Loic Dachary 01d48a270a ceph-deploy: always zap disk before creating an osd ... The existing logic is to ceph-deploy osd create --zap-disk which will zap the data device before preparing it. However it will not zap the journal device (see http://tracker.ceph.com/issues/13291). If ceph-deploy osd create fails, a fall back will zap both the data device and the journal and try prepare again. This could work if the device preparation and activation was synchronous and catch all errors that could be caused by an unclean journal device. However, the activation is asynchronous and it is entirely possible for a device to be prepared successfully and fail to activate in the background. The data and journal device are always zapped before calling ceph-deploy osd create. The logic is simpler and the overhead is low. http://tracker.ceph.com/issues/13000 Fixes: #13000 Signed-off-by: Loic Dachary <loic@dachary.org>		2015-09-30 20:57:24 +02:00
..
buildpackages	buildpackages: user-data must be per os-type/os-version	2015-09-27 23:42:58 +02:00
cephfs	tasks/cephfs: mds allow	2015-09-28 12:38:04 -04:00
tests	rename test to tests, follows best practices for tests dirs	2015-03-02 11:20:35 -08:00
util	tasks: sudo ceph for cli	2015-09-11 12:15:01 -04:00
__init__.py
admin_socket.py
apache.conf.template	Adds mod_proxy_fcgi support and makes it the default setup for rgw	2015-06-23 15:49:27 -05:00
autotest.py
blktrace.py
boto.cfg.template
buildpackages.py	buildpackages: flock must --close	2015-09-26 15:23:03 +02:00
calamari_nosetests.py	calamari_nosetests: verify that calamari_branch exists	2015-07-07 11:03:18 -07:00
calamari_setup.py	calamari_setup: install "cli" utils on Calamari node	2015-08-12 14:52:37 -06:00
ceph_client.py
ceph_deploy.py	ceph-deploy: always zap disk before creating an osd	2015-09-30 20:57:24 +02:00
ceph_fuse.py	tasks/ceph_fuse.py: virtual machines need flexible mount timeout	2015-07-14 13:03:01 +02:00
ceph_manager.py	tasks/ceph_manager: fix logging on failed pool property	2015-09-30 09:24:38 -04:00
ceph_objectstore_tool.py
ceph.py	tasks/ceph: wait for MDS to be active when creating a cluster	2015-09-21 16:05:51 -07:00
cephfs_test_runner.py	tasks: generalise cephfs test runner	2015-04-14 14:13:39 +01:00
cifs_mount.py
cram.py
devstack.py	tasks: sudo ceph for cli	2015-09-11 12:15:01 -04:00
die_on_err.py
divergent_priors2.py	tasks/divergent_priors2: wait for osd to start before tell	2015-09-24 23:34:17 -04:00
divergent_priors.py	tasks/divergent_prior[2]: fix rados get syntax	2015-09-05 09:25:24 -04:00
dump_stuck.py
ec_lost_unfound.py
filestore_idempotent.py	git.ceph.com	2015-04-16 09:05:30 -07:00
kclient.py
locktest.py
logrotate.conf	ceph: support arbitrarily-named daemons in logrotate	2015-06-08 17:36:49 -07:00
lost_unfound.py	Fixes #11013, use time.sleep instead of manager.sleep which isn't there.	2015-03-04 10:09:21 -06:00
manypools.py
mds_creation_failure.py
mds_journal_migration.py
mds_scrub_checks.py
mds_thrash.py
metadata.yaml
mod_fastcgi.conf.template	Adds mod_proxy_fcgi support and makes it the default setup for rgw	2015-06-23 15:49:27 -05:00
mod_proxy_fcgi.tcp.conf.template	Adds mod_proxy_fcgi support and makes it the default setup for rgw	2015-06-23 15:49:27 -05:00
mod_proxy_fcgi.uds.conf.template	Adds mod_proxy_fcgi support and makes it the default setup for rgw	2015-06-23 15:49:27 -05:00
mon_clock_skew_check.py
mon_recovery.py
mon_thrash.py
multibench.py	fix indentation error in docstring	2015-08-06 08:41:17 -04:00
object_source_down.py
omapbench.py
osd_backfill.py
osd_failsafe_enospc.py	tasks: sudo ceph for cli	2015-09-11 12:15:01 -04:00
osd_recovery.py
peer.py
peering_speed_test.py	remove unused imports from peering_speed	2015-02-26 14:52:40 -05:00
populate_rbd_pool.py	remove unused imports from populate_rbd	2015-02-26 14:52:58 -05:00
qemu.py	tasks/qemu: avoid adding duplicated entries to /etc/exports	2015-08-31 22:27:38 +08:00
rados.py	tasks/rados: make {write,append}_excl conditional	2015-07-08 10:43:22 -04:00
radosbench.py	tasks/radosbench: allow cleanup: false	2015-09-11 16:52:41 -04:00
radosgw_admin_rest.py	ensure summary is looked for the user we need (part 2)	2015-03-24 13:37:41 -04:00
radosgw_admin.py	ensure summary is looked for the user we need (part 2)	2015-03-24 13:37:41 -04:00
radosgw_agent.py	git.ceph.com	2015-04-16 08:37:56 -07:00
rbd_fsx.py	tasks: add support for running fsx under valgrind	2015-04-28 09:44:05 -04:00
rbd.py	rbd.xfstests: remove bad assert	2015-04-27 11:29:19 -07:00
recovery_bench.py
reg11184.py	tasks/reg11184.py: fix rados get syntax	2015-09-05 09:23:54 -04:00
rep_lost_unfound_delete.py
repair_test.py	remove unused import from repair_test	2015-02-26 15:04:02 -05:00
rest_api.py	tasks: sudo ceph for cli	2015-09-11 12:15:01 -04:00
restart.py	git.ceph.com	2015-04-16 08:37:56 -07:00
rgw_logsocket.py
rgw.py	tasks: sudo ceph for cli	2015-09-11 12:15:01 -04:00
s3readwrite.py
s3roundtrip.py
s3tests.py	Use old branches as is without 'ceph-' for firefly, hammer etc	2015-09-16 13:21:51 -04:00
samba.py	task/samba: ignore return code of fuser/losf	2015-04-21 11:33:30 +08:00
scrub_test.py	tasks/scrub_test: sudo ls ...	2015-09-02 22:01:58 -04:00
scrub.py
tgt.py
thrash_pool_snaps.py	rados/thrash: add test for radosgw with snaps	2015-05-28 15:36:39 -07:00
thrashosds.py	thrashosds: adds a sighup_delay option, defaulted to 0.1	2015-07-28 14:46:35 -05:00
userdata_setup.yaml	Use an NFS mount instead of virtio-9p in the qemu task	2015-03-02 15:45:10 -08:00
userdata_teardown.yaml
watch_notify_same_primary.py	tasks/watch_notify_same_primary: wait for watch before notify	2015-03-26 16:51:56 -07:00
watch_notify_stress.py
workunit.py	git.ceph.com	2015-04-16 08:37:56 -07:00