Revert "packages.yaml: reflect python-ceph package split"

This reverts commit ec4b0c3d6b057e688363d966a83a037c3585b32f.

References: http://tracker.ceph.com/issues/17075
Signed-off-by: Nathan Cutler <ncutler@suse.com>

Merge pull request #1120 from ceph/wip-rgw-jewel-17025

rgw: data_extra_pool is unique per zone

Reviewed-by: Casey Bodley <cbodley@redhat.com>

rgw: data_extra_pool is unique per zone

Fixes: http://tracker.ceph.com/issues/17025
Signed-off-by: Orit Wasserman <owasserm@redhat.com>
(cherry picked from commit c6cdd0905fbb3a9571400a5227b34b7e2e738830)

Merge pull request #1095 from smithfarm/wip-python-split-jewel

packages.yaml: reflect python-ceph package split

Merge pull request #1101 from SUSE/wip-osd-recovery-jewel

osd_recovery: add delay to ensure maps propogate

Reviewed-by: Samuel Just <sjust@redhat.com>

Merge pull request #1100 from ajarr/wip-manila-backports-jewel

jewel: backports for OpenStack Manila

Reviewed-by: John Spray <john.spray@redhat.com>

Merge pull request #1103 from ceph/wip-fix-bad-check-jewel

backport - Wip fix bad check

tox.ini: exclude .tox dir from flake8

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 50f32756a472c29cc6c18a128e5b6cc99c9371e5)

Fix rechecking of health in loop

Use the newer stdout value for checking
ceph health again.

Signed-off-by: Vasu Kulkarni <vasu@redhat.com>

osd_recovery: add delay to ensure maps propogate

Signed-off-by: Samuel Just <sjust@redhat.com>
(cherry picked from commit ddabdb7f722b192435b87bc31fe587b93cc53924)

tasks/cephfs/test_volume_client: test authentication metadata

Test the usecases for the authentication metadata stored
by the volume client:

* Obtain the list of auth IDs having access to a volume.

* Restrict volume access to auth IDs of a single (OpenStack)
  tenant to enforce strong tenant isolation of volumes.

Signed-off-by: Ramana Raja <rraja@redhat.com>
(cherry picked from commit 08e3a49fa52d49a16585493d05cf51c2b31edece)

tasks/cephfs: test read-only authorization for volumes

Test whether mount access to volumes can be restricted to read-only.

Fixes: http://tracker.ceph.com/issues/15614
Signed-off-by: Ramana Raja <rraja@redhat.com>
(cherry picked from commit aab9a03c4c1d5ef98033b654e1a11da7ddf8aa9a)

tasks/cephfs: add function to configure guest credentials

In the test_volume_client module, refactor code to add a function
that configures credentials for a guest to mount a volume. The
function could be used by tests to setup a guest client to mount
a volume with fewer lines of code.

Signed-off-by: Ramana Raja <rraja@redhat.com>
(cherry picked from commit 02fc6d6acc1246881c173d693773658b210f1ad3)

packages.yaml: reflect python-ceph package split

Signed-off-by: Ricardo Dias <rdias@suse.com>
(cherry picked from commit 0216330568fc1dec2e866bbfd1e49dc42266e1ab)

Merge pull request #1088 from ceph/jewel-mds-snap-failover

fs: add snapshot tests to mds thrashing

Reviewed-by: John Spray <john.spray@redhat.com>

fs: add snapshot tests to mds thrashing

Signed-off-by: Yan, Zheng <zyan@redhat.com>
(cherry picked from commit b9b18c7b4ec17f1461b335e0c4ef3f36be70f74e)

Merge pull request #1071 from ceph/wip-16551-formatted-output

rbd: pull formatted-output.t from the relevant branch

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

rbd: pull formatted-output.t from the relevant branch

Fixes http://tracker.ceph.com/issues/16551

Signed-off-by: Loic Dachary <ldachary@redhat.com>

Merge pull request #1070 from smithfarm/wip-drop-ceph-devel-jewel

jewel: packages.yaml: drop ceph-devel

Reviewed-by: Loic Dachary <ldachary@redhat.com>

packages.yaml: drop ceph-devel

There are no "-dev" packages listed in the deb section of this file, so it
serves no purpose to have ceph-devel here.

