Ceph Releases (general)

Active stable releases

VersionInitial releaseLatestEnd of life (estimated)
nautilusMar 201914.2.5
mimicJun 201813.2.82020-06-01
luminousAug 201712.2.122019-06-01

Understanding the release cycle

Starting with the Nautilus release (14.2.0), there is a new stable release cycleevery year, targeting March 1st. Each stable release series will receive a name(e.g., ‘Mimic’) and a major release number (e.g., 13 for Mimic because ‘M’ isthe 13th letter of the alphabet).

Releases are named after a species of cephalopod (usually the commonname, since the latin names are harder to remember or pronounce).

Version numbers have three components, x.y.z. x identifies the releasecycle (e.g., 13 for Mimic). y identifies the release type:

  • x.0.z - development releases (for early testers and the brave at heart)

  • x.1.z - release candidates (for test clusters, brave users)

  • x.2.z - stable/bugfix releases (for users)

This versioning convention started with the 9.y.z Infernalis cycle. Prior tothat, versions looked with 0.y for development releases and 0.y.z for stableseries.

Development releases (x.0.z)

Each development release (x.0.z) freezes the master development branchand applies integration and upgrade tests before it is released. Oncereleased, there is no effort to backport fixes; developer focus is onthe next development release which is usually only a few weeks away.

  • Development release every 8 to 12 weeks

  • Intended for testing, not production deployments

  • Full integration testing

  • Upgrade testing from the last stable release(s)

  • Every effort is made to allow offline upgrades from previousdevelopment releases (meaning you can stop all daemons, upgrade, andrestart). No attempt is made to support online rolling upgradesbetween development releases. This facilitates deployment ofdevelopment releases on non-production test clusters withoutrepopulating them with data.

Release candidates (x.1.z)

There is a feature release roughly eight (8) weeks prior to the plannedinitial stable release, after which focus shifts to stabilization andbug fixes only.

  • Release candidate release every 1-2 weeks

  • Intended for final testing and validation of the upcoming stable release

Stable releases (x.2.z)

Once the initial stable release is made (x.2.0), there aresemi-regular bug-fix point releases with bug fixes and (occasionally)small feature backports. Bug fixes are accumulated and included inthe next point release.

  • Stable point release every 4 to 6 weeks

  • Intended for production deployments

  • Bug fix backports for two full release cycles.

  • Online, rolling upgrade support and testing from the last two (2)stable release(s) (starting from Luminous).

  • Online, rolling upgrade support and testing from prior stable pointreleases

For each stable release:

Lifetime of stable releases

The lifetime of a stable release series is calculated to be approximately 24months (i.e., two 12 month release cycles) after the month of the first release.For example, Mimic (13.2.z) will reach end of life (EOL) shortly after Octopus(15.2.0) is released. The lifetime of a release may vary because it depends onhow quickly the stable releases are published.

In the case of Jewel and Kraken, the lifetime was slightly different thandescribed above. Prior to Luminous, only every other stable release was an “LTS”release. Therefore,

  • Upgrade scenarios “Jewel -> Kraken -> Luminous” and “Jewel -> Luminous” wereexpected to work.

  • Upgrades from Jewel or Kraken must upgrade to Luminous first before proceedingfurther (e.g., Kraken -> Luminous -> Mimic but not Kraken -> Mimic).

  • Jewel was maintained until Mimic was released (June 2018).

  • Kraken is no longer maintained.

Detailed information on all releases, past and present, can be found at Ceph Releases (index)

Release timeline

Datedevelopmentnautilusmimicluminouskrakenjewelinfernalishammergiantfireflyemperor
Dec 201913.2.8
Dec 201914.2.5
Nov 201913.2.7
Sep 201914.2.4
Sep 201914.2.3
Jul 201914.2.2
Jun 201913.2.6
Apr 201914.2.1
Apr 201912.2.12
Apr 201915.0.0
Mar 201914.2.0
Mar 201913.2.5
Mar 201914.1.1
Feb 201914.1.0
Jan 201912.2.11
Jan 201913.2.4
Jan 201913.2.3
Nov 201812.2.10
Nov 201814.0.1
Nov 201812.2.9
Sep 201813.2.2
Sep 201812.2.8
Jul 201810.2.11
Jul 201813.2.1
Jul 201812.2.7
Jul 201812.2.6
Jun 201813.2.0
May 201814.0.0
May 201813.1.0
Apr 201812.2.5
Apr 201813.0.2
Feb 201812.2.4
Feb 201812.2.3
Feb 201813.0.1
Dec 201712.2.2
Oct 201710.2.10
Sep 201712.2.1
Aug 201711.2.1
Aug 201712.2.0
Aug 201713.0.0
Aug 201712.1.4
Aug 201712.1.3
Aug 201712.1.2
Jul 201710.2.9
Jul 201710.2.8
Jul 201712.1.1
Jun 201712.1.0
May 201712.0.3
Apr 201710.2.7
Apr 201712.0.2
Mar 201710.2.6
Mar 201712.0.1
Feb 20170.94.10
Feb 201712.0.0
Jan 201711.2.0
Jan 201711.1.1
Dec 201610.2.5
Dec 201610.2.4
Dec 201611.1.0
Oct 201611.0.2
Oct 201611.0.1
Sep 201610.2.3
Aug 20160.94.9
Aug 20160.94.8
Jun 201610.2.2
Jun 201611.0.0
May 201610.2.1
May 20160.94.7
Apr 201610.2.0
Apr 201610.1.2
Apr 201610.1.1
Mar 201610.1.0
Mar 201610.0.5
Mar 201610.0.4
Feb 20169.2.1
Feb 20160.94.6
Feb 201610.0.3
Jan 201610.0.2
Dec 201510.0.1
Nov 20159.2.0
Nov 20150.80.11
Nov 201510.0.0
Oct 20150.94.5
Oct 20150.94.4
Oct 20159.1.0
Aug 20150.94.3
Aug 20159.0.3
Jul 20150.80.10
Jul 20159.0.2
Jun 20150.94.2
Jun 20159.0.1
May 20159.0.0
Apr 20150.94.1
Apr 20150.94
Apr 20150.87.2
Mar 20150.80.9
Feb 20150.87.1
Feb 20150.93
Feb 20150.92
Jan 20150.80.8
Jan 20150.91
Dec 20140.90
Dec 20140.89
Nov 20140.88
Oct 20140.87
Oct 20140.80.7
Oct 20140.80.6
Oct 20140.86
Sep 20140.85
Aug 20140.84
Jul 20140.80.5
Jul 20140.80.4
Jul 20140.80.3
Jul 20140.80.2
Jul 20140.83
Jun 20140.82
Jun 20140.81
May 20140.80.1
May 20140.80
Apr 20140.79
Mar 20140.78
Feb 20140.77
Jan 20140.76
Jan 20140.75
Dec 20130.72.2
Dec 20130.74
Dec 20130.73
Nov 20130.72.1
Nov 20130.72