Skip to content

Latest commit

 

History

History
118 lines (92 loc) · 3.54 KB

RELEASE_NOTES.md

File metadata and controls

118 lines (92 loc) · 3.54 KB
Raw

execution_engine2 (ee2) release notes

=========================================

0.0.15

  • Update JobRunner and Execution Engine to substitute internal urls to go through cloudflare

0.0.14

  • Update clients to work with mongo7 by updating pymongo and mongoengine
  • Fix linting issues from flake8 and black
  • Update outdated libraries that can no longer be pip installed
  • The latest Dockerfile build contains conda which installs python 3.12.4 so these changes will pin python to 3.10 and save us headache in the future
  • Speed up tests by moving docker build step out
  • Create matrix of tests to run against mongo3.6 and mongo7

0.0.13

  • Bump version in impl file and sync branches

0.0.12

  • Forcing black to 22.1.0 to make sure that GHA doesn't suddenly fail
  • Prevent jobs that never ran from submitting job execution stats

0.0.11

  • Add ability for kbase user to contact condor via token

0.0.10

  • Fixes bug with ee2 not recording all jobs with the catalog during the process of finishing a job
  • Updates GHA with black and flake8
  • Fix flake8 and black formatting issues by formatting MANY files
  • Updated docs for installing htcondor
  • Update many python libs in requirements.txt

0.0.9

  • Update GHA with latest actions, remove old actions
  • Change job defaults to result in
  • NJS and KB_UPLOAD 5 Jobs
  • Bigmem nodes with 250GB of ram take 1 job
  • Bigmem nodes with 1TB of ram take 4 jobs
  • Remove Jars from built image / cleanup for trivy

0.0.81

  • Updated HTCondor Clients, New Base Image
  • Use default GH actions
  • Updated precommit hooks

0.0.8

  • Fixed a bug that could, seemingly rarely, cause job and log updates to be applied to the wrong Mongo collection.
  • Removed docker shaded jar that causes log4j scan to appear positive with Trivy

0.0.7

  • Fixed a bug that could cause missing queued timestamps if many jobs were submitted in a batch

0.0.6

  • Release of MVP

0.0.5

New Endpoints

  • run_job_batch
  • retry_job
  • retry_jobs
  • abandon_children

BugFixes

  • Fix a bug that caused job requirements from the catalog in CSV format to be ignored other than the client group

Other features and refactoring

  • Refactor run_jobs_batch endpoint to cache catalog calls for batch jobs, submit entire batch to condor in one transaction
  • Refactored tests
  • Removed slack messages for running jobs
  • Added CreatedJobsReaper
  • Added retry_job and retry_jobs endpoint along with ADRs
  • Full EE2 admins can now submit job requirements when running jobs via run_job_batch and run_job. See the SDK spec for details.
  • Added ADRs for retry endpoint

0.0.4

  • Fix up tests
  • Remove dependency on slack
  • Add batch endpoints, cancel_jobs now cancels child jobs
  • Rename prod branch to "main"

0.0.3.4

  • Change 7 day periodic_remove to 7 day hold
  • Prevent reaper from prematurely exiting

0.0.3.3

  • Change log appends from update to $push
  • Change first log linepos to position 0 instead of 1

0.0.3.2

  • Change 2 db state updates for startjob into 1
  • Rework Perms Model
  • Stop Pymongo/Mongoengine clients on each request, and make it an instance variable for each worker thread.

0.0.3.1

  • Logging endpoint returns success value now
  • Added script to fix backfilled records
  • Logs transferred for held jobs ON_EXIT_OR_EVICT
  • Enabled job log monitor again

0.0.3

  • Fix Quay Build

0.0.2

  • Fixed bug with service version displaying release instead of git commit
  • Updated clients
  • Update transfer script to handle failures

0.0.0

  • Module created by kb-sdk init
  • We Were Young