The influxdb module won't run if python-influxdb is not present (but it will be
graceful about not running). That means python-influxdb is a dependency of
that module, not mgr itself. However, we are not (yet) packaging the modules
separately. (When we do, this could become a Requires: of the module.)
RPM itself does not support "Recommends", and ignores this line. Higher-level
tools may or may not support it, so put this line in a SUSE-only conditional
since we know that zypper supports it.
Signed-off-by: Nathan Cutler <ncutler@suse.com>
Signed-off-by: Tim Serong <tserong@suse.com>
when osd is more than 14 in one host,it report error that aio is not enough.
As the default aio-max-nr is 65536, one OSD needs 4096 aios and other programs may use aios.
This patch change aio-max-nr to 1048576 when install ceph-osd rpm package and debian package
Signed-off-by: chenliuzhong <liuzhong.chen@easystack.cn>
os/bluestore: allow reconstruction of osd data dir from bluestore bdev label
Reviewed-By: Alfredo Deza <adeza@redhat.com>
Reviewed-by: Kefu Chai <kchai@redhat.com>
ceph-osdomap-tool into ceph-osd subpackage
ceph-monstore-tool into ceph-mon subpackage
ceph-kvstore-tool into the ceph-base subpackage
Fixes: http://tracker.ceph.com/issues/21762
Signed-off-by: Nathan Cutler <ncutler@suse.com>
As reported by Ademar de Souza Reis Jr <areis@redhat.com>:
This symlink should not be necessary anymore. QEMU is properly linked to
ceph/librbd these days (look at the qemu-block-rbd sub-package in Fedora and
the respective package in RHEL).
The symlink was a hack from a time when librbd was distributed and supported by
ceph and the rbd driver would be enabled at runtime only when/if the symlink
was present.
So the right fix is to actually get rid of the symlink altogether and never
touch (much less own) /usr/lib64/qemu.
Signed-off-by: Nathan Cutler <ncutler@suse.com>
The ceph-fuse(8) man page describes using the "fusermount" tool to mount
CephFS. This utility is in the "fuse" package. Prior to this change, the
ceph-fuse RPM only depends on "fuse-libs", not "fuse".
The Debian packaging has always depended on the full "fuse" (previously
"fuse-utils") package, so we should make the RPM packaging do the same.
Fixes: http://tracker.ceph.com/issues/21057
Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
rpm expands all macros in a .spec file, even those in comments. Drop the
percent signs so rpm will not expand these.
This change silences rpmlint's warning about macros in comments.
Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
Header files are currently installed under a hardcoded "include" path.
Use CMAKE_INSTALL_INCLUDEDIR instead, which defaults to "include" but
allows for customisation alongside other (already used) CMAKE_INSTALL_X
paths.
Signed-off-by: David Disseldorp <ddiss@suse.de>
Add a set of new tests for the case when public_addr and public_bind_addr
are different for a mon. In order to test this properly I had to employ
port forwarding with socat. This helps simulate what would happen in a
environment like Kubernetes. socat is now a build dependency.
Also, moved jq_success to ceph-helpers.sh and refactored run_mon to enable
creating the mons without creating the rbd pool immediately.
Signed-off-by: Bassam Tabbara <bassam.tabbara@quantum.com>
If FIRST_ARG is undefined, the rpms will show an error on upgrade
because the condition in the postun script gets broken.
This was a regression introduced by commit for issue 20077 that moved
ceph-disk unit files to ceph-base.
Fixes: http://tracker.ceph.com/issues/20077
Signed-off-by: Boris Ranto <branto@redhat.com>
RHEL's ceph-common package is "1". Move our upstream packages' epoch
ahead of that, so that upstream will always override whatever is in
RHEL, even if it happens to have a higher NVR one day.
Fixes: http://tracker.ceph.com/issues/20508
Signed-off-by: Ken Dreyer <kdreyer@redhat.com>
Although under normal packaging circumstances this would probably not be done,
in Ceph there is no reasonable deployment scenario where multiple major
versions of libcephfs would need to be present at the same time.
Signed-off-by: Nathan Cutler <ncutler@suse.com>
All options to en/disable inclusion of libraries or other software
are of the format WITH_ so that the Cmake commaind like ahs all
WITH_* options. The WITH_=ON option will result in a HAVE_ setting
in CMAKE so that tests can use that variable.
Last "abusers" to actually follow this format.
- HAVE_BABELTRACE
- HAVE_ZFSLIB is fixed in #15907
Signed-off-by: Willem Jan Withagen <wjw@digiware.nl>
9164804416 introduced a build error:
ceph-common-12.1.0+git.1498286248.2fcedc7b3d-1.1.x86_64.rpm: directories not
owned by a package:
- /usr/share/doc/packages/ceph
The %docdir directive is a way of flagging anything in that directory as being
documentation. It does not actually package the directory. And we don't need
it because we're not dumping a large number of files into this directory.
For more information, see the "Directory-related Directives" section of
http://ftp.rpm.org/max-rpm/s1-rpm-inside-files-list-directives.html
Signed-off-by: Nathan Cutler <ncutler@suse.com>
This implements a create-self-signed-cert command for the ceph-mgr
restful interface.
