve the problem!)
    - sudo chmod 777 -R app/cache
    - sudo chmod 777 -R app/logs
    # update composer (works fine)
    - php ./composer.phar selfupdate

install:
    # update dependencies (fails with obscure permission denied error when using builtin composer)
    # composer update
    - php ./composer.phar update
    - cd ..

before_script:
    # make sure the result folders exist
    - mkdir -p shippable/codecoverage
    - mkdir -p shippable/testresults
    - cd imgboard

script:
    # don't fail for code smells (|| true)
    # PHP mess detector HTML output for offline use
    - bin/phpmd src/ html cleancode,codesize,controversial,design,naming,unusedcode --reportfile ../shippable/testresults/phpmd.html --strict || true
    # PHP mess detector xml output to be displayed in testresults tab
    - bin/phpmd src/ xml cleancode,codesize,controversial,design,naming,unusedcode --reportfile ../shippable/testresults/phpmd.xml --strict || true
    # PHP CS text output for offline use only until above works
    - bin/phpcs --report-file=../shippable/testresults/phpcs.txt src/ || true
    # PHPUNIT Tests anc code coverage in different formats for display in testresults tab and for offline use
    - phpunit -c app/ --log-junit ../shippable/testresults/phpunit.xml --coverage-xml ../shippable/codecoverage --coverage-html ../shippable/codecoverage --testdox-html ../shippable/testresults/junit.html

after_script:
    # Prove the files were written correctly
    - cd ..
    - ls -a shippable/testresults
    - cat shippable/testresults/junit.xml
    # check in test results for offline use 
    - git add shippable
    - git add imgboard/composer.*
    # suppress infinite loop on ci
    - git commit -m 'update test results [skip ci]'
    - git push

notifications:
    email: false

```

Thank you very much for your help.

Martin,
1550,build timeouts,magwas, , ,OPEN,I have a build which consists of some python unit tests followed with some selenium tests in python for chrome and firefox. For the selenium tests I run an apache on ports 8888 and 8889, and a fake smtp server on 1025, all on 127.0.0.1. The build runs nicely on travis. When it finishes on shippable, it runs for 13 minutes.
However there are a lot of cases when it just times out.
I suspect that it has something to do with selenium.
Build history:
https://app.shippable.com/projects/55771730edd7f2c0520e8d4c,
1543,PHP 7 support,hranicka, , ,OPEN,The first public alpha of PHP 7 was released.
http://php.net/archive/2015.php#id2015-01-11-6

Do you plan support for early PHP 7 versions?
Or waiting for final builds?

Thank you for support!

There is only current stable versions support and HHVM:
http://docs.shippable.com/languages/#php,
1542,About ~50% of builds fail,nZac, , ,OPEN,I keep on getting an issue with builds failing during setup of the database.  The same build run two different times will fail once and succeed the next.

The lines that seem to fail are:
```yaml
- sudo echo "<REDACTED>" > /etc/postgresql/9.3/main/pg_hba.conf
- sudo echo "<REDACTED>" >> /etc/postgresql/9.3/main/pg_hba.conf
```,
1541,Your disk is out of space. Free some space to be able to install your bundle.,joshuabremer, , ,OPEN,Hi all, about 50% of our builds are running out of space. Is this a configuration we need to set? It surprises me because it's just a Rails app. There are some large gems like libv8, but they're not *that* big. Also it seems weird  that it's not happening every time.  Ideas?

https://app.shippable.com/builds/55792e757915e70b005962c5,
1540,Request for FQN in coverage report,jariwiklund, , ,OPEN,I would really like to have FQN (Fully Qualified Name) in coverage reports.
I'm using phpunit as examplified:
phpunit --log-junit shippable/testresults/junit.xml --coverage-xml shippable/codecoverage --configuration phpunit.xml
But my coverage report only shows the class name and I have a few Classes called Controller(in different namespaces), so from the coverage report, it is impossible to know which of the Controller classes is covered.,
1538,Unable to use phantomJS,jamesjwarren, , ,OPEN,Our FE test suite, like many I'm sure, runs with PhantomJS. For some reason I am unable to get this working within shippable despite it installing ok.

I receive the following error:
```
PhantomJS threw an error:ERROR
>> /root/workspace/src/.../node_modules/grunt-contrib-jasmine/node_modules/grunt-lib-phantomjs/node_modules/phantomjs/lib/phantom/bin/phantomjs: error while loading shared libraries: libfreetype.so.6: cannot open shared object file: No such file or directory 0 [ '/root/workspace/src/.../node_modules/grunt-contrib-jasmine/node_modules/grunt-lib-phantomjs/node_modules/phantomjs/lib/phantom/bin/phantomjs: error while loading shared libraries: libfreetype.so.6: cannot open shared object file: No such file or directory' ]
>> In order for this task to work properly, PhantomJS must be installed locally
>> via NPM. If you're seeing this message, generally that means the NPM install
>> has failed. Please submit an issue providing as much detail as possible at:
>> https://github.com/gruntjs/grunt-lib-phantomjs/issues
```

My shippable.yml config is something along the lines of:
```
# Language setting
language: node_js

# Node version
node_js:
 - 0.12

# Specify the build_image
build_image: shipimg/ubuntu1404_nodejs

services:
 - selenium

# Setup node.js environment
before_install:
 - source ~/.nvm/nvm.sh && nvm install $SHIPPABLE_NODE_VERSION
 - node --version
 - npm install -g bower
 - npm install -g grunt-cli
 - npm install -g protractor && webdriver-manager update

# Install the required dependencies
install:
 - npm install

# Running npm test to run your test cases
script:
 - grunt test:development
 - grunt test
 - grunt e2e --ci
```

Example build experiencing this issue: https://app.shippable.com/builds/55b559e84897fe0b00f5a87c,
1536,Exclude Tag / Branch names with wildcards,TheEadie, , ,OPEN,Is it possible to exclude tags for building via wildcards eg. To not build anything beginning "bulid/" or "release/"

I've tried using the travis style regex in the excludes section of the shippable.yml but this doesn't seem to prevent the tag triggering a build.

```
branches:
  except:
    - /^build.*$/
```,
1535,~50% Build Failures,jchamberlain, , ,OPEN,Hello,

I'm using a dedicated host, and in the last week or so I've been having at least half of builds fail for seemingly random reasons, including:
- Failing to connect to the Docker registry ("tcp: lookup registry-1.docker.io on 69.20.0.164:53: no such host")
- Doing nothing at all. All it ever shows is the "save reports" step. This is probably the most common error.
- Timing out with no message.

Has anything changed?

Update: I just deployed a fresh server for shippable, and nothing's changed.,
1534,constantly getting throttled from github,spyrospph, , ,OPEN,I am usually getting the below error message when tryign to log in through github.

This usually occurs every GMT morning.

failed to fetch user profile (status: 403 data: {"message":"API rate limit exceeded for spyrospph.","documentation_url":"https://developer.github.com/v3/#rate-limiting"}) at /home/shippable/fe/node_modules/passport-github/lib/passport-github/strategy.js:90:28 at passBackControl (/home/shippable/fe/node_modules/passport-github/node_modules/passport-oauth/node_modules/oauth/lib/oauth2.js:123:9) at IncomingMessage.<anonymous> (/home/shippable/fe/node_modules/passport-github/node_modules/passport-oauth/node_modules/oauth/lib/oauth2.js:142:7) at IncomingMessage.emit (events.js:117:20) at _stream_readable.js:944:16 at process._tickCallback (node.js:448:13)

-S.,
1530,Shippable hangs with no response when running velocity tests with Meteor.js,imarichardson, , ,OPEN,I am trying to run shippable against my meteor.js app.
My tests are in cucumber running via meteor's testing architecture, velocity. One of the things core to testing in meteor is that the app will need to run a cloned instance of itself so the tests can actually run. When I run the test my app perpetually runs; I've decided to not stop the latest build to see how long it will run and at this point it appears to be around 35 minutes. The build continues to show processing with no feedback.

Account name: imarichardson
Project: ve-client-meteor
See build 8. 

Here is my config:
https://gist.github.com/eac8b92e37942f9bbfe6

Thanks.,
1529,Branch name with semantic versioning skipped,andrerom, , ,OPEN,Hi,

we have a branch name called "5.4" and shippable keeps on skipping it, even if we added a shippable.yml in master to say this branch should be tests. older branches in format like "stable-5.3" does not have this issue.

i.e.:
```
Webhook Received for a Branch Excluded by YML in ezsystems/ezpublish-kernel-ee  Today at 11:32 PM Low

