These packages are needed in order to scrape device health metrics from
devices used by OSD and MON daemons.
smartmontools' smartctl is what we use in order to scrape devices' SMART
attributes and general health metrics.
In addition, we use nvme-cli tool on NVMe devices, which fetches
vendor specific NVMe related health metrics.
Ceph rely on these tools for proper functioning of the underlying layers
of devicehealth mgr module, and other mgr modules which use devicehealth
functionality (such as diskprediction_local, telemetry, dashboard).
Essentially, most of devicehealth commands rely on proper functioning of
smartctl, otherwise they lack the device health metrics.
For example, in case smartctl is missing, the commands:
ceph device scrape-daemon-health-metrics <who>
ceph device scrape-health-metrics [<devid>]
will not be able to scrape health metrics, and the command:
ceph device predict-life-expectancy <devid>
will not provide any meaningful output (since there are no metrics).
In short, when we scrape a device by its daemon (be it an OSD or a MON):
ceph device scrape-daemon-health-metrics <who>
The devicehealth module command eventually invokes a
block_device_get_metrics() call in either osd/OSD.cc or mon/Monitor.cc,
which wraps calls to both
block_device_run_smartctl() (spawns smartctl)
block_device_run_vendor_nvme() (spawns nvme)
in common/blkdev.cc.
Minimum version requirements:
'smartmontools' is the package name, which contains two utility
programs: 'smartd' and 'smartctl'. Ceph uses the latter.
Version 6.7 of smartctl first introduced the --json option (beta), which
allows to output the metrics in a JSON format. Since then a few
adjustments were made and the feature officially launched in smartctl
version 7.0.
Since we rely on the JSON format to process the metrics, we must have
smartmontools' smartctl version >= 7.
That said, we choose not to specify smartmontools version here on
purpose, since there might be a scenario where:
We specified smartmontools version to be >= 7.
smartmontools 7 is not available yet in rhel 8 / centos 8.
A user installs via rpm ceph-osd, for example.
smartmontools will not be installed (since version >= 7 is not available
in this repo yet).
Then the user upgrades to 8.3 (which should have smartmontools >= 7),
but smartmontools will not get upgraded (since it's not installed).
In the scenario where we do not specify a version, smartmontools 6.6
will be installed, but it will be upgraded to >= 7 when a user upgrades
(and if it's a fresh installation - version >= 7 would be installed
anyway).
nvme-cli does not have a minimum version.
We use 'Recommends' for both rpm and deb packages since we do not want
the installation to fail in case of conflicts. 'Recommends' weakens the
dependency to be installed in case possible, but ignores it in cases of
conflicts with other dependencies.
It's worth mentioning that smartmontools and nvme-cli dependencies exist
in ceph-container builds.
We add them here for the cases of bare metal installations.
In the future we will add a separate package (with smartmontools and
nvme-cli dependencies) that can be installed on any node (running
rbd-mirror, rgw, mds, mgr, etc.), in order to be able to collect the
health metrics of its devices and offer their life expectancy
prediction.
Fixes: https://tracker.ceph.com/issues/47479
Signed-off-by: Yaarit Hatuka <yaarit@redhat.com>
libzbd-devel is not available in any distros official repo at the time
of writing. but we packaged it for RHEL8/CentOS8. and uploaded its
packages to http://apt-mirror.front.sepia.ceph.com/lab-extras/8/. to
build test it to avoid the bitrot of ZBD bluestore backend at a minimal
level. but we cannot do more than this. because
- the kernel shipped by RHEL/CentOS8 does not have zoned block device
support enabled. see https://zonedstorage.io/distributions/linux/
- and we don't have zoned block device for testing in our lab.
so, in this change, "--with-zbd" option is added so we can pull
in the libzbd-devel package on demand.
Signed-off-by: Kefu Chai <kchai@redhat.com>
rpm,deb: drop /etc/sudoers.d/cephadm
Reviewed-by: Sebastian Wagner <sebastian.wagner@suse.com>
Reviewed-by: David Disseldorp <ddiss@suse.de>
Reviewed-by: Kefu Chai <kchai@redhat.com>
We've upgraded to Boost version 1.73
(1169dee259), and this version provides
working support for boost::context for s390x. Enable this in the RPM
packaging.
Fixes: https://tracker.ceph.com/issues/47304
Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
Current behavior (without this patch) is:
1. cephadm package installs cephadm at /usr/sbin/cephadm
2. cephadm package installs /etc/sudoers.d/cephadm
3. !!! BUT this file refers to a non-existent executable (/usr/bin/cephadm) !!!
4. the PR that introduced this sudoers file (and this discrepancy) was merged in 2019
5. nobody noticed the discrepancy until now
My conclusion: the file /etc/sudoers.d/cephadm is not needed for cephadm to
work.
Fixes: https://tracker.ceph.com/issues/47112
Signed-off-by: Nathan Cutler <ncutler@suse.com>
This reverts commit a55c1dd0fa.
Kept to keep upgrades from older point releases working.
This module can be removed as soon as we no longer
support upgrades from old octopus point releases.
Revert "build/debian: remove osd_support"
This reverts commit 8ff2824beb.
Fixes: https://tracker.ceph.com/issues/47109
Signed-off-by: Sebastian Wagner <sebastian.wagner@suse.com>
the service offered by https://www.diskprophet.com is not accessible
anymore. before we have a replacement of it, let's drop
mgr-diskprediction-cloud. and remove its packaging.
Signed-off-by: Kefu Chai <kchai@redhat.com>
since seastar dropped support from C++14, we have to move to a compiler
with a decent C++17 support.
in this change, gcc-toolset-9 is used for compiling ceph if seastar is
enabled. use version >= 9.2.1-2.2, because of
https://bugzilla.redhat.com/show_bug.cgi?id=1853900
Signed-off-by: Kefu Chai <kchai@redhat.com>
mgr plugin to deploy and configure MDSs in response to degraded file system
MDS instance management as per changes to:
* 'max_mds' option
* 'standby_count_wanted' option
* mds liveness and transitions from standby to active
mds_autoscaler plugin test credit goes to Sebastian Wagner.
Fixes: https://tracker.ceph.com/issues/40929
Signed-off-by: Milind Changire <mchangir@redhat.com>
Since e5b585d15d ceph-volume depends on
python-ceph-common. This commit introduces this dependency for the
ceph-osd rpm (which includes ceph-volume) and installs the dependency
for tox runs.
Fixes: https://tracker.ceph.com/issues/46772
Fixes: e5b585d15d
Signed-off-by: Jan Fajerski <jfajerski@suse.com>
seastar uses setjmp() and longjmp() to implement coroutine, but
longjmp() is defined as ____longjmp_chk() by GCC if _FORTIFY_SOURC is
defined. ____longjmp_chk() simply bails out with an error message if
the dest stack pointer is higher than the src stack pointer, or the dest
stack pointer is not in the sigaltstack. in the case of seastar, the dst
%sp is not necessarily higher than src stack pointer, and it's not
handling a signal for switching the thread context. that's why we have
the "longjmp causes uninitialized stack frame" error when running
crimson-osd on RHEL/CentOS 8 using the prebuilt rpm packages.
the optflags rpm macro adds -D_FORTIFY_SOURCE=2 to CFLAGS and CXXFLAGS,
so even seastar tries to pass -U_FORTIFY_SOURCE to GCC, there is chance
that cmake append CXXFLAGS at the end of the option list passed to GCC.
and this renders seastar's attempt to undefine _FORTIFY_SOURCE useless.
another way to address this issue is to undefine this macro in
seastar:src/core/thread.cc. but since seastar tries neutralize the macro
in its cmake script instead of source file, i assume they have their
considerations. let's drop it in the rpm recipe instead.
Signed-off-by: Kefu Chai <kchai@redhat.com>
in hope to simplify the enabling of crimson-osd. as we will use cephadm
as the way for deploying crimson-osd in future. it's unlikely we will
have classic osd and crimson osd co-located in the same system. so
this is most simple way to enable crimson-osd with minimum efforts.
Signed-off-by: Kefu Chai <kchai@redhat.com>
We no longer need it, as distributions are converging on /usr/libexec
since FHS 3.0 has specified it as the acceptable path for libexecdir.
Fedora has been using /usr/libexec since the beginning, and
openSUSE has accepted software installed into /usr/libexec since
September 2019 in openSUSE:Factory. openSUSE Leap 15.x still uses
/usr/lib for %_libexecdir, but the rest of the spec should handle the
change gracefully.
Signed-off-by: Neal Gompa <ngompa13@gmail.com>
This mimics the setting passed to CMake in other package builds,
and ensures that systemd units are installed into the right place.
Signed-off-by: Neal Gompa <ngompa13@gmail.com>
mgr/cephadm: use jinja2 to render templates
Reviewed-by: Ernesto Puerta <epuertat@redhat.com>
Reviewed-by: Sebastian Wagner <sebastian.wagner@suse.com>
otherwise, "osc build" fails like:
[ 5545s] libcephfs-devel-16.0.0-2172.ga59fb9ffac.x86_64.rpm: directories not owned by a package:
[ 5545s] - /usr/include/cephfs/metrics
this addresses a regression introduced by
c669de12e0
Signed-off-by: Kefu Chai <kchai@redhat.com>
The librbd parent cache is an initial testbed for librbd plugin
support. The shared library will be packaged with librbd for the
time being.
Signed-off-by: Jason Dillaman <dillaman@redhat.com>
* refs/pull/34596/head:
test: add a new program for testing ino_release_cb
client: add a new inode release request callback
client: expose ceph_ll_register_callbacks via libcephfs
client: move callback typedefs and arg struct into ceph_ll_client.h
client: rename ceph_statx.h to ceph_ll_client.h
client: make client_dentry_callback_t more friendly for C
client: only override umask_cb with non-NULL values
Reviewed-by: Xiubo Li <xiubli@redhat.com>
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Reviewed-by: Zheng Yan <zyan@redhat.com>
Reviewed-by: Greg Farnum <gfarnum@redhat.com>
Adds a `radosgw-admin` subcommand and walks the associated bucket
indices and manifests to generate the list of rados objects that
represent the rgw objects in the bucket(s).
Also adds a tool named `rgw-orphan-list`, which uses the radoslist
subcommand, that produces a list in a local file of what appear to be
rgw orphans.
Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
...so we can use it to include other definitions and types that need to
be shared with in-tree code that doesn't want to include libcephfs.h.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