It is designed so that it will try to restart the module once the cert
is created.
Signed-off-by: Boris Ranto <branto@redhat.com>
The SUSE and Fedora packaging guidelines specify that subpackages _may_ have
their own License line if it is necessary. (Hopefully it's not, because
maintaining one License line is much easier.)
Signed-off-by: Nathan Cutler <ncutler@suse.com>
Although "which" is only used in the tests, we need the dependency
to be picked up by install-deps.sh.
This is a stopgap measure until we can get rid of the dependency entirely by
replacing "which" with "type -P" in the tests.
Fixes: http://tracker.ceph.com/issues/20127
Signed-off-by: Nathan Cutler <ncutler@suse.com>
Modify cmake files to take advantage or re-organization of dmclock's
cmake set-up, specifically not having dmclock's cmake files call
add_test. Remove dmclock tests from being dependencies on ceph's
"test" target. Make change so dmclock tests are not built unless
WITH_TESTS and WITH_DMCLOCK_TESTS are both set. This is so openSUSE
Leap will build correctly.
Signed-off-by: J. Eric Ivancich <ivancich@redhat.com>
We need to have ceph-base installed before ceph-selinux to use ceph-disk
in %post script. The default ordering is random and so the installation
randomly failed to relabel the files.
Fixes: http://tracker.ceph.com/issues/20184
Signed-off-by: Boris Ranto <branto@redhat.com>
The SELinux package now requires the ceph-disk binary but that one was
part of the ceph-osd package. The ceph-disk python library is already
packaged in ceph-base so moving ceph-disk to ceph-base seems like a
reasonable next step.
Signed-off-by: Boris Ranto <branto@redhat.com>
We can take advantage of ceph-disk fix subcommand when doing a package
install. We will keep using the differential fixfiles command otherwise.
We also need to add relabel for /usr/bin/ daemons so that we could use
this.
Fixes: http://tracker.ceph.com/issues/20077
Signed-off-by: Boris Ranto <branto@redhat.com>
Since the Beast frontend uses boost::context which is not supported on
s390x.
Signed-off-by: Nathan Cutler <ncutler@suse.com>
Signed-off-by: Kefu Chai <kchai@redhat.com>
This is the simplest way to generate the keys and probably the least
likely to cause trouble in the future.
Signed-off-by: Boris Ranto <branto@redhat.com>
This allows users to create their own certificates that are signed by a
CA. If the keys are not present then a self-signed certificate will be
created and distributed amongst all the mgr nodes.
It also allows us to get rid of the pyOpenSSL dependency.
Signed-off-by: Boris Ranto <branto@redhat.com>
The new rest API uses pecan for the restful functionality and simplifies
the code significantly. It should be mostly equivalent in functionality
to the django-based rest API. The api is self-documenting via /doc
endpoint.
Signed-off-by: Boris Ranto <branto@redhat.com>
This commit moves "BuildRequires: python-sphinx" down to the RH/CentOS/Fedora
distro conditional and adds a "BuildRequires: python-Sphinx" to the SUSE
conditional.
Signed-off-by: Jan Matejek <jmatejek@suse.com>
Signed-off-by: Nathan Cutler <ncutler@suse.com>
This commit moves the radosgw-admin command from
the radosgw package to ceph-common.
This allows for a better user expierence since the
radosgw-admin command can be run on any node.
Fixes: http://tracker.ceph.com/issues/19577
Signed-off-by: Ali Maredia <amaredia@redhat.com>
mgr/pybind/fsstatus already required this, but ceph.in does now.
Add as a build dependency as well to handle tests.
Signed-off-by: Dan Mick <dan.mick@redhat.com>
Creates an installable version of "src/include/rados/objclass.h" that allows
object classes to be built outside of the Ceph tree. cls_sdk is an example
of such an object class.
Signed-off-by: Neha Ojha <nojha@redhat.com>
ca40e12845 added the following lines to the spec
file:
%dir %{_libdir}/ceph/crypto
%{_libdir}/ceph/crypto/libceph_*.so*
and 350932979b made it so those build artifacts
are generated on x86_64 only.
The result is a build failure on non-x86_64 architectures:
error: Directory not found: /home/abuild/rpmbuild/BUILDROOT/ceph-12.0.2+git.1493119152.181baf6-1.1.ppc64le/usr/lib64/ceph/crypto
error: File not found by glob: /home/abuild/rpmbuild/BUILDROOT/ceph-12.0.2+git.1493119152.181baf6-1.1.ppc64le/usr/lib64/ceph/crypto/libceph_*.so*
Signed-off-by: Nathan Cutler <ncutler@suse.com>
ceph-create-keys unit file was removed here:
* 8bcb4646b6
* dc5fe8d415
As a consequence the systemctl preset command now fails to run since the
unit does not exist anymore. Due to the redirection in /dev/null we
don't know what's happening.
Ultimately the mon unit doesn't get enabled and the mon service won't
start after reboot.
Removing the old/non-existent unit makes the command succeed now.
Signed-off-by: Sébastien Han <seb@redhat.com>
