ec2-fleet-3.2.0

🎉 Major features and improvements

• Add option for scaling executors by node hardware in a new dropdown selection. Dropdown options for no scaling and scale by weights replace the "Scale Executors by Weight" checkbox (#415) @GavinBurris42

🐛 Bug fixes

• Add backwards compatibility for executor scaling feature (#424) @GavinBurris42
• Revert class name changes. Provides backwards compatibility for cloud configurations (#423) @GavinBurris42

📝 Documentation updates

• Update CasC documentation with use cases of executor scaling (#424) @GavinBurris42
• Update README to reflect support for EC2 Fleets created through through the CreateFleet API (#415) @GavinBurris42
• Update README with new permissions needed for the IAM Role (#415) @GavinBurris42
• Update CasC documentation to reflect new executor scaling feature (#415) @GavinBurris42

🚦 Tests

• Update Fleet Label Cloud and created tests for CasC (#418) @GavinBurris42

ec2-fleet-3.1.0

🎉 Major features and improvements

• Add support for fleets created through the CreateFleet API (#410) @GavinBurris42

ec2-fleet-3.0.2

🚨 Removed

• Revert "Update constructor to follow CasC best practices" (#416) @vineeth-bandi
  • Added missing validation to ensure maxSize is less than or equal to minSize. MaxSize that doesn't comply is automatically modified to minSize. (#406) @vineeth-bandi
  • Update constructor to follow CasC best practices (#406) @vineeth-bandi
  • Update casc docs to reflect new defaults, add unlisted fields to examples, and include EC2FleetLabelCloud examples (#406) @vineeth-bandi

📝 Documentation updates

• Update bug report issue template (#413) @pdk27

ec2-fleet-3.0.1

🐛 Bug fixes

• Added missing validation to ensure maxSize is less than or equal to minSize. MaxSize that doesn't comply is automatically modified to minSize. (#406) @vineeth-bandi

📝 Documentation updates

• Update constructor to follow CasC best practices (#406) @vineeth-bandi
• Adding missing config, warning for Spot Fleet usage, other minor changes (#404) @pdk27

👻 Maintenance

• Create release-drafter.yml (#408) @pdk27
• Update constructor to follow CasC best practices (#406) @vineeth-bandi

ec2-fleet-3.0.0

What's Changed

🚀 What’s New?

• oldId is no longer used and nodes & computers are no longer re-assigned to the new cloud objects (after configuration save) because they have been causing numerous hard-to-debug issues. The tracking of clouds that the nodes & computers belong to is instead done using cloud name.
• Cloud name:
  • Is required
  • Jenkins Cloud names are validated for uniqueness. Error messages are displayed in the UI for non-unique names used.
    
  • UI: unique cloud names are unmodifiable and a unique default name is suggested for new clouds e.g. FleetCloud-QXlsET6X
  • JCasC: users can now ask plugin to generate a unique default cloud name by specifying “” for name.

🐛 Bug Fixes

• #360 #322 Remove references of oldId and EC2FleetCloudAware class, remove tracking of cloud objects. Replaced cloud object tracking with tracking by name making cloud name a special configuration
• #397 MaxTotalUses didn't show the correct value in the UI, instead showed the default -1.
• #335 Resubmitted build uses the parameters of the last failed build instead of the last build

📓 Others

• Minor refactoring:
  • Renamed FleetNode#name to instanceId for clarity, added getDescriptor to return sub type
  • Added a check to not provision if Jenkins is quieting down or terminating
• Updated documentation
• Added AWS's warning about SpotFleet usage to ReadME.

