Skip to content

Commit 2b085e8

Browse files
authored
Merge pull request ceph#61427 from zdover23/wip-doc-2025-01-17-cephfs-disaster-recovery-experts-2
doc/cephfs: edit disaster-recovery-experts (2 of x) Reviewed-by: Anthony D'Atri <[email protected]>
2 parents 893a9bd + 4f3a69e commit 2b085e8

File tree

1 file changed

+20
-16
lines changed

1 file changed

+20
-16
lines changed

doc/cephfs/disaster-recovery-experts.rst

Lines changed: 20 additions & 16 deletions
Original file line numberDiff line numberDiff line change
@@ -61,40 +61,44 @@ store state.
6161
Journal truncation
6262
------------------
6363

64-
If the journal is corrupt or MDSs cannot replay it for any reason, you can
65-
truncate it like so:
64+
Use a command of the following form to truncate any journal that is corrupt or
65+
that an MDS cannot replay:
6666

67-
::
67+
.. prompt:: bash #
6868

69-
cephfs-journal-tool [--rank=<fs_name>:{mds-rank|all}] journal reset --yes-i-really-really-mean-it
69+
cephfs-journal-tool [--rank=<fs_name>:{mds-rank|all}] journal reset --yes-i-really-really-mean-it
7070

71-
Specify the filesystem and the MDS rank using the ``--rank`` option when the file system has/had
72-
multiple active MDS.
71+
Specify the filesystem and the MDS rank using the ``--rank`` option when the
72+
file system has or had multiple active MDS.
7373

7474
.. warning::
7575

76-
Resetting the journal *will* lose metadata unless you have extracted
77-
it by other means such as ``recover_dentries``. It is likely to leave
78-
some orphaned objects in the data pool. It may result in re-allocation
79-
of already-written inodes, such that permissions rules could be violated.
76+
Resetting the journal *will* cause metadata to be lost unless you have
77+
extracted it by other means such as ``recover_dentries``. Resetting the
78+
journal is likely to leave orphaned objects in the data pool. Resetting
79+
the journal may result in the re-allocation of already-written inodes,
80+
which means that permissions rules could be violated.
8081

8182
MDS table wipes
8283
---------------
8384

8485
After the journal has been reset, it may no longer be consistent with respect
8586
to the contents of the MDS tables (InoTable, SessionMap, SnapServer).
8687

87-
To reset the SessionMap (erase all sessions), use:
88+
Use the following command to reset the SessionMap (this will erase all
89+
sessions):
8890

89-
::
91+
.. prompt:: bash #
9092

9193
cephfs-table-tool all reset session
9294

93-
This command acts on the tables of all 'in' MDS ranks. Replace 'all' with an MDS
94-
rank to operate on that rank only.
95+
This command acts on the tables of all MDS ranks that are ``in``. To operate
96+
only on a specified rank, replace ``all`` in the above command with an MDS
97+
rank.
9598

96-
The session table is the table most likely to need resetting, but if you know you
97-
also need to reset the other tables then replace 'session' with 'snap' or 'inode'.
99+
Of all tables, the session table is the table most likely to require a reset.
100+
If you know that you need also to reset the other tables, then replace
101+
``session`` with ``snap`` or ``inode``.
98102

99103
MDS map reset
100104
-------------

0 commit comments

Comments
 (0)