Skip to content

Latest commit

 

History

History
52 lines (38 loc) · 2.37 KB

CHANGELOG.md

File metadata and controls

52 lines (38 loc) · 2.37 KB

Changelog

[0.1.8] - 2019-1-17

Changed

Added

  • [Add a gradual weight when adding/removing an OSD][pr#119]
  • [Remove the journal when removing an OSD][pr#121]
  • [Add a Bluestore OSD manually][pr#124]
  • [Remove a Manually made Bluestore OSD][pr#125]

[0.1.7] - 2019-12-24

Changed

  • [Use the new blkid function get_tag_value which is safer than do_probe][pr#103]
  • [Cargo clippy and fmt fixes][pr#104]
  • [stdout, stderr, and pid files are now configurable in the bynar.json file][pr#106]
  • [Daemon exits the process gracefully on Bynar error while daemonized][pr#107]
  • [Bynar and disk-manager now check if a daemonized process exists first so only a single instance of disk-manager or bynar runs at a time][pr#109]
  • [Change osd_id get function to query the osd metadata for the osd id][pr#111]
  • [Change metadata checks for devices and osd ids for Bluestore Lvms so they use ceph-volume lvm list to get the proper metadata for checks][pr#113]

Added

  • [Made Bynar main program into a daemon process when passed the --daemon flag. Current signals handled are SIGINT, SIGCHLD, SIGHUP, and SIGTERM, the rest remain with default behavior][pr#105]
  • [Made Disk-Manager into a daemon process when passed the --daemon flag. Signal handling and config files handled the same ways as the Bynar daemon][pr#108]
  • [Add partprobe call if blkrrpart call fails][pr#110]
  • [Add check for osd_id in the cluster. If it IS in the cluster skip adding in the osd][pr#112]
  • [Add smartpassed logging][pr#114]

[0.1.6] - 2019-11-22

Changed

  • [Updates to the crate versions][pr#88]
  • [Revert ZMQ to ~0.8 to retain CurveKeyPair representation][pr#89]
  • [Change filter_disks so it filters out /boot and /boot/efi partitions and disks][pr#90]
  • [Updated pnet and lock file][pr#91]
  • [Upgrade ceph crate to latest version][pr#94]
  • [Replace ceph_safe_disk crate with Ceph command call osd safe-to-destroy][pr#95]
  • [Change ceph backend remove_disk helper functions to follow proper removal procedure][pr#96]
  • [Update dependencies for the debian][pr#100]

Added

  • [Added smartctl health scan check][pr#93]
  • [Add a special case for Disk type block devices so they only undergo Scan transition][pr#97]
  • [Add unmounts and mounts to functions and clean up temporary mounts][pr#98]
  • [Add Attribute to CephConfig to filter out non-OSD devices and add Protobuf message type to handle skipped disk behavior][pr#100]