✅ PRs

• #383
• #390
• #392
• 
#395

• #403

• #394
• #388
• #401
• #402
• #404

🗒️ Upgrade Steps

This release changes the way clouds and their agents are tracked in a backwards incompatible way. The following changes to cloud configurations are required:

• JCasc users:
  Fix configuration and reload configuration under Manage Jenkins -> Configuration as Code
  • remove usage of oldId, it is no longer used
  • make cloud names unique. JCasc users can ask plugin to name their clouds by specifying name: "" (Tip: Use the UI to identify clouds with duplicate names)
• UI users:
  • rename clouds with non-unique name
  • optionally, use dynamically generated default for new clouds

🤝 Thanks to the contributors for reporting issues and sharing helpful information:

🌟 @wzur-r3, @wosiu, @mwos-sl, @icep87, @limasilet, @benipeled, @tim-goto, @decoursin, @fluffysquirrels, @roded
🌟 @vineeth-bandi's first contribution

Full Changelog: ec2-fleet-2.6.0...ec2-fleet-2.7.0

ec2-fleet-2.7.1

What's Changed

🚀 What’s New?

• Plugin no longer auto-selects a fleet in Cloud configuration (under Manage Jenkins -> Configure Clouds -> Add new cloud)

🐛 Bug Fixes

• #347 #364 Removed auto-selection of fleet from drop-down in Cloud configuration form
• #329 Replaced usage of 'slave' with 'agent'

📓 Others

• Minor refactoring

✅ PRs

• #381 by @pdk27
• #385 by @naraharip2017
• #386 by @naraharip2017

🤝 Thanks to the contributors for reporting issues and sharing helpful information:

🌟 @tofanadrian3000, @benipeled, @buliverd53, @darthkisling
🌟 @naraharip2017's first contribution

Full Changelog: ec2-fleet-2.6.0...ec2-fleet-2.7.0

ec2-fleet-2.7.0

What's Changed

🚀 What’s New?

• Track and log all termination reasons for plugin initiated instance terminations.
• For the case when maxTotalUses is specified, remaining uses will be tracked separately instead of using maxTotalUses for it, maxTotalUses is now read-only.
• Add logs in post job action to expose results of tasks. Previously, tasks terminated with problems were not exposed.

🐛 Bug Fixes

• Fix maxtotaluses decrement logic (stopped decrementing limited uses all the way to -1, as -1 has a special meaning i.e. unlimited)
• #363 Skip termination of scheduled instances until next update cycle when busy.
• #322 Fix lost state on configuration change by persisting data, and rebuilding the state in periodic checks.

✅ PRs

• Fix 322 and 363 by @pdk27 in #376
• Fix unnecessary decrement of maxTotalUses and add clarity to logs around terminations triggered by plugin by @pdk27 in #375

📓 Others

• Minor refactoring / renaming
• Add new integration tests

🤝 Thanks to the contributors for reporting issues and sharing helpful information:

@wosiu, @qibint, @mwos-sl, @fdaca, @xocasdashdash, @igtsekov, @avikivity, @HDatPaddle, @naorc123, @ikaakkola, @kintarowins, @mtn-boblloyd, @pd-andrew-ferguson, @kherpel, @blackliner, @calumgilchristmodulr, @benipeled, @aluczakSumo, @limasilet, @iurisman, @fazdamoa

Full Changelog: ec2-fleet-2.6.0...ec2-fleet-2.7.0

ec2-fleet-2.6.0

What's Changed

• Make planned node name unique per EC2FleetCloud(#370) @pdk27
• Fix logs and help @pdk27
• Fix computation of excess workload and available capacity @pdk27 @cjerad

Issues Linked

#322
#359

ec2-fleet-2.5.3

What's Changed

• Upgrading the minimum required Jenkins version (#368) @pdk27
• Fix typo in jcasc configuration example (#353) @uriinf
• Updating AWS security policy example (#348) @Ar-Kan

ec2-fleet-2.5.2

What's Changed

• Fix getCurrentSpareInstanceCount - check busy nodes in given fleet, not whole Jenkins instance by @h-okon in #343

New Contributors

• @h-okon made their first contribution in #343

Full Changelog: ec2-fleet-2.5.1...ec2-fleet-2.5.2