Webhook received for branch 5.4, which is not included in branches: only tag in YML.
```

No idea why.,
1527,Cache builds failing at 'commit for cache' stage,adrianbk, , ,OPEN,https://app.shippable.com/builds/55743aeb0d553e0c002eb819

It appears as if the `commit for cache` stage is not being executed and fails the build, despite the build script succeeding. Any idea what's happening here? 

__Failing build__
![image](https://cloud.githubusercontent.com/assets/1910629/8024278/41c8da7e-0d71-11e5-8149-d30b16d736c6.png)

__Successful build__
![image](https://cloud.githubusercontent.com/assets/1910629/8024287/a81b485c-0d71-11e5-80db-3fed83a62227.png)
,
1525,Build failing because of error saving reports,adrianbk, , ,OPEN,Hi, 

I'm on the startup plan and my build is consistently failing when saving test reports:

![image](https://cloud.githubusercontent.com/assets/1910629/8019950/90dabac8-0caa-11e5-8117-4da59f483a3a.png)

Is there an upper limit on the the size of the test reports or the number of test results that shippable can handle?

Here's my build config: https://github.com/adrianbk/gradle/blob/shippable/shippable.yml,
1524,Do you support Packer builds?,gamename, , ,OPEN,Hi,

Do you support [packer](https://www.packer.io/) builds? If so, which platforms? I'm particularly interested in [vagrant](https://www.vagrantup.com/) builds for virtualbox and amazon aws.

Thanks,
-T,
1523,Any known issues with mocha-junit-reporter over-reporting number of passed tests?,tinisi, , ,OPEN,Folks,
Loving your service, getting very close to a useful CI solution.

We have a node app with about 1100 server side mocha tests.

Ever since we have switched to using mocha-junit-reporter with your service, the total number of tests shown in your web report is much inflated, by about a factor of 20.

When I run the tests locally with tap reporter, I see what I would expect.

We have tried a number of different set ups, letting mocha discover all our tests in a folder, and using a shell script and the mocha --grep option to run portions of the tests, but the number of passed tests is always high.

The failures seem to be reported correctly.

Are there any known issues that might cause this?

Thanks!
Tinisi,
1521,Build timeout and some errors by build Android Project,mario-naether, , ,OPEN,Hi , recently I get some errors on build my Andriod Project.

https://app.shippable.com/projects/5551c34aedd7f2c052e982f2

Command "sudo apt-get update" output "W: GPG error: http://ppa.launchpad.net precise Release: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY B70731143DD9F856" 


Command: "echo yes | android update sdk --filter platform-tools --no-ui --force > /dev/null" outputed "Error: Ignoring unknown package filter 'platform-tools'"

Command: "echo yes | android update sdk -a -u -t sys-img-armeabi-v7a-android-10" outputed "Error: Ignoring unknown package filter 'sys-img-armeabi-v7a-android-10'  "

ShellScript "android-wait-for-emulator" outputed "
Command timed out
Command thread timed out"


Can anyone tell me what is wrong on my shippable config?,
1520,Git sync clones into wrong directory when pulling from bitbucket repo,vially, , ,OPEN,The `git_sync` step clones the repo into: `~/workspace/src/bitbucket.com` instead of `~/workspace/src/bitbucket.org`

This is causing issues with the Go language because the import paths have to match the file locations on disk.

I know you can fix it by defining the `SHIPPABLE_REPO_DIR` global env var but I think that should be the default behaviour so I don't have to clutter the `shippable.yml` file for every Go project
```yaml
env:
  global:
    - SHIPPABLE_REPO_DIR="$HOME/workspace/src/bitbucket.org/$REPO_NAME"
```

Related issues: #1226 ,
1515,How To Revert To build_image In shippable.yml?,jchamberlain, , ,OPEN,The settings panel under a project warns that its settings will override those in the shippable.yml. The build image, however, cannot be left empty. I must have previously set a build image, because I can't get it to use the one specified in my shippable.yml and am instead stuck setting it project-wide in the settings panel. This is a problem because I need to use a different image for one branch of my code, as reflected in its shippable.yml. Thoughts?,
1513,got unexpected error for a mysql sentence when executing before script,richhl, , ,OPEN,don't know what happens, mysql sentence is well formed! thx in advance. 

 before_script ?
 composer install -n
 mysql -e "CREATE USER 'ticketuser'@'localhost' IDENTIFIED BY 'evarV3';"
 mysql -e "GRANT ALL PRIVILEGES ON *.* TO 'ticketuser'@'localhost';"
/home/shippable/862d9511-f5db-407d-a31a-fd6ccd4a2cac.sh: eval: line 85: unexpected EOF while looking for matching `"'',
1510,Frequent issues creating a virtualenv,prabhuramachandran, , ,OPEN,Been using shippable for a while now and of late I have found that several of my builds routinely fail with a strange virtualenv error.  Here is an example:

https://app.shippable.com/builds/556cbd0e20addd0f00f67a18

If you notice on this: https://app.shippable.com/projects/540e849c3479c5ea8f9f030e

There are several builds that have failed with a similar error.  My shippable.yml has not changed in any significant way in many months.,
1508,build keeps hanging on "pull",fommil, , ,OPEN,I've restarted this build a few times, but it keeps hanging during the docker image pull, can you please have a look?

https://app.shippable.com/builds/556c33ed20addd0f00f614b3,
1507,Support clover html coverage (phpunit),eventbase-web, , ,OPEN,Are there any plans to support html coverage reports?

We've been using a local install of Jenkins for our PHP CI builds, but *much* prefer Shippable for a variety of reasons. Right now the XML coverage reports are the only deal breaker between me and moving us wholesale to Shippable.,
1506,docker image user disappeared,tal-moshayov, , ,OPEN,I'm building using a custom docker image that has a postgres user pre-created in it. 
When I try to use that image in shippable, that user does not exist (invalid user: 'postgres').
What's going on here? Is shippable changing the target docker? 
Any hint for debug?,
1505,Cannot connect to port on dedicated host & custom image,RobertHerhold, , ,OPEN,I bought dedicated host support and created and pushed a public docker image based off of this repo: https://github.com/shippableImages/mongoBase.

I am able to run the container and connect to it just fine on my local machine (port 27017). Is there something special I need to do either on my dedicated host or in my custom image to resolve this issue?

Running `mongo` (Mongo shell) returns 
```
connecting to: test
couldn't connect to server 127.0.0.1:27017...
exception: connect failed
```

I have tried this with a separate database, different port, and different docker images but have never been able to connect.

With this being an issue, dedicated hosts are absolutely useless. Please advise.,
1504,How do I bind a docker build to a database addon?,tal-moshayov, , ,OPEN,I'm using a custom docker image for my build, which requires a db (postgres) to run its tests. So I added:
`addons:  
  postgresql : "9.3"
` to shippable.yml.
Now, assuming shippable kickstarts the database addon in the background, **what is the db host name and port I should bind to?**  
It appears that on a shippable predefined image the addon is automatically bound to localhost, but it doesn't seem to be this way with a custom docker image.,
1503,Permission denied when running COPY FROM queries against PostgreSQL,netj, , ,OPEN,It seems postgres has no permission to access files under /home/shippable/workspace/src/github.com/.  Our tests include `COPY FROM` queries and they are not being able to pass.  See https://app.shippable.com/builds/5569a020e9aab00d00d504df.,
1499,Encrypted env vars not available,sengenbe, , ,OPEN,I have a Rails application in which all builds are failing due to the env vars not being available to a gem called Figaro, which throws errors when required variables are not present.  I followed instructions and verified I used the correct organization and my yml file looks fine.  Are the variables not made available at this step?,
1497,Docker cleanup affects non-shippable containers,rogierslag, , ,OPEN,We run our build server now just on one of the other docker hosts (thats what docker is for right!) but it appears then the "cleaning" of old containers also will try to remove any container. So also containers which should not be cleaned by shippable (such as other regular development stuff).

Can it please be possible to disable this? Using docker we do no longer need static ci servers. Please don't replace those static CI servers with static docker servers!,
1490,Build Dockerfile and push to Quay.io,howarddierking, , ,OPEN,Hey all - sorry if this question is a repeat, but in reading through the other threads, I'm having a hard time teasing apart the requests for custom _build_ containers vs. support for building container images from my project artifacts. It is the latter scenario that I'm interested in.

I need to support the following workflow:

* Build the project (e.g. current workflow)
* Build an image from the Dockerfile (need to also be able to tag it with the build number)
* Login to Quay.io (robot account)
* Push image to Quay.io

Maybe down the road I'll bake in automated deployment, but because we do deployments at the entire auto-scaling group level and need to support multi-container orchestration, I'll probably continue to do this out of band for some time.

So, is this workflow supported in Shippable today?,
1489,API errors,pmoust, , ,OPEN,Trying to get  current running builds while testing a new client for shippable API. 
I get an error response, to reproduce:

```
curl -H "Authorization: apiToken $apiToken"  https://api.shippable.com/projects/54e2287a5ab6cc13528c39d2/runningBuilds
```

response:
headers:
```
HTTP/1.1 500 Internal Server Error
Access-Control-Allow-Credentials: true
Access-Control-Allow-Headers: Accept,Content-Type,Authorization,Cookie
Access-Control-Allow-Methods: GET,PUT,POST,DELETE
Access-Control-Allow-Origin: *
Content-Type: application/json; charset=utf-8
Date: Wed, 27 May 2015 16:00:57 GMT
ETag: "1123644452"
X-Powered-By: Express
Content-Length: 155
Connection: keep-alive
```
respBody:
```json
{
  "id": 9999,
  "methodName": "shipError",
  "link": {
    "name": "MongoError",
    "$err": "Can't canonicalize query: BadValue unknown top level operator: $nin",
    "code": 17287
  }
}
```,
1488,Feature Request: Coverage,Nalum, , ,OPEN,On the coverage view can you show the extra data that is provided?

Example: Python coverage goes into detail about which lines are covered, partially and not covered.

It would be nice to see this in the coverage tab and not have to either create the HTML output from coverage or view the console output to get that information.,
1486,AWS Opsworks integration,thiagonache, , ,OPEN,Hi there,

I'm integrating shippable with opsworks, but I could not find a way to create deployment after container push. Is that possible?
I deploy the container on opsworks, so if I use after_success I'm not sure if deployment will run before or after the push to dockerhub.,
1485,Managing invites on new projects,sdlarsen, , ,OPEN,When creating a new project, invites are sent to everyone who has/had access to the repositories and has been signed in to Shippable (I guess). How can we manage this when people are leaving a team (or a company)?,
1482,PhantomJS not starting anymore. I haven't changed my configuration.,nathanmarks, , ,OPEN,Hey guys,

I haven't touched any configuration related to PhantomJS, but it's suddenly not starting in my builds anymore.

```
WARN [launcher]: PhantomJS have not captured in 60000 ms, killing.
INFO [launcher]: Trying to start PhantomJS again (1/2).
WARN [launcher]: PhantomJS have not captured in 60000 ms, killing.
INFO [launcher]: Trying to start PhantomJS again (2/2).
```

Example failed build: https://app.shippable.com/builds/55633688d88a130e00213bbb
Example success build: https://app.shippable.com/builds/555617d5d81dea0d0039ebed,
1481,Google App Engine deployments failing from the minion,aproxs, , ,OPEN,I have a deployment script in my repo "build/deploy_app.sh", with the following line:

echo "$SECRET" | $GAE_DIR/google_appengine/appcfg.py -e "$EMAIL" --passin update moduleone1.yaml moduletwo.yaml

where:
 - SECRET and EMAIL are secure variables and GAE_DIR is correct.

If I run exactly that command in my local computer, the deployment goes fine
However, when the minion runs that command I get the following output:

03:15 AM Host: appengine.google.com
03:15 AM Application: applicationID; module: moduleone; version: 1
03:15 AM 
Starting update of app: applicationID, module: moduleone, version: 1
03:15 AM Getting current resource limits. 

an hour after, the deployment times out, but I get nothing else.

I have no clue what's going on... :(. Is this isolated to me? or is it a known issue?,
1480,Install fails with 403,Nalum, , ,OPEN,The install phase of the build is failing with:
```
package bitbucket.org/[group]/[repo]/[sub-module]: https://api.bitbucket.org/1.0/repositories/[group]/[repo]: 403 FORBIDDEN
```

This is a private repo do I need to do something to allow access?

Here is my `shippable.yml` file:

```
language: go

go:
  - 1.4

env:
  - CI_REPORT=shippable/testresults/junit.xml

before_install:
  - export GOPATH=$SHIPPABLE_GOPATH
  - source $HOME/.gvm/scripts/gvm
  - gvm install go$SHIPPABLE_GO_VERSION
  - gvm use go$SHIPPABLE_GO_VERSION
  - go get github.com/t-yuki/gocover-cobertura
  - go get github.com/onsi/gomega
  - go get github.com/onsi/ginkgo
  - go get golang.org/x/tools/cmd/cover

# Make folders for the reports
before_script:
  - mkdir -p shippable/testresults
  - mkdir -p shippable/codecoverage

script:
  - go test -coverprofile=coverage.txt -covermode count ./
  - $GOPATH/bin/gocover-cobertura < coverage.txt > shippable/codecoverage/coverage.xml
```,
1478,git submodules false has no effect,sbalko, , ,OPEN,Hi,

I have a project hosted on GitHub with several submodules. For the shippable build, I do not want these submodules to be fetched, so I included a

```
git:
   submodules: false
```

clause into ```shippable.yml```. However, that does not seem to have any effect. I continue to see the following build steps:

```
 ssh-agent bash -c 'ssh-add /root/.ssh/project_rsa; git clone git@github.com:sbalko/libav.js.git /root/workspace/src/github.com/sbalko/libav.js'
 git config --get user.name || git config user.name 'Shippable Build'
 git config --get user.email || git config user.email 'build@shippable.com'
 rm -rf /root/workspace/src/github.com/sbalko/libav.js/shippable
 mkdir -p /root/workspace/src/github.com/sbalko/libav.js/shippable/testresults
 mkdir -p /root/workspace/src/github.com/sbalko/libav.js/shippable/codecoverage
 mkdir -p /root/workspace/src/github.com/sbalko/libav.js/shippable/buildoutput
 git checkout -f dev --
 git reset --hard f78c31dbe932080e7811bcb780e4dc73287bd87a
 git submodule init
 git submodule sync
```

Why are the latter two steps happening? Isn't that what is supposed to be suppressed by the "submodules: false" clause?

Any help would be greatly appreciated!
Soeren
,
1463,Can't build with local import paths for Go,boj, , ,OPEN,The following import:

    import "myproject/subpath"

Fails with:

    import "myproject/subpath": import path does not begin with hostname

If you try the following:

    import "github.com/boj/myproject/subpath"

The build will fail on a private project because it expects to be able to import it like it was public.

My shippable.yml:

```
language: go

go:
  - 1.4

script:
  - export GOPATH=$SHIPPABLE_GOPATH
  - source $HOME/.gvm/scripts/gvm
  - gvm use go$SHIPPABLE_GO_VERSION
  - go get -d -v ./... && make integration

services:
  - rethinkdb
```,
1453,"No tests found.[test(AVD) - 4.4.2]" on android unit testing exampleApp,mario-naether, , ,OPEN,Hi,
i has a small issue on Shippable.
I write a Sample Android app and hostet on bitbucket. (https://bitbucket.org/mnaether/sampleapp)
my builds failed in the shippable Project with following error on command:
"gradlew connectedAndroidTest build"

```
com.android.builder.testing.ConnectedDevice > No tests found.[test(AVD) - 4.4.2] FAILED 

No tests found. This usually means that your test classes are not in the form that your test runner expects (e.g. don't inherit from TestCase or lack @Test annotations).

:app:connectedAndroidTestDebug FAILED


FAILURE: Build failed with an exception.


* What went wrong:

Execution failed for task ':app:connectedAndroidTestDebug'.

> There were failing tests. See the report at: file:///home/shippable/workspace/src/bitbucket.com/mnaether/sampleapp/app/build/reports/androidTests/connected/index.html
```

All tests running completely on my Local Android Stuido.
What is wrong on my test or shippable config?,
1450,save_coverage_reports error,jonmer85, , ,OPEN,Hi,
I recently changed some of my coverage configuration to include some omissions of files that I don't want to count towards my test coverage. I now see that the save_coverage_reports step is failing and I don't get any errors to help me determine the problem. I confirmed that the changes on my local environment work just fine and the produced coverage files are of similar format.

The build in question is https://app.shippable.com/builds/555413c64240ea0b00c2aad6

Thanks,
Jon,
1444,Mock not working in Python,lfrodrigues, , ,OPEN,Hi,

I'm using 
```
@mock.patch('storages.backends.s3boto.S3BotoStorage', FileSystemStorage)
```
In my test code.

Locally my test code is working fine but when it is run at Shippable mock doesn't patch correctly and S3BotoStorage gets used. Any idea what could be wrong?
,
1434,Build number format for docker images,rvdwijngaard, , ,OPEN,Hi, 

According to your blog article http://blog.shippable.com/immutable-containers-with-version-tags-on-docker-hub all pushed docker images are provided with a tag which contains the build number. 

My images are tagged with the branchname + build number (without the revision number).

In my yaml I want to update some metadata file in s3 with the docker image version: 
- sed -e "s;%BUILD_NUMBER%;$BUILD_NUMBER;g" ecs/machinelogservice.json > ecs/tmp.json

The build number contains the build number + revision (in my case 60.1). 

Is there any environment variable which contains the tag / version info of my docker image? 
,
1430,Build status emails in plain text?,rollcat, , ,OPEN,Currently build status emails are delivered as HTML-only. HTML is more difficult to parse; also some MUAs handle it quite badly. Adding a plain text part to a MIME message would let it degrade gracefully.
,
1420,Handling multiple projects in a single repository,marcua, , ,OPEN,It looks like `shippable.yml` must be in the root of the git repo.  Say I have one large repository with multiple codebases (e.g., two django projects, or some python code and some js code), and for concreteness let's say they are in `gitrepo:/python_project1` and `gitrepo:/python_project2`.  The questions that arise for me are:

  - How do I get tests in both directories to run---can I have multiple `script` and `install` tags for each project?
  - Will the coverage and test reports for each project be reported separately?

If you have an example `shippable.yml` that handles this case, I'd be very grateful!
,
1417,rails selenium specs failing though they are passing on local dev,murtyk, , ,OPEN,Build ID: murtykorada / wtracker / Build Group: 34

When full suite of specs are run, all js: true specs are failing. Just to show the isolated problem, ran only one spec:

    require 'rails_helper'
    describe 'signin page' do
      it 'no js' do
        visit '/login'
        expect(page).to have_text 'Remember Me'
      end
      it 'with js', js: true do
        visit '/login'
        expect(page).to have_text 'Remember Me'
      end
      it 'full url with js', js: true do
        visit 'http://www.localhost.com:7171/login'
        expect(page).to have_text 'Remember Me'
      end
    end

both the js: true specs are failing with Capybara::ElementNotFound: Unable to find xpath "/html" 

````shippable.yml:
# {Rails.root}/shippable.yml
# shippleable yml for CI
language: ruby

cache: true

archive: true

rvm:
  - 2.2.0

hosts:
  - www.localhost.com
  - brookdale.localhost.com
  - njit.localhost.com

env:
  - CI_REPORTS=shippable/testresults COVERAGE_REPORTS=shippable/codecoverage CAPYBARA_PAGE_PATH=shippable/capybara
  - secure: mykeys
services:
 - mongodb
 - selenium

# ensure the test output and coverage dirs are created
before_script:
  - mkdir -p shippable/testresults
  - mkdir -p shippable/codecoverage
  - mkdir -p shippable/capybara
  - cp config/shippable.database.yml config/database.yml
  - psql -c 'DROP DATABASE IF EXISTS wtracker_test;' -U postgres
  - psql -c 'create database wtracker_test;' -U postgres
  - RAILS_ENV=test rake db:migrate
  - RAILS_ENV=test rake testprep:load_data
  - "export DISPLAY=:99.0"
  - "/etc/init.d/xvfb start"
# write the rspec tests to the output dir
script:
  # - rspec -f JUnit -o shippable/testresults/results.xml
  - rspec --format RspecJunitFormatter -o shippable/testresults/results.xml spec/features/js_and_no_js_spec.rb

notifications:
    email:
      recipients:
        - myname@yahoo.com
````

````ruby
source 'https://rubygems.org'
group :production, :staging, :integration do
  ruby ENV['CUSTOM_RUBY_VERSION'] || '2.2.0'
end

gem 'rails', '4.2.0'
gem 'protected_attributes', '1.0.7'
# Bundle edge Rails instead:
# gem 'rails', :git => 'git://github.com/rails/rails.git'
# gem 'bootstrap-sass', '~> 2.1.0.0' #trying new one

group :development, :test do
  gem "thin"
end
group :production, :staging, :integration do
  gem 'unicorn', '4.8.3'
  gem 'unicorn-worker-killer'
  gem 'rails_12factor', '0.0.2'
end

gem 'pg'
gem 'mongo'
gem 'bson_ext'
gem 'mongoid', github: 'mongoid/mongoid'

gem 'attr_encrypted', '1.3.3'
gem 'draper', '1.3.1'
gem 'simple_form', '~> 3.1.0'
gem 'devise', '~> 3.4.1'
gem 'gmaps4rails', '1.5.6'
gem 'fusion_tables', '0.4.1'
gem "zip-zip", "~> 0.2"
gem 'rubyzip', '1.0.0'
gem 'roo', '1.12.1'
gem 'geocoder', '1.2.6'
# gem 'google_places', '0.16.0'
gem 'mechanize', '~> 2.5.1'
# gem 'ox', '2.0.0'
gem 'httparty', '0.12.0'
gem 'rest-client', '1.6.7'
gem 'aws-sdk', '1.33.0'
gem 'fuzzy-string-match_pure', '0.9.5'

gem 'will_paginate', '3.0.3'
gem 'bootstrap-will_paginate', '0.0.6'

gem 'pundit', '0.2.1'
gem 'newrelic_rpm'
gem 'memcachier', '0.0.2'
gem 'dalli', '2.6.4'
gem 'icalendar', '1.4.3'
gem 'humanizer'
gem 'ransack', github: 'activerecord-hackery/ransack', branch: 'rails-4.2'

gem 'delayed_job_active_record'
gem 'clockwork'

# Gems used only for assets and not required
# in production environments by default.
gem 'sass-rails', '~> 5.0'
gem 'bootstrap-sass', '~> 2.3.0.0'
gem 'bootstrap-datepicker-rails', '1.1.1.6'
gem 'font-awesome-sass-rails', '3.0.2.2'
gem 'coffee-rails', '~> 4.1.0'

# See https://github.com/sstephenson/execjs#readme for more supported runtimes
# gem 'therubyracer', :platforms => :ruby

gem 'uglifier', '>= 1.3.0'
  # gem 'jquery-ui-rails'

group :development do
  gem 'rack-mini-profiler'
  gem 'web-console', '~> 2.0'
end

group :test, :development do
  gem 'byebug'
  # gem 'debugger'
  # gem 'wdm', :platforms => [:mswin, :mingw], :require => false
end

gem 'quiet_assets', group: :development

group :test do
  gem 'rspec-rails', '~> 3.1'
  gem 'rspec-retry'
  gem 'factory_girl_rails'
  gem 'selenium-webdriver', '2.43.0'
  gem 'headless', '1.0.2'
  gem 'capybara', '~> 2.4.1'
  gem 'capybara-screenshot'
  # gem "capybara-webkit"
  gem 'faker', '1.0.1'
  gem 'guard-rspec', '~> 4.2.10'
  gem 'shoulda'
  gem 'launchy'
  gem 'database_cleaner'
  gem 'webmock', '1.11.0'
  gem 'vcr', '2.9.2'

  gem 'simplecov' #, require: false
  gem 'simplecov-csv'
  gem "rspec_junit_formatter"
end
gem "parallel_tests", group: :development
gem 'jquery-rails', '4.0.3'
gem "jquery.fileupload-rails", '1.11.0'
````,
1405,Automated deployment to Digital Ocean,naveensky, , ,OPEN,Hi,

I am looking forward to setup automated deployment to Docker instances on Digital Ocean. Do you have a tutorial/setup guide in regards to the same.

I need to deploy a Java app

Naveen,
1404,gulp: ../deps/uv/src/unix/core.c:422: uv__close: Assertion `fd > 2' failed.,andrewmichaelsmith, , ,OPEN,A gulp task to SCP some files after a successful load now consistently fails.

For the past 4 months this task has worked perfectly for over 80 builds. Now the past 3 builds have all failed. I have changed nothing on the host machine and the tasks succeeds from other servers.

The task:

```
gulp.task('upload', shell.task([                                                 
    'chmod 0600 id_rsa',                                                         
    'scp -iid_rsa -rpv www/* user@host:/path/'           
]));
```

The error:

```
[13:25:56] Starting 'upload'...
gulp: ../deps/uv/src/unix/core.c:422: uv__close: Assertion `fd > 2' failed.
[13:25:56] 'upload' errored after 228 ms
```,
1399,Transparant caching in Shippable,schmmd, , ,OPEN,It's great to have caching in Shippable because it speeds our builds by more than 50%.  Shippable must have a slow connection to Maven Central because downloading dependencies is **slow**.  And downloading all dependencies for each build is insane for a complex JVM application.

However, Shippable's caching mechanism is far from perfect.  It's clear you have sacrifice development costs for product quality.  At our company, our caches frequently need to be invalidated because dependencies change and builds start to slow down again because of the new dependencies.  However, it's hard to stay on top of that, and it's yet another thing I need to educate the development team on.  Plus, when we reset our cache, the next ~5 builds are excruciatingly slow.

Transparent caching would be a huge boon to our company and a differentiating feature for Shippable.

On a much smaller scale, it would be great to have a button to invalidate the cache.  Presently, we notice the cache needs to be invalidated on Shippable, and then we switch over to Git, push a commit to a bogus branch with "[reset_minion]" in the title, and then delete the branch.  A button would be much more straightforward ;-),
1393,accessibility problems prevent me from getting to settings tab,zersiax, , ,OPEN,I am a screenreader user and currently unable to reach the settings -> Deployment keys portion of the settings area. The only settings I appear to be able to access are the general account settings. I need the deployment key for Heroku access, I cannot do work without it. Can this please be fixed, soon? Or at least give me a URL I can go to to reach this page directly? Thanks,
1380,encrypted env vars doesn't work with forked repos,hackerhasid, , ,OPEN,our development flow works like this:

1 - developers fork our repo
2 - developers work on their fork
3 - developers create pull requests when they're ready to merge back into the main repo

right now every user needs to encrypt variables themselves and this means that everyone needs their own `shippable.yml` file with these (differently) encrypted values! and none of the PRs can build because you guys won't decrypt env variables from forked repos for security reasons!

the solution codeship, heroku, and aws opsworks use for secure variables is storing them on the website (instead of in the shippable.yml). so as the admin for my company i'd set these variables inside shippable.com's ui.,
1371,Shorter custom timeout,winglian, , ,OPEN,There have been a lot of random timeouts lately. All successful builds take less than 15minutes, and if it runs longer than that, it times out. Is there a setting we can use in shippable.yml that we can use to simply set the timeout to 30 minutes instead of 2 hours so we aren't waiting for these random timeouts?
,
1370,MongoDB fails to start,nathanpeck, , ,OPEN,Trying to start MongoDB with the following shippable.yml:

```
build_image: shippableimages/ubuntu1404_nodejs

language: node_js

# language version
node_js:
  - 0.12.00

before_script:
  - sudo apt-get install -y redis-server mongodb
  - sudo /etc/init.d/redis-server start
  - sudo mkdir -p /data/db
  - sudo /etc/init.d/mongodb start

script:
 - npm install
 - npm test
```

After a series of failures I changed the run command to this to see what was happening:

```
 - sudo /usr/bin/mongod -f /etc/mongodb.conf
```

This allowed me to see that Mongo is failing with this stack trace.

```
Mon Apr 27 22:24:00.920 terminate() called, printing stack (if implemented for platform):
0x981b16 0x582265 0x7fa38ac70836 0x7fa38ac70863 0x7fa38ac70aa2 0x7fa38acc2677 0x7fa38ac8e0c4 0x7fa38ac832e9 0x7fa38ac8430d 0x7fa38b6012c2 0x7fa38b602f4e 0x7fa38b5faeed 0x74a1bb 0x56ac9a 0x7fa38a352ec5 0x58210c 
 /usr/bin/mongod(_ZN5mongo15printStackTraceERSo+0x26) [0x981b16]
 /usr/bin/mongod(_ZN5mongo11myterminateEv+0x45) [0x582265]
 /usr/lib/x86_64-linux-gnu/libstdc++.so.6(+0x5e836) [0x7fa38ac70836]
 /usr/lib/x86_64-linux-gnu/libstdc++.so.6(+0x5e863) [0x7fa38ac70863]
 /usr/lib/x86_64-linux-gnu/libstdc++.so.6(+0x5eaa2) [0x7fa38ac70aa2]
 /usr/lib/x86_64-linux-gnu/libstdc++.so.6(_ZSt21__throw_runtime_errorPKc+0x67) [0x7fa38acc2677]
 /usr/lib/x86_64-linux-gnu/libstdc++.so.6(+0x7c0c4) [0x7fa38ac8e0c4]
 /usr/lib/x86_64-linux-gnu/libstdc++.so.6(_ZNSt6locale5_ImplC1EPKcm+0x49) [0x7fa38ac832e9]
 /usr/lib/x86_64-linux-gnu/libstdc++.so.6(_ZNSt6localeC1EPKc+0x60d) [0x7fa38ac8430d]
 /usr/lib/x86_64-linux-gnu/libboost_filesystem.so.1.54.0(_ZN5boost10filesystem4path7codecvtEv+0x52) [0x7fa38b6012c2]
 /usr/lib/x86_64-linux-gnu/libboost_filesystem.so.1.54.0(_ZNK5boost10filesystem4path14root_directoryEv+0xde) [0x7fa38b602f4e]
 /usr/lib/x86_64-linux-gnu/libboost_filesystem.so.1.54.0(_ZN5boost10filesystem8absoluteERKNS0_4pathES3_+0x2d) [0x7fa38b5faeed]
 /usr/bin/mongod(_ZN5mongo27initializeServerGlobalStateEb+0x2fb) [0x74a1bb]
 /usr/bin/mongod(main+0x26a) [0x56ac9a]
 /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf5) [0x7fa38a352ec5]
 /usr/bin/mongod() [0x58210c]
Mon Apr 27 22:24:00.921 Got signal: 6 (Aborted).

Mon Apr 27 22:24:00.923 Backtrace:
0x981b16 0x58272b 0x7fa38a367d40 0x7fa38a367cc9 0x7fa38a36b0d8 0x58226a 0x7fa38ac70836 0x7fa38ac70863 0x7fa38ac70aa2 0x7fa38acc2677 0x7fa38ac8e0c4 0x7fa38ac832e9 0x7fa38ac8430d 0x7fa38b6012c2 0x7fa38b602f4e 0x7fa38b5faeed 0x74a1bb 0x56ac9a 0x7fa38a352ec5 0x58210c 
 /usr/bin/mongod(_ZN5mongo15printStackTraceERSo+0x26) [0x981b16]
 /usr/bin/mongod(_ZN5mongo10abruptQuitEi+0xfb) [0x58272b]
 /lib/x86_64-linux-gnu/libc.so.6(+0x36d40) [0x7fa38a367d40]
 /lib/x86_64-linux-gnu/libc.so.6(gsignal+0x39) [0x7fa38a367cc9]
 /lib/x86_64-linux-gnu/libc.so.6(abort+0x148) [0x7fa38a36b0d8]
 /usr/bin/mongod(_ZN5mongo11myterminateEv+0x4a) [0x58226a]
 /usr/lib/x86_64-linux-gnu/libstdc++.so.6(+0x5e836) [0x7fa38ac70836]
 /usr/lib/x86_64-linux-gnu/libstdc++.so.6(+0x5e863) [0x7fa38ac70863]
 /usr/lib/x86_64-linux-gnu/libstdc++.so.6(+0x5eaa2) [0x7fa38ac70aa2]
 /usr/lib/x86_64-linux-gnu/libstdc++.so.6(_ZSt21__throw_runtime_errorPKc+0x67) [0x7fa38acc2677]
 /usr/lib/x86_64-linux-gnu/libstdc++.so.6(+0x7c0c4) [0x7fa38ac8e0c4]
 /usr/lib/x86_64-linux-gnu/libstdc++.so.6(_ZNSt6locale5_ImplC1EPKcm+0x49) [0x7fa38ac832e9]
 /usr/lib/x86_64-linux-gnu/libstdc++.so.6(_ZNSt6localeC1EPKc+0x60d) [0x7fa38ac8430d]
 /usr/lib/x86_64-linux-gnu/libboost_filesystem.so.1.54.0(_ZN5boost10filesystem4path7codecvtEv+0x52) [0x7fa38b6012c2]
 /usr/lib/x86_64-linux-gnu/libboost_filesystem.so.1.54.0(_ZNK5boost10filesystem4path14root_directoryEv+0xde) [0x7fa38b602f4e]
 /usr/lib/x86_64-linux-gnu/libboost_filesystem.so.1.54.0(_ZN5boost10filesystem8absoluteERKNS0_4pathES3_+0x2d) [0x7fa38b5faeed]
 /usr/bin/mongod(_ZN5mongo27initializeServerGlobalStateEb+0x2fb) [0x74a1bb]
 /usr/bin/mongod(main+0x26a) [0x56ac9a]
 /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xf5) [0x7fa38a352ec5]
 /usr/bin/mongod() [0x58210c]
```,
1364,Explanation of build times?,nathanpeck, , ,OPEN,Shippable always shows times that look phenomenal when you view the build detail page:

![img](http://i.imgur.com/SjBWFWI.png)

It looks like that build took just 12 seconds.... But that is completely untrue, as the overall time was over 3 minutes according to the build list for the project. Additionally if I had started a stop watch the moment the buildimage step began and stopped it the moment it finished it would have been about 1 minute in length. The same goes for the other steps.

Can you explain how the timing works, and why the build details thinks it only took 12 seconds, while the build list clearly shows times of several minutes in length?,
1362,Git tag fails to start build,jchamberlain, , ,OPEN,On Travis, I could push a tag up and a build would start, just like with any other commit. On Shippable, the build tries to start but complains that there's no shippable.yml. Builds work fine for everything else, just not pushed tags. I'm only deploying when a tag is pushed, so right now I have to manually start a build whenever I push a tag.

In some ways, this is the opposite request of #905. I'd like to have the same behavior as Travis.,
1354,Build stalls after phpunit, and then stops with timeout,antonienko, , ,OPEN,I'm trying out shippable, and I want to see how everything works before upgrading the account to a paying one, but I'm getting stuck with one problem:

the build seem to freeze after the phpunit is completed, and finally after so many minutes it gives a timeout.

You can see the build here:
https://app.shippable.com/builds/553a228bbf6d0a0c00523d18

What am I doing wrong?

Thanks in advance.,
1350, No such file or directory: '/home/shippable/builds/XYZ',caarlos0, , ,OPEN,I'm randomly getting this error while building some java projects.

```
[Errno 2] No such file or directory: '/home/shippable/builds/54d10b015ab6cc13528aaf77': Traceback (most recent call last):
  File "/home/shippable/dpd/step_listener.py", line 99, in handle_message_callback
    self.execute_step(build_plan, current_step, False)
  File "/home/shippable/dpd/step_listener.py", line 165, in execute_step
    raise e
OSError: [Errno 2] No such file or directory: '/home/shippable/builds/54d10b015ab6cc13528aaf77'
```

How can I proceed?,
1348,Issue with the "cache: true" option,hildingur, , ,OPEN,I have two private repo's in java where one depends on the other.

The upstream repo, builds find, but the build artifacts aren't cached for the downstream repo.

Can you please investigate.

Thanks.,
1341,Artifacts download URL with Free plan for a private repo,cchantep, , ,OPEN,Hi,

Is it possible to get (if yes how) the download URL for the artifacts (the archive) created by a build, using a free Shippable plan with a private Bitbucket repository?

Best regards,
1338,can't reset settings override of shippable.yml,DanLipsitt, , ,OPEN,The settings in the web interface say they override the contents of shippable.yml. The problem is that there is no way to unset that override. This is compounded by the fact that you can't change some settings without changing others. This means you can't set enable-privileged-mode without also setting build-image, which I want to leave at the default value in shippable.yml.,
1336,Latest Python 2.7.x support,joaonvfigueiredo, , ,OPEN,Hi,

[This warning](https://urllib3.readthedocs.org/en/latest/security.html#insecureplatformwarning) yields due to using an older version of Python 2.7 SSL implementation. I'm using Python 2.7.8 which works without this warning.

The module that issues the warning (urllib3) is quite used for python web programming, and used by other modules such as requests. I'd say some more people are getting the warning.

Currently I see that the min/v2 container uses the Python 2.7.3 version, however the shippable.yml only seems to be able to select the major versions.

Is this possible:
@shippable.yml

    python:
      - 2.7.8
?

If not, am I forced to use a different build image that uses an updated version of Python?,
1325,No sufficient output ,fisera123, , ,OPEN,Hello, 
I'm trying to integrate shippable with my project, in the repository project is not the root folder, I couldn't find any explanation how to change directories on shippable docs webpage.
This is my shippable.yml 
```
language: ruby

rvm:
  - 1.9.3

env:
  - CI_REPORTS=shippable/testresults COVERAGE_REPORTS=shippable/codecoverage

before_script:
  - mkdir -p shippable/testresults

install:
  - cd ./taxiapp/
  - bundle install --gemfile="Gemfile"
script:
  - rspec -f RspecJunitFormatter -o shippable/testresults/results.xml

notifications:
 email:
     recipients:
         - example@gmail.com
```
Result is failure but no information is provided, thus I don't know what I did wrong. I've also tried adding `- cd ./taxiapp/` to the `script` in that case output was directory not found
https://app.shippable.com/builds/5531c6143ef75c0c00a39a86,
1321,Update PHP composer version,claar, , ,OPEN,The composer version in shippable is from November 2014 (1.0.0-alpha10, hash ffffab3).

This old version doesn't understand current composer syntaxes, such as ^5.2, so I've had to add `composer self-update` to my build script, adding an unnecessary delay in build times and processing work/bandwidth for the build host.,
1315,In triggerBuild request: Passing a parameter from one build to another,vkuusk, , ,OPEN,We are doing chaining of several builds and need to pass "a string value" from one build to the next.
Currently we are using external location (private S3 bucket) to store this value (one build puts there, the next one reads from there).

the data we are transferring is just a small string.

Is it currently possible to pass a key:value pair to the triggerBuild request, so it becomes an environmental variable in the "next" build.

If it's not currently possible, will you consider adding this feature at some point in future?

Thanks,
1313,Failing on Install part of build,sojharo, , ,OPEN,Shippable sometimes fails to pull the libraries during the build process. After this, I go to the shippable website and run the build manually from there and then it gives no error.

![shippable](https://cloud.githubusercontent.com/assets/5811465/7176266/44244592-e436-11e4-8f42-1bb5dc7afb78.PNG)
,
1304,curl: (22) The requested URL returned error: 404,kaspernj, , ,OPEN,Please see 35.4 on this build:
https://app.shippable.com/builds/552b9ba96206490c00972945

I am getting those errors randomly very often.

"shippable.yml" and all other files are available from the public repository.,
1293,How to add ssh keys,z2z, , ,OPEN,How to add private ssh keys...  objective is to set up private bitbucket repos inside composer?,
1290,coverage: highlighted source code,jamestalmage, , ,OPEN,In Jenkins, the cobertura coverage output includes highlighted version of the source-code that allows you to easily see which lines are not covered. Using basically the same build process in Shippable (i.e. same result.xml file) - I am seeing basic coverage information information for each file, but no highlighted sources.

Is this a feature that Shippable does not support, or am I doing something wrong?,
1283,Include branch name in subject line of notification email,fzumstein, , ,OPEN,It would be great if that could be configured.,
1277,does shippable run builds for each git tag push on bit-bucket?,geshan, , ,OPEN,I think it does not run the build/tests for git tags pushed on bit-bucket can this be checked. In my opinion it should run the build on each git tag push only only branch push. thanks.,
1268,Primary/largest link in emails should be to the build results,catskul, , ,OPEN,I've been using Shippable for several months now and have found the email notifications a bit frustrating in their UX.

Virtually any time I want to go somewhere from the email it's going to be the details of a test failure.

However the test failure "details" link is the smallest most non-obvious link on the page, and when I'm in automatic mode, I always end up clicking one of the other links on the page, and end up annoyed.

IMO, the first and most obvious link on the page should be to the test results:
 * Not to shippable.com, no one who already uses the service would want to go there from this email
 * Not to the repo url, not sure what I would do from there
 * The commit hash is potentially useful but certainly not in comparison to why the build failed.

If you agree, please re-think the UX of your email alerts.

The easiest fix IMO would be link to move the results to the top, and the context information to the bottom, and to link the green/red "succeeded"/"failed" marker to the details. And un-link the Shippable icon. 
 ,
1267,Breadcrumbs linking is inconsistent,catskul, , ,OPEN,The bread crumb links in the top left do not follow a simple consistent scheme.

While looking at a project the text is:

    Team / project

If I click on project, I get taken to bitbucket.
If I click on team, I get taken to shippable team page

Once on the shippable team page the text is:
 
     Team

If I click on the team, I end up on the bitbucket page.

I regularly click on links on Shippable, and in Shippable emails and end up on bitbucket without meaning to. It's extremely frustrating, and despite having used Shippable for months now I still find myself ending up on pages I don't mean to be.

IMO virtually all links should link back to places within the shippable app, unless there's a compelling reason not to (links on commit hashes). I see no utility in arbitrarily linking back to revision control site. 

I'll open a separate ticket for the email UX.
,
1249,MongoDB version,malchmih, , ,OPEN,Hello!

Base docker image has outdated version of Mongo - 2.4.12. Could you update it to 2.6+? My project depends on the features of this new version, so I can't really use Shippable now.

Thank you!,
1248,Shippable for java dependent projects,reddy6ue, , ,OPEN,I am trying to solve the following problem. 

I have three projects, a persistence layer library, a service layer library and a web application.

Shippable currently doesn't have a feature to kickoff cascading builds like Jenkins does. So, I decided to use git submodules to setup appropriate dependencies. Does anybody have a sample that makes git submodules work with a bitbucket git repo? 

Or alternately, has anybody solved the cascading dependent jobs problem like the one Jenkins provides?,
1240,[cosmetic] log output doesn't play well with control sequences,fommil, , ,OPEN,The log output is quite horrible when running any scala builds, e.g. this is typical output:

```
[info] Loading project definition from /root/workspace/src/github.com/fommil/freeslick/project
[info] Updating {file:/root/workspace/src/github.com/fommil/freeslick/project/}freeslick-build...
[info] Resolving com.typesafe.sbt#sbt-scalariform;1.3.0 ...
A
2K[info] Resolving org.scalariform#scalariform_2.10;0.1.4 ...
A
2K[info] Resolving org.scala-lang#scala-library;2.10.4 ...
A
2K[info] Resolving org.scoverage#sbt-scoverage;1.0.4 ...
A
2K[info] Resolving org.scoverage#scalac-scoverage-plugin_2.10;1.0.4 ...
A
2K[info] Resolving org.scoverage#sbt-coveralls;1.0.0.BETA1 ...
A
2K[info] Resolving com.fasterxml.jackson.core#jackson-core;2.4.2 ...
A
2K[info] Resolving com.fasterxml.jackson.module#jackson-module-scala_2.10;2.4.2 ...
```

with the

```

A
2K
```

being artefacts that should not be shown. Also, there is a lot of double newline printing happening when there should only be one newline.,
1239,Instructions to optimise a docker image for use in shippable,fommil, , ,OPEN,In our docker images, we try to guess what the Shippable minion environment wants to install to speed up the build: https://github.com/ensime/ensime-docker/blob/shippable-1.0/Dockerfile#L89

```
RUN\
  apt-get install -y locales git-core python-pip ssh &&\
  apt-get clean &&\
  pip install virtualenv &&\
  pip install --install-option="--prefix=/root/ve" pika boto glob2 &&\
  echo "en_US.UTF-8 UTF-8" >> /etc/locale.gen &&\
  locale-gen
```

but there is still a significant amount of installation happening when the minion starts up that I can't pre-compute.

What else can we be doing to help startup speed? (and, especially with github being under attack, reduce network failure rate), e.g. today all our builds are failing because of 

```
Downloading/unpacking virtualenv==1.11.4

  Cannot fetch index base URL https://pypi.python.org/simple/

  Could not find any downloads that satisfy the requirement virtualenv==1.11.4

Cleaning up...

