This is necessary as iscsi daemons expect the iscsi-gateway.cfg
which includes this list to match across the iscsi daemons
Fixes: https://tracker.ceph.com/issues/57078
Signed-off-by: Adam King <adking@redhat.com>
mgr/cephadm: don't use "sudo" in commands if user is root
Reviewed-by: John Mulligan <jmulligan@redhat.com>
Reviewed-by: Redouane Kachach <rkachach@redhat.com>
Previously, due to the way it was just splitting the file name on
dashes to try and get the profile name, any profile name with dashes
was not getting properly matched and would therefore get marked
stray and removed. This new strategy instead tries to match the actual
expected file name.
Fixes: https://tracker.ceph.com/issues/57405
Signed-off-by: Adam King <adking@redhat.com>
mgr/cephadm: Adding --storage.tsdb.retention.size prometheus option
Reviewed-by: Adam King <adking@redhat.com>
Reviewed-by: Anthony D'Atri <anthonyeleven@users.noreply.github.com>
doc/monitoring: add min vers of apps in mon stack
Reviewed-by: Anthony D'Atri <anthony.datri@gmail.com>
Reviewed-by: Ernesto Puerta <epuertat@redhat.com>
This was accidentally changed to "nfs"--which breaks various
ceph orchestration assumptions.
Found by Ernesto Puerta <epuertat@redhat.com>
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
Mounting files with extra container args seems to not
work with some of the more intuitive syntaxes.
See https://tracker.ceph.com/issues/57338
Signed-off-by: Adam King <adking@redhat.com>
mgr/dashboard: fix snapshot creation with duplicate name
Reviewed-by: Ernesto Puerta <epuertat@redhat.com>
Reviewed-by: Nizamudeen A <nia@redhat.com>
Reviewed-by: Pere Diaz Bou <pdiazbou@redhat.com>
When generating tags the order of endpoints wasn't taken into account.
Two endpoints with the same url prefix, for example `/api/cluster/` and
`/api/cluster/user`, have different docs and the tags is generated from
a doc of one of these two, and since the order of these endpoints might
vary it is imperative to sort them to have a deterministic output.
Signed-off-by: Pere Diaz Bou <pdiazbou@redhat.com>
AppMain::shutdown() had borrowed the unregister_async_signal_handler()
text from librgw, but of course, that's wrong.
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
Don't rely on the conf-default value of rgw_nfs_frontends to
provide "rgw-nfs," as that's not optional when running under
librgw.
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
Reorganize some dependencies shared by the librgw.so and radosgw
main units. The librgw.so build no longer builds a main(...)
unit.
Build time should be slightly reduced, as several shared dependencies
have been moved into rgw_a.
There are some other cleanups, e.g., rgw_signal.{h,cc}.
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
s/rgw::InitHelper/rgw::AppMain/;
Move program state from RGWLib into AppMain, preparing it to
manage program state for all RGW instances (e.g., main()).
Signed-off-by: Matt Benjamin <mbenjamin@redhat.com>
If the hostname is selected for the primary devices, then we can
preselect the hostname filter for the db/wal devices because osds will
be deployed only on the hostname of the primary device. If preselected
it'll be clear that only this devices will be used to deploy.
Addition to this, usually ssd devices are used for db/wal devices. So I
am preselecting these too in the filters.
Fixes: https://tracker.ceph.com/issues/57118
Signed-off-by: Nizamudeen A <nia@redhat.com>
