2014-10-08 sprint

10/08/2014 (Sprint start date 09/18/2014)


167 issues   (167 closed — 0 open)

2014-10-08 sprint

Arvados Contributor Wiki

This wiki has information related to the open source project. It's a work in progress, so come back often for updates and feel free to become a contributor.

Platform Overview

Release Notes

Key Components

Using Arvados

Contributing and hacking



Time tracking
Estimated time 78.50 hours
Issues by










Related issues
# Subject Story Points
Bug #3381 [Workbench] "jobs" view in Workbench shows failed job job but green progress bar 1.0
Feature #4042 [Crunch] run-command script supports running MxN tasks (e.g., M directories x N chromosomes) 1.0
Feature #2881 [OPS] Basic node manager that can start/stop compute nodes based on demand 5.0
Bug #4085 [Workbench] On pipeline instance page, queued jobs should have Cancel buttons 0.5
Bug #3825 [Crunch] crunch-job should save (very) large log files without filling up tmp space 1.0
Support #3902 [Support] Support: fix bugs and write tests second half 5.0
Bug #3261 [Keep] arv-put fails if permission system is enabled 1.0
Feature #3894 [Documentation] Document the run-tests.sh program on the Hacking pages on the wiki 1.0
Bug #3990 [Workbench] "Re-run pipeline" buttons should put the new pipeline-instance in the same project as the original 0.5
Bug #3882 [Workbench] fiddlesticks trying to cancel job that is already cancelled 0.5
Feature #3775 [Crunch] arv-crunch-job should work in local mode with an arvados-hosted repository 1.0
Bug #3828 [Keepproxy] insufficient error checking on "defer listener.Close()" lines in the tests, cf. jenkins run 846 where the keepproxy tests failed like that 0.5
Bug #4012 [Crunch] crunch-job bandaid: use eval/retry to wrap api calls that are likely to fail a long-running job due to a transient error condition 1.0
Bug #4046 [API] All jobs claim to be "next in queue". Bug in queue_position calculation in services/api/app/models/job.rb. It needs to increment index. 0.5
Bug #3634 [Workbench] Page content is the same as where you left it when navigating (back) to it with browser history 1.0
Bug #4025 [Workbench] Make it clear that "Move to project" button in the upper right of the project page moves the current project, not the current selection.
Bug #4036 [Workbench] When combining collections in a project, the new collection is created in Home project, but the user is still in current project.
Bug #4016 [Workbench] "Make sharing link" missing from collection pages.
Bug #4045 [Workbench] Provenance graph display regressed (possibly with #3036) and does not connect jobs/components via input/output pdh 0.5
Bug #3769 [API] In crunch-dispatch, throttle by bytes_per_minute or _node_minute 1.0
Bug #3782 [Workbench] Browser prompts about long-running JavaScript when trying to view a large Job log in viewer 0.5
Bug #3052 [Crunch] stdout from non-task programs invoked by crunch-job should propagate to job log. 0.5
Bug #4000 [API] "re-run with latest" yields fiddlesticks if pipeline template has been updated with new components 1.0
Bug #4004 Restore "All jobs/pipelines/collections" buttons to dashboard 0.5
Feature #4038 merges to master may use "fixes #" in addition to "closes #" 0.5
Bug #3961 [API] 9tee4: cannot create subproject under Home 0.5
Bug #3866 [SDK] Python KeepClient needs to construct URLs smarter 0.5
Bug #3848 [SDK] Python SDK should always set proper version when built from git 0.5
Bug #3889 [API] [Workbench] Refactor any functional tests that call multiple actions into either integration tests or independent functional tests. 1.0
Bug #3616 [Workbench] Workbench uses huge amounts of RAM when user downloads a large (> 1 GiB) file. 1.0
Bug #3871 [FUSE] Crashes on non-ASCII folder names 1.0
Bug #3038 Clean up recommended/default port numbers for API server and Workbench 1.0
Feature #3899 [Crunch] crunch-job and Workbench use the new job state attribute (see #3898) instead of paying attention to (and updating) running and success flags. 2.0
Bug #3168 [Crunch] Source git repository for a job should not be fetched into internal arvados repository over and over again even when requested node profile is not available 0.5
Bug #3949 [SDK] arv-get consumes excessive memory on very large collections 0.5
Story #3413 [Keep] Keep Server accepts list of trashable blocks 1.0
Bug #3890 [API] dots (and probably other valid characters) are being stripped from git repository names. The name field in the output of arv repository get_all_permissions is wrong (it strips characters that are valid) 0.5
Story #3187 [Workbench] Better information and features on "show pipeline instance" page 3.0
Support #3893 [Documentation] Explain in public wiki how to submit good/safe support requests and bug reports without accidentally leaking private information 0.5
Story #3898 [API] Job model has a single state attribute that can be updated and read instead of paying attention to running and success flags. 0.5
Story #3661 [Workbench] Add "Move" and "Copy" buttons to top of #show page for every kind of object that goes_in_folders 0.5
Bug #3147 [SDKs] Python clients should automatically retry failed API and Keep requests (including timeouts), in order to survive temporary outages like server restarts and network blips. 1.0
Feature #3338 [Workbench] Show number of items in each tab of the project page. 0.5
Bug #3927 [API] Db migration for User.owner_uuid NOT NULL constraint 0.5
Bug #3384 [Crunch] Termination of jobs due to 'Connection timed out'?