No distributions at all found for virtualenv==1.11.4

Storing debug log for failure in /root/.pip/pip.log

converted 'https://raw.github.com/pypa/pip/master/contrib/get-pip.py' (ANSI_X3.4-1968) -> 'https://raw.github.com/pypa/pip/master/contrib/get-pip.py' (UTF-8)

--2015-03-29 11:23:54--  https://raw.github.com/pypa/pip/master/contrib/get-pip.py

Warning: Boot step is taking longer than expected
```

which I presume is because `raw.github.com` is not reachable... but we already have `pip` installed so I don't think this stage should even be necessary.,
1235,enable selected branches,fommil, , ,OPEN,I have some branches on my repos that I don't want shippable to build or complain about if there is missing `shippable.yml`.,
1233,don't require sudo,fommil, , ,OPEN,Longer term, you should be looking at dedicated host installations that **do not** require `sudo` permissions.

Given that you only:

* install python packages into your home directory
* launch docker

there is no reason for you to `sudo`, if I have given you access to run docker. I realise this means that you would not be able to run docker in privileged mode, but I'm very happy with that.
,
1214,intermittent build failures due to submodule,smaiyaki, , ,OPEN,my configuration is as thus:

* In the repository, I have added a submodule. I have however not staged the submodule to avoid my repo from being unnecessary big
* In shippable, in the before_script tag, I added these 2 steps:
  * git submodule init
  * git sumodule update
* At this stage, the submodule should be checkout during the build
* I will then cd in to the submodule and run a script that starts another build in the background

While this step works in most of the cases, it occasionally fails to run my script with the error that the script file could not be found. I have added some extra line to list out all the file listing before running the script file and it shows that the submodule has not been cloned and thus the failure. a rebuild or a couple of rebuilds will fix the issue

In summary, the submodule feature is a buggy I guess,
1208,Rebuild all open pull requests on push to destination branch,sigvef, , ,OPEN,Hi,

Is it possible to rebuild all open pull requests on push to the destination branch? By this I mean that when a  branch (say, `master`) is updated, all open pull requests into that branch are rebuilt.

Thanks!,
1200,limit concurrent jobs on a dedicated node,fommil, , ,OPEN,The number of jobs running on a dedicated host appears to be unbounded, this can take the box down. It would be good to limit this on a per-host basis.,
1199,cancelling a job does not stop it on the node,fommil, , ,OPEN,On my dedicated nodes, if I start up a bunch of jobs and then cancel them, I can still see them running to completion on the nodes, eating up CPU cycles and precious resources.,
1198,support coveralls.io in pull requests,fommil, , ,OPEN,to get coveralls.io support in pull requests, I need to put my `COVERALLS_REPO_TOKEN` in the `shippable.yml` file. I think it is correct that the encrypted vars should not be visible in pull requests, but it would also be good if some transient thing were made available to work with coveralls.io. Travis is able to do this by using their `TRAVIS_JOB_ID` but I think they have buy-in from `coveralls.io` to do this.,
1197,TypeError: on_connection_closed() takes exactly 2 arguments (4 given),fommil, , ,OPEN,While debugging another issue, I found a lot of these errors in the shippable logs:

```
TypeError: on_connection_closed() takes exactly 2 arguments (4 given)
WARNING:pika.connection:Disconnected from RabbitMQ at msg.shippable.com:5671 (0): Not specified
```

You might want to look at http://stackoverflow.com/questions/19255014 to help with this.,
1196,connection error on dedicated node,fommil, , ,OPEN,I have two dedicated nodes: both successfully deployed, but one is working and the other is not. Both are hosted by the same provider, but are running different flavours of Linux: the succeeding one is running Ubuntu, the failing one is running Debian.

This is an example of a failure https://app.shippable.com/builds/550d5d8f21d6a90b00366cf4 which is using the normal shippable docker image. It also fails for custom images.

```
 Docker login
Warning: Docker login failed for user fommil with error HTTPConnectionPool(host='127.0.0.1', port=4243): Max retries exceeded with url: /v1.15/auth (Caused by <class 'socket.error'>: [Errno 111] Connection refused)
 Docker pull shippable/minv2:latest
HTTPConnectionPool(host='127.0.0.1', port=4243): Max retries exceeded with url: /v1.15/images/create?tag=latest&fromImage=shippable%2Fminv2 (Caused by <class 'socket.error'>: [Errno 111] Connection refused): Traceback (most recent call last):
  File "/home/shippable/dpd/step_listener.py", line 99, in handle_message_callback
    self.execute_step(build_plan, current_step, False)
  File "/home/shippable/dpd/step_listener.py", line 165, in execute_step
    raise e
ConnectionError: HTTPConnectionPool(host='127.0.0.1', port=4243): Max retries exceeded with url: /v1.15/images/create?tag=latest&fromImage=shippable%2Fminv2 (Caused by <class 'socket.error'>: [Errno 111] Connection refused)
```

If I manually log in to the box as the `shippable` user, I am able to issue this command

```
docker pull shippable/minv2 
```

which works (although, notably, it had to download everything first).

It would help me greatly if you could let me see the script that is running and failing... there is probably some missing unix utility or a python/shell version mismatch.,
1191,Bump maven version,velo, , ,OPEN,Currently shippable is using  maven 3.2.1.

This version has some bugs to run parallel builds [MNG-5750](http://jira.codehaus.org/browse/MNG-5750) and to download artifacts [MNG-5723](http://jira.codehaus.org/browse/MNG-5723)

I will build a custom image for my project, but would be way better keep using shippable default image!

Thanks!,
1181,How to copy build output to cloud storage,affmobapp, , ,OPEN,How to copy build output to cloud storage like DropBox or google drive,
1180,Not able to update Bitbucket commit with tag,affmobapp, , ,OPEN,After successful build not able to update Bitbucket commit with tag.
Not able to use git push. Not able to add SSH.,
1179,Setting user for custom image,thephred, , ,OPEN,I am trying to run a custom docker build image via a dedicated host.  It has a user setup in it for the build and I would like shippable to launch the build with that user similar to how minv2 runs as shippable.  However, it only runs as root.  There was a reference in another ticker to the fact that you could set the user in the shippable.yml but I couldn't find how to do this in the docs and guessing how to do it has failed.  Any clue on this?,
1173,RFE: exit build with message to appear on github, etc,fommil, , ,OPEN,It would be really great if scripts could be written with some extra information that would appear in PRs if they fail, e.g. instead of the current

```bash
  - sbt ++$SCALA_VERSION clean test:compile it:compile doc &&
    sbt ++$SCALA_VERSION test it:test &&
    sbt ++$SCALA_VERSION publishLocal &&
    ensime-emacs/test/run_emacs_tests.sh &&
```

imagine if we could write

```bash
  - sbt ++$SCALA_VERSION clean test:compile it:compile doc || failed "compilation"
  - sbt ++$SCALA_VERSION test it:test || failed "tests"
  - sbt ++$SCALA_VERSION publishLocal &&
    ensime-emacs/test/run_emacs_tests.sh || failed "integration tests"
```

it would be even better if the PR comment included a link to the relevant part of the log.

Under Travis rules, if a line fails the others will still be evaluated. I'd like a way of just exiting immediately with the message. Anything that speeds up the cycle is good.,
1172,failure in after_success should not influence the results,fommil, , ,OPEN,related somewhat to #1165 I just had a build fail because of a failure in `after_success` https://app.shippable.com/builds/5504edc19b228b0d005bdfbe

This shouldn't happen: the build has succeeded at this point and failures in `after_success` should be ignored (perhaps a notification?),
1169,jdk flag is ignored,fommil, , ,OPEN,I changed our builds to use the Ubuntu 14.04 (instead of 12.02) docker images, but now it's clear that the config entry

```
jdk:
  - openjdk6
```

is being ignored because our tests always fail on these two unit tests if the java sources are not installed, and I only installed `openjdk-6-source` https://app.shippable.com/builds/5504c2e29b228b0d005bde28

I can only presume that `JAVA_HOME`, `JDK_HOME` and all the various `java` binaries are not pointing at my selected java.

Workaround is to manually type something like this in `install`, which is what the `jdk` entry should be doing behind the scenes.

```
sudo update-java-alternatives -s java-1.6.0-openjdk-amd64 &&
export JAVA_HOME=$(readlink -f /usr/bin/java | sed "s:bin/java::") &&
export JDK_HOME=$(readlink -f /usr/bin/javac | sed "s:bin/javac::")
```
,
1168,deactivate dedicated node,fommil, , ,OPEN,I have a node that is not working out, and I'd like to disable it until we can fix the problem. However, there is no option to do this... so I've had to delete it and when I want to try and get it working again I'll have to add all the details, delete the other nodes (because I'd like to isolate this node) and then add everything again.,
1165,result is not published to github when `script` stage finishes,fommil, , ,OPEN,I have some big `after_success` jobs that are slow and trigger downstream behaviours (e.g. coverage reports from coveralls.io) but Shippable seems to be waiting for **everything** to finish before reporting success, especially back to github (e.g. updating a PR status or badge).

The correct behaviour should be that when the `script` stage finishes, the triggers to update github and badges should be fired.,
1160,RFE: travis-style tail log,fommil, , ,OPEN,Can you please implement the ability to follow the tail of the log, the way travis does?,
1159,ignoring travis apt_packages,fommil, , ,OPEN,With your docker support, I'm certain you have a much better way of doing this, but for the awesome "10 seconds to migrate from travis" claim, you might want to fix this.

My project uses the Travis-CI `sudo: false` with the `apt_packages` addon.

https://github.com/ensime/ensime-server/blob/master/.travis.yml

it would appear that you do not support this, otherwise this would have worked

https://app.shippable.com/builds/550458419b228b0d005bd68c,
1157,Can't merge client+server coverage report together,topikachu, , ,OPEN,I have a node project which uses mocha for server tests and karma+jasmine for client tests. Both create coverage report, with different file name, under shippable\codecoverage. It's OK if there's only a single coverage report. But if I create two reports together, shippable only displays the server one. ,
1155,save_reports fails when no code for coverage analysis ,nbucic, , ,OPEN,Right now our code is not well tested but we want to implement code coverage analysis for further bundles that we will create. When I put current Symfony bundles in phpunit.xml <whitelist><exclude> and make phpunit test localy, empty code coverage analysis is generated.
However, when I deploy to github and Shippable automatically runs build, script finally fails with error:

```
save_reports
  parse_coverage_reports
    Error parsing coverage reports.
```

This is what I get when I run phpunit locally with coverage-text:


```
Code Coverage Report:     
  2015-03-13 22:00:40     
                          
 Summary:                 
  Classes: 100.00% (14/14)
  Methods:        (0/0)   
  Lines:          (0/0)   
```

If I enable at least one of current bundles, script succeeds.,
1152,Shared local maven repository between projects,akribopo, , ,OPEN,Hi, 

I have 5 different java projects. During the build of No.1 project, mvn install is executed in order to install the artifacts of the build to the local maven repository. Cache is also enabled. 

However, when the build for project No.2 (which has a dependency in project No.1)  is executed, it fails because there aren't any artifacts of No.1 project in the local repository. 

Is there any solution to share local maven repositories between projects/minions? 
,
1142,Grunt + Compass + PNG Error,egermano, , ,OPEN,I've got some issues with grunt, compass and png (when generate sprites).

This already worked in past, but now, don't.

[That's my build](https://app.shippable.com/builds/54ff539c50bbf40d00dd6822)

That's the error I've got:

    Warning: ChunkyPNG::SignatureMismatch on line 102 of /var/lib/gems/1.8/gems/chunky_png-1.3.4/lib/chunky_png/datastream.rb: PNG signature not found, found "\211PNG\n\032\n\000" instead of "\211PNG\r\n\032\n"!,
1123,mariadb 10 with tokudb engine,sheershoff, , ,OPEN,Please add MariaDB 10 with TokuDB engine support. MariaDB 5 with TokuDB could be also a good add-on.,
1122,Error Decrypting Secure Variables ,stasyanx, , ,OPEN,Im trying to integrate with Slack for notifications and getting error `Error Decrypting Secure Variables ` while trying to run my test.
I encrypted my `slack_token` on shippble like SLACK_TOKEN=my_slack_token
my chunk of shippable.yml
`env:
  global:
    -APP_NAME=xxx
    -SLACK_ORG=11 PROJECT=22
    -CI_REPORTS = shippable/testresults COVERAGE_REPORTS=shippable/codecoverage
    -secure: my_encrypted_token here`,
1120,Do not attempt a build if the PR if base<->head have merge conflicts,pmoust, , ,OPEN,PR contains conflicts with base branch that cannot be automatically resolved.

Expected behavior: no build triggered
Actual behavior: build starts and fails on git merge.

Not that important, but nice to not have a spam to the team members if a PR fails for such reason.,
1095,Timeout when trying to push/deploy to CloudControl using cloudcontrol tools,resurge, , ,OPEN,Related build: https://app.shippable.com/builds/54ea59f7f774120e0020dc48

When trying to push or deploy to CloudControl I get a timeout.
(`cctrlapp <appname>/<branchname> push` & `cctrlapp <appname>/<branchname> deploy`)
More info about the tools can be found here: https://www.cloudcontrol.com/dev-center/Platform%20Documentation

(I could circumvent the push command by using git straightaway to push but the deplot command I can't find an easy alternative to)
In my case, when trying to execute `cctrlapp shippablecloudcontrolexample/default deploy` I get a timeout.
I think it's weird that it completely freezes.
Is it possible that the cctrlapp command is prompting for input and that that line doesn't get logged in the shippable output? ,
1093,nodejs 0.12.0 is building from source ~ 20min build time,thecaddy, , ,OPEN,Is there any way to install from the nvm binary and/or a shippable container with the latest version of stable node?,
1092,postgres 9.4,thecaddy, , ,OPEN,Would appreciate if you'd add an the updated postgres.,
1091,Debian minion image,DanLipsitt, , ,OPEN,Issue #72 requested Debian 7 (wheezy) minions, as well as Ubuntu. The issue was closed when the Ubuntu image was completed, but there is still no Wheezy image.,
1086,Is there a way to deploy containers to Azure?,munjalpatel, , ,OPEN,Hello,

Is there a way to deploy containers to Azure CoreOS machine?,
1084,Don't use a virtualenv for boot.sh on dedicated,pcsforeducation, , ,OPEN,My base container has all the requirements already installed, so tests take like 30 seconds when I also install the requirements.txt in /home/shippable/wrk/requirements.txt and throw a "deactivate" into boot.sh. Otherwise they take quite a bit longer, and I see little benefit over using Travis. It would be much simpler to ship the requirements preinstalled (there's only like 3 of them) in the directory you mount into the container. The dependencies are pinned anyway, so when you want to change them, you can reship them. 

Also, you seem to overwrite the files in /home/shippable/wrk occasionally, so I can't just edit the file and have my tests always work :),
1079,Failed build step time,velo, , ,OPEN,Would be nice to have the time the step took to fail.

![image](https://cloud.githubusercontent.com/assets/136590/6332369/974c42de-bb63-11e4-8c6e-ed398321bdb6.png)
,
1078,Dedicated hosts server,velo, , ,OPEN,Would be nice to have a visual clue showing witch dedicated hosts is being used for a given build.

My build stuck for some reason and I had to connect to all servers and reboot to get things back to normal,
1066,Minify & Combine JS on shippable app,naveensky, , ,OPEN,Hi, 

I am using shippable in India for last 2-3 weeks. I find shippable app to load really slow (15+ secs) on my 16Mbps connection. I guess thats because shippable prod app is not combining js/css, 

- total of 115 unminified scripts, amount to 2.2 MB which can reduce by 75% and we could definitely use CDN for angular, socket and other third party libs
- total of 12 css which could be combine
- We could definitely use CDN to host js/css/images

I am sure shippable team would be knowing about these issues but if they could prioritise in their grand scheme of things (no pun intended, u guys are doing awesome stuff!) it would make my life easier

I am more than willing to help if required, 

Thanks, 
,
1065,Limit builds when using dedicated hosts,velo, , ,OPEN,My build is really greedy.  And I wanna limit how many builds I run per machine.

If I don't linux will start killing processes.

How can I do so?,
1041,Different tests for Server side & Client side,podoler, , ,OPEN,Hi,
We have two different tests & coverage,
one for server side and the other for client side,
Is there any way to see both different reports for tests & coverage for both of them?

Thank you,
Yarin Podoler
StartApp Team,
1036,I need to add additional repositories for maven into settings.xml,venkateshamurthy, , ,OPEN,Is it possible add custom repository urls some where (basically settings.xml customized) during maven build?

Today i am failing in https://app.shippable.com/builds/54d4db5421e8ee0e00db520a.

Please help,
1033,Maven central proxy...,velo, , ,OPEN,Hi guys,

You probably spend a good ammount of money on internet to contact repo1.maven.org/maven2 and repo.maven.apache.org/maven2/

You probably can save a lot by setting up you own "central" using nexus or artifactory.... 

For me, as user, it would save tons of time downloading maven artifacts.,
1016,Shippable IP ranges,pmoust, , ,OPEN,Hello,
would it be possible to provide us (the ones that use shippable in dedicated hosts) some IP ranges that we can whitelist in our firewall rules for shippable to access port 22?,
1003,Support for multiline YAML statements,zanona, , ,OPEN,One cool thing about YAML is the ability to add multiline strings as parameter values.
Unfortunately, it seems Shippable currently does not support it.
If you guys could implement this, it would great for improving config file management and readability as well.

For example:
```yaml
after_script: …
after_success: |
  if [ "$BRANCH" == "master" ]
      then rsync -azP --exclude=node_modules --del "$OUTPUT_DIR/" "$SSH:api"
      ssh "$SSH" "forever restartall"
  fi  
```
,
983,Configuration to run builds in queue for a project,phagunbaya, , ,OPEN,Right now if any changes are pushed to git repo, built is triggered on shippable and runs in parallel. I want to configure it in such a way that for a project it should run a built at a time and rest triggered builds should wait in queue. Any way to do this ?,
953,Multiple dedicated hosts with common IP: Punching through firewall/NAT,taqtiqa-mark, , ,OPEN,Hi, I'd like to know if you have tested, or are aware of users successfully using ngrok to connect to multiple dedicated hosts sharing an IP address of the host.

Related issues are;
#839 @ragsden Establishes that Shippable : 
- needs general outbound access
- needs SSH access to `<subdomain>.ngrok.com:<number>`

#837 Establishes that Shippable can (since 6-Dec-2014) utilize multiple dedicated hosts with same IP and different ports.

The question arises when combining #839 and #837.  That is, multiple dedicated hosts on the same IP adderess behind firewall/NAT:
- Does Shippable require SSH access via a port that is dedicated to each dedicated host...?

That is, in the example in #837, I would set up ngrok to tunnel TCP traffic as follows:
````
mysubdom.ngrok.com:10122 -> 1.1.1.1 port 10122
mysubdom.ngrok.com:10222 -> 1.1.1.1 port 10222
````
,
942,UI/UX enhancement: update head title with something meaningful,pmoust, , ,OPEN,Operating in multiple tabs tends to be annoying when the head>title does not contain useful info.

In this example multiple tabs with different builds are open, yet it is not possible to distinguish which is what.

![screenshot 2015-01-05 15 59 29](https://cloud.githubusercontent.com/assets/2493339/5613934/fb249f2c-94f3-11e4-9183-874b3e78115e.png)

Would be very nice to have the information available at a glance (i.e. Repo name / # ID)
Extra candy would be to update the DOM with the build status.,
926,Any way to get timestamp with built console output,phagunbaya, , ,OPEN,Is there any way to get the timestamp with console output during built time ? The problem is sometimes built take a longer than expected (even longer than an hour. Our tests output have a time stamp so they are not the culprit. If we get the console output with timestamp, we can find the abnormal condition.,
924,Add support for a shell scripts,bassamtabbara, , ,OPEN,Please add support for shell scripts builds. I.e. in shippable.yml language is set to "bash" or "shell". Also would be great for there to be a generic base image with no services and language runtimes installed. I tried using shippableimages/ubuntu_base but that was not supported. Finally, having an icon for the generic image in your web UI would be great, as it stands the default is the ruby gem icon.
,
905,Git tags for deploys start new test,rogierslag, , ,OPEN,We'd like to keep track of which code was running in production at which point in time. However, if our deploy script pushes a new tag to github this triggers a new build on shippable. Can we switch this off somehow? It kinda pollutes the overview and causes regular builds to hold