Also, once https://github.com/ceph/ceph/pull/9744 is merged the presence of
"ceph-devel" in this file will break teuthology for rpm targets.

Signed-off-by: Nathan Cutler <ncutler@suse.com>
(cherry picked from commit af9e05709af76c931e4dd50349cac3a4fc87a575)

Merge pull request #1042 from ceph/wip-li-rados-jewel

Jewel: Backport changes needed for list-inconsistent-* format changes

Reviewed-by: Loic Dachary <ldachary@redhat.com>

Merge pull request #1056 from ceph/wip-rbd-nbd-jewel

jewel: rbd: added rbd-nbd fsx test case

Reviewed-by: Josh Durgin <jdurgin@redhat.com>

rbd: added rbd-nbd fsx test case

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit 7425e2d7a64e2962f6a7bc6f6ba09ee9137e7a14)

Merge pull request #1015 from ceph/wip-jewel-update-standby

(jewel) tasks/cephfs: update failover test for standby changes

Merge pull request #1016 from ceph/wip-jewel-multicluster-fixes

(jewel) backport cephfs fixes for test broken by multi-cluster support

Merge pull request #1047 from ceph/wip-16113-jewel

suites/rados: add test for 16113

Initial check in for a complete jewel-x suite

Fixes http://tracker.ceph.com/issues/15754
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
Signed-off-by: Yuri Weinstein <yweinste@redhat.com>
(cherry picked from commit 6ad42ef5796baed50c107877133128be4ae267f3)

scrub_test: Shards are not marked in error when osd guesses

Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 68400e95524181f6bdd1a250b72b1cb38f543aea)

scrub_test: Handle list-inconsistent-obj changes in pull #8983

Add checking of error in the shards

Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 5992384700f335a970c1f7cb392d5b71cecfda5c)

Merge pull request #1035 from ceph/wip-sleep-jewel

[jewel]: tasks: fix non-existent sleep function

Reviewed-by: Josh Durgin <jdurgin@redhat.com>

suites/rados: add test for 16113

Signed-off-by: Samuel Just <sjust@redhat.com>

tasks: fix non-existent sleep function

CephManager has no sleep function. Use time.sleep() instead.

Ran into this while testing a branch. Apparently it doesn't happen
much since this hasn't changed in years, but the error was copied
into several tasks.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit cfc48e36da74cd8b840270e388cc4b7e3efc1b58)

Merge pull request #1030 from ceph/wip-manila-backports

jewel: backports needed for Manila

Reviewed-by: John Spray <john.spray@redhat.com>

fs/volume_client: exercise the configurable prefix and ns_prefix.

Also ensure namespace_prefix and vol_prefix are set to default values
if they are not explicitly passed.

Fixes #15417

Signed-off-by: Xiaoxi Chen <xiaoxchen@ebay.com>
(cherry picked from commit 42f63dc6a6bfbe7667ad2293946fedeb097fabce)

tasks/cephfs: add TestVolumeClient.test_purge

For reproducing error handling paths during purge,
such as #15266 (ASCII exception)

Fixes: http://tracker.ceph.com/issues/15266
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 798c50e9bcd6c62b7e1687c85db718bca6883b3e)

tasks/cephfs: test 'df' output in volumeclient

Fixes: http://tracker.ceph.com/issues/15599
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 4e9635e8558aa69e66ec79b53469b03cd3eb12e6)

tasks/cephfs: make mount point more configurable

The mount points of the test clients were created based on the auth
ID they used. So two clients using the same auth ID could not create
different mount points in a straight forward way. Expose the
directory name of the mount point as a public attribute allowing
client mount points to be easily changed whenever needed.

Signed-off-by: Ramana Raja <rraja@redhat.com>
(cherry picked from commit 64869a3e4d5722e84f791ee3a07bf445d4402d32)
(cherry picked from commit dcdccfcd57aa6061620c8796c7cd90829cae03aa)

tasks/cephfs: test volume client eviction

Test whether the CephFSVolumeClient can evict a client based on its
auth ID and the volume path it has mounted.

Fixes: http://tracker.ceph.com/issues/15855
Signed-off-by: Ramana Raja <rraja@redhat.com>
(cherry picked from commit cbadb384f2e6f6fa2ebeab0d60af591a6de252a4)
(cherry picked from commit fab28a97e487ead9cc837ef73bb9798c1b24d99c)

suites: allow four remote clients for fs/recovery

test_volume_client requires four clients.

Signed-off-by: Ramana Raja <rraja@redhat.com>
(cherry picked from commit b970f97dbc5af6e331282c4eb61c9faaff1e586e)
(cherry picked from commit f539587df992fc1b77196a43c8a54fab345fc0c9)

Merge pull request #1029 from ceph/wip-rbd-openstack

rbd: provision volumes to format as XFS

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

rbd: provision volumes to format as XFS

Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from commit 45bd6cbf57c6772c47906f13eb27bf8586067213)

Merge pull request #1027 from ceph/wip-ctx-manager-jewel

tasks/create_verify_lfn_objects: adjust to new ctx.manager location

Reviewed-by: Josh Durgin <jdurgin@redhat.com>

tasks/create_verify_lfn_objects: adjust to new ctx.manager location

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 1cb2681cd02d7782c9dea602c9cf2b0fb80e6fae)

Merge pull request #1025 from ceph/wip-rbd-mirror-stress-jewel

jewel: suites/rbd: added replication stress test workunit

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

suites/rbd: added replication stress test workunit

Signed-off-by: Jason Dillaman <dillaman@redhat.com>
(cherry picked from commit ecbda33b52c1a139cdf31562ac81a433b7e1dc68)

Merge pull request #1021 from smithfarm/wip-buildpackages-jewel

openstack: ovh renamed flavors : s/eg/hg/

Reviewed-by: Loic Dachary <ldachary@redhat.com>

openstack: ovh renamed flavors : s/eg/hg/

http://travaux.ovh.net/?do=details&id=17532

Signed-off-by: Loic Dachary <loic@dachary.org>
(cherry picked from commit 4d47b2d1b0fae2efcc331fbf2a6f8c576bb4aad3)

tasks/cephfs: find processes correctly

Don't mistake the daemon-helper line for the actual
daemon process.

Fixes: http://tracker.ceph.com/issues/15900
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 66901ccc04077c86dbb713ddb4c1ab6de075f482)

tasks/cephfs: fix mount wait in test_mount_conn_close

Fixes: http://tracker.ceph.com/issues/15900
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 76809b1af562558dc5229be4b5d63fa2d043c4ae)

tasks/cephfs: fix ps usage to not truncate cmds

Now that daemon commandlines include "--cluster" they
easily overflow ps's default behaviour.

Fixes: http://tracker.ceph.com/issues/15857
Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit 8e6a164dfb957a838482207a911289ffab0936f2)

tasks/cephfs: update failover test for standby changes

Previously, an MDS could be used as a replay even
if it had "standby replay = false", this is changed
now so that replays will only happen explicitly.

Signed-off-by: John Spray <john.spray@redhat.com>
(cherry picked from commit d9309b57e3c12579f64939ee78161f7e99a062c8)

Merge pull request #1011 from ceph/wip-jewel-multi-cluster

jewel: multi-cluster support and rbd mirroring tests

Reviewed-by: Jason Dillaman <dillaman@redhat.com>

suites/teuthology/multi-cluster: make them pass again

Use xfs to avoid object length limits and be explicit about which
clients to run on, so the workunit git clone doesn't race with another
client on the same machine.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit ed9ee7a4c8c0cce357ed0c2fd7a0f0368210f3cc)

tasks: fix {testdir}/data paths

The multi-cluster branch updated these to have a '$cluster.' prefix

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 26c7dab57ef7a6bd8377441bb1063a38e096002d)

tasks/ceph.healthy: allow None as config

This is used by the pg-removal-interruption test.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit a67a123f0c0c3b9e33021ebd94a454961e4f46dd)

suites: add multi-cluster tests to the teuthology suite

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit bdd8ffddec219414977240c5692a87a820df6ab6)

tasks: update ctx.ceph.conf readers to use per-cluster conf

Just use the default cluster name for now, since these tasks aren't used
for multi-cluster tests yet.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit a9872182d362ded03a3f23679a389e686806be5d)

suites/rbd: add basic mirroring suite

Just one workunit to run so far.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 67d7d997c213cfb554b5b1dda4acef9fac309ff8)

tasks: add an rbd-mirror task

This simply runs the rbd-mirror daemon. Nothing special is needed
since the runtime files (logs, asok, etc) go to the standard ceph
locations.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit ed33f86589ff28a1fff62157de2a82a28ca9699e)

tasks/ceph: ignore EEXIST for the archive data dir creation

With multiple clusters this will be called multiple times.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit bccdef665070f043082e33665fd17739bcbadd62)

tasks/ceph: pull each mon dir only once

No need to pull all mon dirs for the host for each mon on the host.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 4c2e7309db40bb6f87a9f2930371522164513bb4)

tasks/ceph: only run ceph_log and valgrind_post once

These setup and parse logs on all hosts, so they should be run only
for the first cluster setup. This cluster will be torn down last, so
the cleanup happens after all clusters are shutdown as well.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 3203b76792e6ab7f35d465cebe602243681d5d9e)

tasks/ceph: make scrubbing cluster-aware

Simplify implementation by using manager and teuthology.misc methods
instead of reinventing them here.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 96e7724e2360f86bf22e5a51de19659b74db9448)

tasks/watch_notify_same_primary: adjust to new ctx.manager location

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 34a58dda5606f565237c095577caa6133a2c01f1)

tasks/thrash_pool_snaps: adjust to new ctx.manager location

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 9f5b0370dd07c571c58653d03a2eabe53b2e4793)

tasks/repair_test: clean up manager usage, adjust to new location

Only repair_test_2 actually uses ctx, the rest of the functions only
need the manager.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit a4267dba2c39819da759b971b3ba65830ce10647)

repair_test: Disable scheduled scrubbing to not mess up do_pg_scrub()

Fixes: http://tracker.ceph.com/issues/15679
Signed-off-by: David Zafman <dzafman@redhat.com>
(cherry picked from commit 5c0edbae3eb35654b4efb2e6eb097014953aea12)

tasks/reg11184: adjust to new ctx.manager location

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 6f465efa041a32caf4f121b946c8a999d90a7069)

tasks/radosbench: adjust to new ctx.manager location

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 55fe7de7be6a243325679e8c74bc68f39b59cf94)

tasks/rados: adjust to new ctx.manager location

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 39578a6a357e3379cbaedeb5301f3eb2ffe4bc7f)

tasks/populate_rbd_pool: adjust to new ctx.manager location

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 563c0e99b9b2cda07110f1724004c69eceea3752)

tasks/peering_speed_test: adust for new ctx.manager location

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 691af92f674b4be5fa657830e4542be4b2ce50e0)

tasks/osd_failsafe_enospc: adjust for new ctx.manager location

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 645a0f5d20fd7a4297ecf5def779778c08f00fcb)

tasks/divergent_priors*: adjust to new ctx.manager location

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 70db2726c577a5b6a1cde2be30189bf8bf065c5b)

tasks/ceph_objectstore_tool: use existing manager

This is only used after the ceph task, which already createds a
manager for the cluster. Use the default 'ceph' cluster manager for
now.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit e00437ae897741a10545ab1bcc32f94b530831c4)

tasks/ceph_manager: make utility_task cluster-aware

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 274d79ade39bddfcc13d3b9cb6829bac8b693c1f)

tasks/ceph: make restart subtask cluster-aware

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 3948f108a847744de4197333ccb50f8a3c6fada5)

tasks/ceph: make wait_for_mon_quorum cluster-aware

Accept a 'daemons' list like other ceph subtasks, so it can get an
optional 'cluster' setting too.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 9ad65769c77586332d81c9a7414b2f8d90e391b7)

tasks/ceph: make wait_for_osds_up cluster-aware

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit bb76deaf5784dfaf3a7ea0a7bb4a601f89d681d4)

tasks/ceph: update ctx.manager usage to ctx.managers

Not sure this function is ever used (no users in ceph-qa-suite yamls
or tasks.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit ef619062be841bb93dd6747b31fbb55ee8e1cd23)

tasks/workunit: work with roles that include cluster

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 1b2087231d36bc05db247cccbe2d217cde771fad)

tasks: move find_remote to util, rename and add helper

This is a useful for any cluster-aware task.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 256ebf8a12815f8650b2b2b7517190145bd02ea7)

tasks/blktrace: make cluster-aware

Need to adjust to lookup devices by role rather than id, so might as
well add a cluster config option while we're at it. Remove the
confusing default config that is not actually used.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit a3096e8600a4933d3420f9e498d0da7707eb686d)

tasks/ceph_manager: make Thrasher cluster-aware

Just a few spots need to know to lookup only osds in this cluster, or
prefix a filename with the cluster. Use CephManager.find_remote() to
avoid a bunch of repetition and look only in the intended cluster.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit bba323834be5ace438042ce6fa370a6cd80c37d3)

tasks/ceph_manager: make mount_osd_data() cluster-aware

Use a cluster-specific mount point, and address osds by full role,
rather than just id, in the ctx.disk_config structures.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 713e717fda482cbe6396ee8af65ac7d81132852a)

tasks/ceph_manager: add cluster param to write_conf()

Only used by cephfs right now, so don't bother changing callers.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 524e6d7a5e29ff325edd2673e3774fad757bb83e)

tasks/ceph_manager: simplify remote lookup, and make it cluster aware

Re-implement find_remote() using ctx.cluster.only() with a matcher
function that includes the manager's cluster, and use it instead of
miscellaneous ctx.cluster.only() calls elsewhere.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit ff49deb6f0beb99b0115d606cc1bc4f4bf293f5c)

tasks/ceph_manager: parameterize CephManager with cluster

Add --cluster arguments, pass cluster to get_daemon() and
iter_daemons_of_role, replace 'ceph' with cluster in paths, and use
ctx.ceph[cluster] instead of ctx.ceph.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 141c73d3998c9edb45a15d86f1f9029b4ea3a41b)

tasks/thrashosds: adjust to per-cluster managers

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 05bddf90766aeeabaf4e2274c66eaa8b4db425c9)

tasks/thrashosds: add an option to specify the cluster to thrash

Thrashing must be aware of the cluster it is using, since it maintains
some invariants (e.g. enough osds up when waiting to recover).

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit b88d595571f4415b17e50afa4dc47cbe6ae87bc4)

tasks/ceph: store cluster config in a per-cluster dict

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 0acbafe3719e773a670aece2ba578f72100a4e31)

tasks/ceph: create a CephManager per cluster

Thrashings tasks will be updated to use ctx.managers indexed by
cluster later.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 25ff14af7421bfbb91c5e60f51fe6a6f924ec30e)

tasks/ceph: make healthy() cluster-aware

ceph.healthy may be used as a standalone task, so it may not always
have the cluster name in its configuration.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit d1354703051bb5fa27ecc1f0e4f18de21c80e6b4)

tasks/ceph: make cephfs_setup() cluster-aware

Note that cephfs tests using the Filesystem abstractions will need to
be converted to understand multiple clusters later. This just updates
the ceph task portion.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit b0dd04736eed2822d1584eb8f6f7839e7f02ffc9)

tasks/ceph: make crush_setup() cluster-aware

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 44097101025a3053bb7938062ee527a76cedc29f)

tasks/ceph: adapt to cluster-aware daemon.resolve_role_list

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit e19e9e2fa31bd544894b3e36ade7c09ed8af73d6)

tasks/ceph: make run_daemon() cluster-aware

Pass --cluster where appropriate and include the full role in file
names.

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 26b8a1f3acdc53122767918cec86d36fdf74337c)

tasks/ceph: convert cluster creation to work with multiple clusters

Add a cluster option to the ceph task, and pass that through to
cluster(). Make sure monitors and clients don't collide by adding
their cluster to paths they use.

This assumes there is one ceph task per cluster, and osds from
multiple clusters do not share hosts (or else the block device
assignment won't work).

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit a54ff597faa379c8fd369352fb924651be269f95)

tasks/ceph_client: add cluster param

Signed-off-by: Josh Durgin <jdurgin@redhat.com>
(cherry picked from commit 466678b980e0527947f5c436d83da6933b216a25)

Merge pull request #1000 from smithfarm/wip-opensuse-make-rpm-jewel

make-rpm.sh: support openSUSE

Reviewed-by: Loic Dachary <ldachary@redhat.com>