News / 04 November 2016

iRODS 4.1.10 is released


The iRODS Consortium and RENCI are pleased to announce iRODS 4.1.10.

This is a bugfix release which addresses free_space calculation and updating, izonereport and irods-grid behavior, dynamic PEP behavior, and other small issues.

This is a required upgrade from 4.1.9 for multi-server Zones. Bug [#3306] prevented multi-server Zones from performing the free_space check correctly, as the calculation was executed too early (on the wrong server). iRODS 4.1.8 is not affected by this bug.

This release consists of 42 commits from 4 contributors and closed 24 issues.

The latest files are available at ftp://ftp.renci.org/pub/irods/releases/4.1.10/.

The release notes include:

Feature

  • New microservice, static PEPs to update unixfilesystem resource free_space [#3307] [#3312]

  • Note:
    Upgrading to 4.1.10 will not automatically add the two new static PEPs
    to core.re. To avoid spurious DEBUG messages, add the following two
    empty definitions to core.re:

    acPostProcForParallelTransferReceived(*leaf_resource) {}
    acPostProcForDataCopyReceived(*leaf_resource) {}

    Note:This updated feature (along with [#3306]) changes the optional
    unixfilesystem context string keyword from 'high_water_mark' to its
    semantic complement 'minimum_free_space_for_create_in_bytes'. Using
    the deprecated 'high_water_mark' or 'required_free_inodes_for_create'
    will write a LOG_NOTICE to the server log.
    (Read More)

    Bug Fixes

  • Fix for microservice parameter limitation [#3092] [#3095]

  • Fixes for unixfilesystem and free_space check [#3247] [#3305] [#3306] [#3311] [#3340]

  • Fix for reading past end of buffer [#3255]

  • Better debugging [#3260] [#3308] [#3313] [#3348] [#3351]

  • Fixes for izonereport [#3294] [#3303]

  • Fix for irods-grid [#3301]

  • Fix for list() microservice in rule engine [#3304]

  • Fix for dynamic PEP documentation [#3314]

  • Fix for random resource hierarchy logic [#3315]

  • Fix for ilocate whitespace in results [#3332]

  • Fix for delay rule in dynamic PEPs [#3342]


  • Terrell Russell