```
git tag "deploy_SOMEVAL"
git push --tags
```,
901,Notifications should support sending to committer and predefined list of e-mails,HenrikOssipoff, , ,OPEN,Extending upon #900, I would like to add a feature request to be able to add a new way of doing notifications.

In our use case, we don't want to spam the entire development team whenever our build breaks. We have HipChat for this. However, we would like to do the following, which is not currently possible in Shippable:

- Send an e-mail to the last committer so they definitely know they broke something
- Send an e-mail to a predefined list of e-mails -- we have a shared e-mail folder which we use to keep track of such things,
887,npm install with package.json reference to private github repo,chriskinsman, , ,OPEN,How do I handle this?  I have shared code referenced in my package.json in a private repo.  npm install fails with: permission denied: (public key).,
857,Continuous Deployment to a Chef Server for cookbooks,behemphi, , ,OPEN,It would be ideal to have a way to upload a cookbook to a Chef server after it passes test.  

* light weight - do not install the entire chef gem or omnibus package, only what is needed to the the upload
* connect to any Chef server (i.e. Host Chef, Private Chef, Enterprise Chef)
,
834,Using environment variables in commit_container,jamesjwarren, , ,OPEN,Is it possible to do something like:
`commit_container: dockerusername/$project:$tag`

setting `$project` and `$tag` via environment variables? I have tried this but get invalid repository name as it thinks the repo name is $project...

Ideally I would like to set the tag based on the branch.,
824,[Consolidated feature request] Push to docker registry feature requests,moylop260, , ,OPEN,I have dedicated host configured done.
All works fine.

Now, I need connect to my minion build container to debug my test.
With other tools ci we can enable "ssh connection" for this purpose.

![example_ci](https://cloud.githubusercontent.com/assets/6644187/5220263/2c3caedc-7628-11e4-974b-13f042f9045c.jpg)

However, shippable use docker container.
Can we use this docker container built by shippable?
,
820,Only include services based on current build matrix,hanloong, , ,OPEN,Is it possible to only include services for specific builds in the matrix?

For example we have unit tests and integration tests.
 - They both require `postgres` and `redis` - this works now
 - Only the INTEGRATION tests require `selenium` which can take a long time to start

Also I noticed `mysql` is started by default, is there a way to stop this?

Thanks in advance,
803,[question] Using more than one custom images,moylop260, , ,OPEN,I have defined into ```.travis.yml``` file
```yml
build_image: my_custom_image
```

But We need add other images to test it.
```yml
build_image: 
   - my_custom_image1
   - my_custom_image2
   - shippable/minv2
```
Is possible?,
794,Re-run individual builds environments ,hanloong, , ,OPEN,Is it possible to run individual builds (So just build 19.2 not 19.X which would include (19.1, 19.2, 19.3 etc)).

We have setup environments to run builds separately.
 - unit tests
 - integration tests
 - external integration tests (rely on external API's)

Unit tests are super reliable, integration tests less so, and external integration even less than that. On Travis if one of these fails we can just trigger a re-run of that test and leave the others as passed.

I know most people tend to mock out external API calls during tests but we have been burnt by our payment gateway API making breaking changes to existing API versions, so this is the best way to make sure our code is stable.,
790,Piped command breaks execution -- or just the log,bittner, , ,OPEN,I'm using a series of piped commands to validate an HTML file as part of the build process. The [YAML file](https://bitbucket.org/bittner/adventsbasar/src/master/shippable.yml#cl-12) validates (with [Travis WebLint](http://lint.travis-ci.org/)) and the build process seems to execute successfully. However, the [build log](https://app.shippable.com/builds/546c825a7aac73c600837567) looks a bit odd:
```
(-)  script                                                                ? 1 seconds
      bash -c '! python html5check.py home/*.html
    (+) csslint --format=compact home/*.css
    Build script completed 
```
## Command is cut off at first pipe

The first script command really is:
```
bash -c '! python html5check.py home/*.html | grep -v "Error:.*frameborder.*attribute.*iframe" | grep "Error:\ "'
```
In the console log only the part before the first pipe shows up. There is no output of the command shown, but that *may* be correct, because with no errors there is no output. Also when I download the console log the text output is not much better: The part starting at the first pipe character is missing completely.

Is this just a text output error or is part of the command *actually* swallowed?

## Side Note

I need to do the weird `bash -c '...'` construct, because the YAML won't validate with a simple negated Bash command line. Explained with examples in https://github.com/mozilla/html5-lint/issues/9.,
774,Rspec & Cucumber output with colors,ArturT, , ,OPEN,It would be great to have colorized output for script command. It's much easier to read rspec or cucumber output when I see what's green or red. Feature badly needed. :),
771,SHIPPABLE_MINION_INDEX and SHIPPABLE_MINION_TOTAL,ArturT, , ,OPEN,Could you provide variables:
* `SHIPPABLE_MINION_INDEX` it should starts from zero.
* `SHIPPABLE_MINION_TOTAL` it says how many minions particular build has.

I created knapsack gem https://github.com/ArturT/knapsack and I'd like to integrate it with shippable. It would be nice to have mentioned envs. CircleCI has similar thing https://circleci.com/docs/parallel-manual-setup#env-splitting,
763,Support for AppFog deployments,shawn9273, , ,OPEN,Is there current support (or plans to support) AppFog deployments?,
707,[Consolidated feature request] Support additional Notifications providers,amobrem, , ,OPEN,https://slack.com/,
693,regex & wildcards in branch exclusion,jeradphelps, , ,OPEN,I'd like to be able to run builds on wildcards for git-flow branches.  For instance, something like this:

~~~
branches:
  only:
    - master
    - develop
    - feature/*
    - release/*
    - support/*
    - hotfix/*
~~~

Is this supported?  It is not documented (though it is for [travis](http://docs.travis-ci.com/user/build-configuration/#Using-regular-expressions)), and in my initial tests, it is not working. ,
679,Don't see build status for each commit on Github,rockerhieu, , ,OPEN,Recently I don't see any icon to indicate build status for each commit on Github from Shippable. My project is https://app.shippable.com/projects/543f4a86b904a4b215678013. I tried to delete and re-enable the project on Shippable, but it didn't fix the issue.,
656,Make the project, build URL human understandable,rockerhieu, , ,OPEN,Let me take Circle CI as an example:
* https://circleci.com/gh/:account/:project/tree/:branch
* https://circleci.com/gh/:account/:project/:build_number
,
616,Support code analysis,rockerhieu, , ,OPEN,The idea is to parse any Lint/FindBugs/PMD code analysis results in `shippable/codeanalysis` and render them like you're doing with `shippable/testcoverage` and `shippable/testresults`.,
604,Support for http://ccmenu.org/ status monitoring,toby-griffiths, , ,OPEN,Hi there,

Just started playing with CI & found it nice & easy to setup the initial repos.  Now to configure the build.

Do you now, or plan to in the future, support feeds for monitoring build status using tools such as [CCMenu](http://ccmenu.org/)?,
601,Shippable account does not use Github primary email,arielschiavoni, , ,OPEN,Hi,

I have a Github account with two different emails configured.
Then when I created a Shippable account linked to Github it seems to be using the last email in that list which is not the primary one.

Thanks.

Ariel.
,
593,.travis.yml does not respect defined python version,raymondwanyoike, , ,OPEN,Setting a specific python version (`2.7.8` instead of `2.7`) in `.travis.yml` has no effect.

```
language: python
python:
    - "2.7.8"
...
```

All builds are being made against python `2.7.3` (build log)

```
Installing setuptools, pip...done.
Running virtualenv with interpreter /usr/bin/python2.7
Python 2.7.3
pip 1.5.4 from /home/shippable/build_ve/python/2.7/local/lib/python2.7/site-packages (python 2.7)
mysql started successfully
```,
578,Login page clobbers original destination,luhn, , ,OPEN,Observed behavior:

* Click on a link to a Shippable build, such as `https://app.shippable.com/#/builds/123456`, with no logged in session or an expired login session
* You are redirected to `http://www.shippable.com/#/builds/123456`.
* Click on "Login" in top right.
* Click "GitHub auth"  (I assume BitBucket auth behaves similarly, but I haven't verified.)
* You are logged in and redirected to `https://app.shippable.com/`.

Desired behavior:  After login, be redirected to the initial destination:, `https://app.shippable.com/#/builds/123456`.

Another suggestion:  It would make it flow a little smoother if we were redirected to a dedicated login page (containing "GitHub auth" and "BitBucket auth" in very large buttons, and little else), rather than the homepage.  It might not be evident from the homepage that a login is being requested, nor is it immediately clear where one is able to log in.,
570,Manually triggered deployments,pniederw, , ,OPEN,Couldn't find a discussion forum so posting here. I'm building an OSS library that needs to get published to Maven OSS/Maven Central. Each build should get published to Maven OSS, but only manually selected builds (releases) should get published to Maven Central.  Does Shippable support manually triggered deployments (i.e. pull model)?,
457,[Consolidated feature request] Support additional platforms,micw, , ,OPEN,Hi,

in the documentaion I found:
(Coming soon) Our Windows minions are based on AWS AMI for Windows 2012.

Do you have any time plan when this feature will be available?

Kind regards,
Michael Wyraz.
,
431,Make project/build status links shorter and human readable,yihui, , ,OPEN,e.g. compare the badge URL's:

- Shippable: https://api.shippable.com/projects/53fa5b48c9a91cb5066f8468/badge/master
- Travis CI: https://travis-ci.org/yihui/formatR.svg

I strongly prefer Travis CI URL's (username/repo instead of a long hash string).,
407,Upgrade Cassandra Service to Datastax Community Edition,tristanwietsma, , ,OPEN,Moving from Shippable/sample_ruby_cassandra#3

The current Cassandra service minion is running 1.1.12, which is over a year behind the current stable release (2.0.9).

Travis-CI's Cassandra service is based on the Datastax Community Edition:
http://docs.travis-ci.com/user/database-setup/#Cassandra
http://planetcassandra.org/cassandra/

Please consider upgrading Cassandra to the latest stable release in the very near future. I'm running into API changes between the preferred client and your minion, rendering the service unusable.,
400,Tailing the logs,hhimanshu, , ,OPEN,Just like Travis-CI provides the way to tail the logs, it would be nice to have that feature in Shippable as well.  

manual scrolling is bad :(

@manishas @rageshkrishna ,
256,Update Git to support subtree command,tinchou, , ,OPEN,I have a project with a structure like this:

````
-
| - dev
| - docs
| - spikes
````

In order to deploy the `dev` folder to Heroku, I was using `git subtree` from my own computer as shown here: https://coderwall.com/p/ssxp5q. Now I'm trying to delegate the deployment step to Shippable.

When the minion finishes building and executes `git subtree push --prefix 'dev' heroku master` (set in the `after_success` section of my `shippable.yml` file), I get the following message:

````
git: subtree is not a git command. See git --help.
````

It seems that you have an outdated version without subtree support. Hope you can update Git to allow us to use this feature.

Also, there is no documentation on how to use a build environment different than Ubuntu 12.04. Should it be as follows? (I tried it, nothing changed)

````
build_environment: ubuntu1310
````

In case you want to check, the ID of my project is `53acd8e1c40d71a900e7e704`.

Thanks,
Martín Coll.,
237,BitBucket Pull Request Support,coreybutler, , ,OPEN,Any timeline on pull request support with BitBucket? Every CI service has Github integration... this would be a nice differentiating factor.,
190,[Consolidated Feature request] Integration with additional tools/services,pmatsinopoulos, , ,OPEN,Does Shippable support Solr integration?,
182,HTTP api for getting build status.,RX14, , ,OPEN,This would be useful for external applications such as an IDE where they can show whether the build is failing or not. Requires #53 ,
174,Code coverage enhancements,epicserve, , ,OPEN,It would be nice if on the coverage tab it showed the source code with hits or misses, when you click on the plus for each file.

Here is a screenshot of the area I'm referring to.

![shippable 2014-05-30 16-10-18 2014-05-30 16-11-38](https://cloud.githubusercontent.com/assets/191620/3136210/01c28838-e83f-11e3-8497-73828d453a66.jpg)
,
146,Add better build artifacts handling.,RX14, , ,OPEN,I would prefer the build artifacts to be handled like Jenkins. You should be able to upload multiple files by where they are instead or simply zipping the shippable folder. There should be an artifacts tab with all the different artifacts you specify and also a download of all of the artifacts (incl. code coverage and test results)
For example:
````yml
#shippable.yml

artifacts:
  - build/libs/Test-*.jar #Import a file
  - build/libs #Zip the libs folder and send it as an artifact
````

This would be a great addition to a service that I have otherwise found faultless (so far...),
105,[Consolidated Feature request] Support additional languages,manuel-rubio, , ,OPEN,Hi,

I'll try to build my ephp project in shippable and I can see an advice to open an issue to support that so, here is my issue! please, give support to erlang language!

My repo is: https://github.com/altenwald/ephp.git

Thanks.,
45,[Consolidated Feature request] Support for additional source control providers,voanhduy1512, , ,OPEN,Is Shippable supported custom git server (Gitalb) or will be supported in near feature?,
30,Schedule Runs,plotti, , ,OPEN,Is there a way to make shippable run an integration test once a week? Sometimes I'dont have code that i push during that week but the underlying apis might have changed, and I'd like to make sure that the software is still working. Any ideas?

Cheers
Thomas,