Ceph blog stories provide high-level spotlights on our customers all over the world
This is the first bugfix release for Kraken, and probably the last release of
the Kraken series (Kraken will be declared “End Of Life” (EOL) when Luminous
is declared stable). It contains a large number of bugfixes across all Ceph
We recommend that all v11.2.x users upgrade.
For more detailed information, see the complete changelog.
In previous versions, if a client sent an op to the wrong OSD, the OSD
would reply with ENXIO. The rationale here is that the client or OSD is
clearly buggy and we want to surface the error as clearly as possible.
We now only send the ENXIO reply if the osd_enxio_on_misdirected_op option
is enabled (it’s off by default). This means that a VM using librbd that
previously would have gotten an EIO and gone read-only will now see a
blocked/hung IO instead.
There was a bug introduced in Jewel (#19119) that broke the mapping behavior
when an “out” OSD that still existed in the CRUSH map was removed with ‘osd rm’.
This could result in ‘misdirected op’ and other errors. The bug is now fixed,
but the fix itself introduces the same risk because the behavior may vary between
clients and OSDs. To avoid problems, please ensure that all OSDs are removed
from the CRUSH map before deleting them. That is, be sure to do:
ceph osd crush rm osd.123
ceph osd rm osd.123
This release greatly improves control and throttling of the snap trimmer. It
introduces the “osd max trimming pgs” option (defaulting to 2), which limits
how many PGs on an OSD can be trimming snapshots at a time. And it restores
the safe use of the “osd snap trim sleep” option, wihch defaults to 0 but
otherwise adds the given number of seconds in delay between every dispatch
of trim operations to the underlying system.