* refs/pull/28831/head:
client: support the fallocate() when fuse version >= 2.9
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Reviewed-by: Zheng Yan <zyan@redhat.com>
This avoids the need to acquire the mds_lock prior to responding to config
changes. The MDSRank can create a finisher context that acquires the lock
later.
Fixes: https://tracker.ceph.com/issues/40694
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
Include hardware details and update language for modern tools.
Fixes: http://tracker.ceph.com/issues/39620
Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
* refs/pull/28978/head:
qa/tasks/mgr/dashboard/test_health: fix test_full_health test
mon: use per-pool stats only when all OSDs are reporting
osd: report whether we have per-pool stats
osd/osd_types: osd_stat_t: include num_per_pool_osds member
Reviewed-by: Igor Fedotov <ifedotov@suse.com>
mgr/dashboard: controllers/grafana is not Python3 compatible
Reviewed-by: Sebastian Wagner <swagner@suse.com>
Reviewed-by: Volker Theile <vtheile@suse.com>
Some callers (ceph-objectstore-tool) instantiate SnapMapper with a bogus
pool id (0) and want to use it to adjust mappings. We have the hobj
available, so use its pool member instead.
Signed-off-by: Sage Weil <sage@redhat.com>
* refs/pull/28945/head:
os/bluestore: StupidAllocator stucks on 4GB allocations
os/bluestore: avoid length overflow in extents returned by Stupid
Reviewed-by: Sage Weil <sage@redhat.com>
* refs/pull/28990/head:
qa: vstart_runner fails because of string index out of range
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
Reviewed-by: Laura Paduano <lpaduano@suse.com>
mgr/dashboard: Fix form validation problems in RGW user form
Reviewed-by: Tatjana Dehler <tdehler@suse.com>
Reviewed-by: Volker Theile <vtheile@suse.com>
RGW basic implementation
- Add one or more hosts running RGW daemon
- Allow to specify a multisite/multizone settings using a config file.
- Remove RGW daemon used to provide service for one zone.
Comments:
I see two use cases:
a. Install/Remove in new hosts the RGW daemons and configure them.
b. RGW management (CRUD zone groups/zones ... etc)
But the current orchestrators API seems to be oriented only to cover use case b.
What i have implemented in this moment covers use case a, despite the API is not friendly with that. I have added the possibility of getting a config file to specify hosts and settings.
rgw-admin commands covers all the functionality for managing RGW... So probably it does not have too much sense to implement anything around RGW management in orchestrators (use case b).
My doubts are:
1. The current API covers RGW management using a default configuration (user cannot configure multisite/multizone). Is this ok?
2. Shoud we add more API endpoints to allow add RGW nodes/hosts and manage the RGW config?
---------------------------------
Modifications suggested by @epuerta:
Mainly:
- New Inventory group class for manage hosts in Ansible Roles inside Ansible Inventory.
- A redundant plus sign was found by @epuerta
Signed-off-by: Juan Miguel Olmo Martínez <jolmomar@redhat.com>
Some labels were not bold, because the css rule didn't apply to them.
I have improved the css rule and also add missing classes in some labels.
Signed-off-by: Tiago Melo <tmelo@suse.com>
Build shows:
[ 53%] Building CXX object src/mds/CMakeFiles/mds.dir/JournalPointer.cc.o
ceph/src/rgw/rgw_rest_s3.cc: In member function ‘RGWOp* RGWHandler_REST_Bucket_S3::get_obj_op(bool)’:
ceph/src/rgw/rgw_rest_s3.cc:3588:5: warning: control reaches end of non-void function [-Wreturn-type]
} }
^
Fixes: 4ffc765c4c
Fixes: http://tracker.ceph.com/issues/40747
Signed-off-by: Jos Collin <jcollin@redhat.com>
The previous implementation didn't work properly when there was more
than once service on a node (e.g. mgr and mon on the same node).
Signed-off-by: Tim Serong <tserong@suse.com>
mgr/mgr_module: Added persistent dict
basically a pythonic interface for a k-v store
mgr/orchestrator: Added common code to implement a cache
mgr/ssh: The cache works by manually adding and removing hosts.
only the data is invalidated
mgr/deepsea: The cache invalidates all at once and only
valid objects are cache
Fixes https://tracker.ceph.com/issues/39990
Signed-off-by: Sebastian Wagner <sebastian.wagner@suse.com>
* Updated troubleshooting.rst as kernel-client-troubleshooting.rst
is dropped.
* Added troubleshooting references in fuse.rst and kernel.rst
Signed-off-by: Jos Collin <jcollin@redhat.com>
* refs/pull/28643/head:
doc: removed old EOL Ceph versions from the list
doc: added openSUSE distros to the list and updated the releases
doc: updated the OS recommendations to reflect the Nautilus release
Reviewed-by: Lenz Grimmer <lgrimmer@suse.com>
Reviewed-by: Patrick Donnelly <pdonnell@redhat.com>
