From b2f51dc733ceac9cd30dc03a5e9eff5150f0bea1 Mon Sep 17 00:00:00 2001 From: Patrick Donnelly Date: Thu, 5 Apr 2018 20:06:29 -0700 Subject: [PATCH] doc: outline steps to take CephFS fs down Fixes: https://tracker.ceph.com/issues/23427 Signed-off-by: Patrick Donnelly (cherry picked from commit 9aa13bf0b0bf20b4fcf4949af60ebfff1e8fea0a) --- doc/cephfs/administration.rst | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/doc/cephfs/administration.rst b/doc/cephfs/administration.rst index e9d9195557d1..71a0e4075d10 100644 --- a/doc/cephfs/administration.rst +++ b/doc/cephfs/administration.rst @@ -74,6 +74,22 @@ appear to be eg. exabytes in size, causing load on the MDS as it tries to enumerate the objects during operations like stats or deletes. +Taking the cluster down +----------------------- + +Taking a CephFS cluster down is done by reducing the number of ranks to 1, +setting the cluster_down flag, and then failing the last rank. For example: + +:: + ceph fs set max_mds 1 + ceph mds deactivate :1 # rank 2 of 2 + ceph status # wait for rank 1 to finish stopping + ceph fs set cluster_down true + ceph mds fail :0 + +Setting the ``cluster_down`` flag prevents standbys from taking over the failed +rank. + Daemons ------- -- 2.47.3