0.5.0 - Improved reallocation, state file, and more

• Makes file management optional in the client. is currently still required by the api, need to change that next. This behavior can also be achieved by including sync: {} in the job definition.
• Uses the imds sdk to perform reallocation at the node, rather than via external tracking with the kelpie api. The maximum number of job failures a node can have before reallocating itself is configured with the environment variable MAX_JOB_FAILURES. defaults to 5. Any in-progress uploads will complete before the instance is reallocated.
• Allows users to set a max idle time. After the configured time, the instance will reallocate itself. This value is configured with the environment variable MAX_TIME_WITH_NO_WORK_S. Defaults to 0, which disables it. THIS IS NOT A SCALE-IN, the node will be replaced with a different one.
• Kelpie keeps a state file tracking in-progress downloads and uploads. this allows the user application to check these things in case they are relevant. The location of this file is passed into the job as environment variable KELPIE_STATE_FILE
• The s3 request timeout has been removed. was previously the default value of 120s. This has mostly not effected users because kelpie uses parallel uploading, so each 10mb chunk typically completes in well under 120s. however, with many files being uploaded over a slow enough connection, it is possible for this timeout value to be reached.
• Passes the kelpie job id into the job as environment variable KELPIE_JOB_ID
• update to a supported fork of pkg which allows using node 20.

Release 0.4.4

Release 0.4.4

Release 0.4.3

Release 0.4.3

Release 0.4.2

Release 0.4.2

Release 0.4.1

Release 0.4.1

Release 0.4.0

Release 0.4.0

Release 0.3.0

Release 0.3.0

Release 0.2.0

Release 0.2.0

Release 0.1.2

Release 0.1.2

Release 0.1.1

Release 0.1.1