Skip to content
This repository has been archived by the owner on Sep 21, 2021. It is now read-only.

Dropwizard082 basiccallbak #1

Open
wants to merge 80 commits into
base: master
Choose a base branch
from

Conversation

sverhoeven
Copy link
Owner

No description provided.

sverhoeven added 30 commits May 27, 2013 17:27
Add test for toString of JobLauncherConfiguration.
The url of the second job was the url of the first job with the second job identifier appended.
Because path() method of uri builder keeps appending.
Fixed by moving path() outside loop with a template variable.
As output like stdout/stderr could contain details about exception.
See xenon-middleware/xenon#73

Now `octopus.files().cancelJob(job)` method returns status so we use that.
See xenon-middleware/xenon#106

So use the Ocotopus job identifier as key.
Replaced usage of uuid with simple string, to prevent expectations that key will be an UUID.
…nding/running jobs.

Fixes #1.

Document and test the different preStage/postStage variants.

When job times out then don't download the sandbox. Only download sandbox for jobs that are done.

After job cancelation remove the sandbox.
…topus(octopus).

Octopus job is more strict so add a JobDescription during testing.
…n expected job description.

Job can be cancelled before the job status has been fetched, added test for it.
…Octopus and runs shell commands.

The test to verify the http client can handle MAC authentication has been refactored.
Instead of using an external http server to test against in now uses an internal one.
The internal http server is part of the apache http client tests.
The integration config file is not needed anymore.
Octopus Sandox requires a Files object instead of an Octopus object.
…e using files engine instead of another path.
blootsvoets and others added 30 commits November 13, 2014 11:27
They are seen as errors by Netbeans
Get Osmium to compile cleanly
Upped version as Osmium is java 8 compliant.
environment now comes after stderr and before the optional callback.
This would be relevant to compute on multiple different types of clusters.
The configuration is backwards compatible, so joblauncher.yml can state:
```
xenon:
  scheduler:
    scheme: local
    queue: multi
  ...
```
or:
```
xenon:
  schedulers:
    mycomputer:
      scheme: local
      queue: multi
    mycluster:
      scheme: ssh
      queue: single
  defaultScheduler: mycomputer
  ...
```
The property `scheduler` is mutually exclusive with `schedulers` and `defaultScheduler`.
The configuration now takes multiple launchers, each with its own
sandbox and scheduler. This way both SSH and local file systems
can be supported, along with the scheduling system.
The mac auth for callback can not be implemented in latest http client, switching to basic auth for callbacks.
…ropwizard.

Upgraded cobertura-maven-plugin so fix error.
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants