suites/kcephfs: put clients on dedicated nodes

Some tests want to use power cycling to reset stuck
mounts, and that needs to not kill OSDs as collateral
damage.

Need to revisit this to avoid unnecessarily using a whole
node for the client for those tests that don't require it
(i.e. those that don't use CephFSTestCase.REQUIRE_KCLIENT_REMOTE)

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

diff --git a/suites/kcephfs/recovery/clusters/4-remote-clients.yaml b/suites/kcephfs/recovery/clusters/4-remote-clients.yaml
index 702bb161e75b5b022278b1e9a6ce1ff05748e6d1..ef0cfd59b539c0f5426ecae73dd8a8c5c242c51c 100644 (file)
--- a/suites/kcephfs/recovery/clusters/4-remote-clients.yaml
+++ b/suites/kcephfs/recovery/clusters/4-remote-clients.yaml
@@ -1,6 +1,8 @@
 roles:
-- [mon.a, osd.0, mds.a, mds.b, client.1, client.2, client.3]
-- [client.0, osd.1, osd.2]
+- [mon.a, osd.0, mds.a, client.2]
+- [osd.1, osd.2, mds.b, client.3]
+- [client.0]
+- [client.1]
 log-rotate:
   ceph-mds: 10G
   ceph-osd: 10G