ceph-authtool 部署ceph-mds

admin

I'm trying to create ceph-mds manually on Centos 7 on Ceph Nautilus 14.2.19. first i created a folder inside /var/lib/ceph/mds in the format of <clusterid>-mds.<hostid> then ran the following commands:

9 Y! {6 g& ~) t, u( W0 Y7 Nceph-authtool --create-keyring /var/lib/ceph/mds/ceph-mds.<hostid>/keyring --gen-key -n mds.<hostid>
$ W2 u' t; @/ c% s0 kceph auth add mds.<hostid> osd "allow rwx" mds "allow *" mon "allow profile mds" -i /var/lib/ceph/mds/ceph-mds.<hostid>/keyring


Then i changed user permissions of contents of /var/lib/ceph/mds/ to ceph:ceph. When i check the keyring and the keyring added on ceph auth list  for mds i can confirm they are matching.
However when I run systemctl start ceph-mds@mds. the daemon doesnt start and on journalctl i get the following error output:
Apr 20 11:38:14 <hostid> ceph-mds[44742]: 2021-04-20 11:38:14.592 7f53bcaef700 -1 monclient(hunting): handle_auth_bad_method server allowed_methods [2] but i only support [2]
Apr 20 11:38:14 <hostid> ceph-mds[44742]: failed to fetch mon config (--no-mon-config to skip).
8 c+ p$ e4 l" S7 \: g
/ I$ E' f; j% y8 H
4 v2 u, o. D% G: z7 k( j, \: YThe entire output of systemctl status ceph-mds@mds.:
' q5 G+ `. J: M8 l1 F● ceph-mds@mds.<hostid>.service - Ceph metadata server daemon
   Loaded: loaded ([url=]/usr/lib/systemd/system/ceph-mds@.service[/url]; disabled; vendor preset: disabled)
" r6 D4 x% g7 S# t   Active: failed (Result: start-limit) since Tue 2021-04-20 12:28:15 +03; 11min ago
7 ~. R/ c3 w' ]7 Z/ q$ w3 d1 C  Process: 15564 ExecStart=/usr/bin/ceph-mds -f --cluster ${CLUSTER} --id %i --setuser ceph --setgroup ceph (code=exited, status=1/FAILURE)
0 L1 P- n6 u6 J( a Main PID: 15564 (code=exited, status=1/FAILURE)
Apr 20 12:28:15 <hostid> systemd[1]: Unit ceph-mds@mds.<hostid>.service entered failed state.
Apr 20 12:28:15 <hostid> systemd[1]: ceph-mds@mds.<hostid>.service failed.
Apr 20 12:28:15 <hostid> systemd[1]: ceph-mds@mds.<hostid>.service holdoff time over, scheduling restart.
Apr 20 12:28:15 <hostid> systemd[1]: Stopped Ceph metadata server daemon.
0 Z1 D# m& q# y5 ~# a! |$ aApr 20 12:28:15 <hostid> systemd[1]: start request repeated too quickly for ceph-mds@mds.<hostid>.service
Apr 20 12:28:15 <hostid> systemd[1]: Failed to start Ceph metadata server daemon.
Apr 20 12:28:15 <hostid> systemd[1]: Unit ceph-mds@mds.<hostid>.service entered failed state.
$ }+ x# V. _+ H  S& a8 m2 R: WApr 20 12:28:15 <hostid> systemd[1]: ceph-mds@mds.<hostid>.service failed.
$ z9 q$ A" l# _/ ~0 b
" w6 F: ~- ?& `* m8 C% a
What might be the reason for this?
- {9 v; j* M3 [* f3 A

admin

[root@ceph-2 ceph-2]# mkdir -p /var/lib/ceph/mds/ceph-mds.ceph-2/
[root@ceph-2 ceph-2]# ceph-authtool --create-keyring /var/lib/ceph/mds/ceph-mds.ceph-2/keyring --gen-key -n mds.ceph-2
7 p9 g: h; y- q  z: zcreating /var/lib/ceph/mds/ceph-mds.ceph-2/keyring

[root@ceph-2 ceph-2]# ceph auth add mds.ceph-2 osd "allow rwx" mds "allow *" mon "allow profile mds" -i /var/lib/ceph/mds/ceph-mds.ceph-2/keyring
' p. }: E) C' F- j5 |- u8 B. i( l" Y
[root@ceph-2 ceph-2]# ceph auth add mds.ceph-2 osd "allow rwx" mds "allow *" mon "allow profile mds" -i /var/lib/ceph/mds/ceph-mds.ceph-2/keyring
% l, @, C3 C' w' fadded key for mds.ceph-2
8 @4 A* e  K9 z0 x% v% H; `[root@ceph-2 ceph-2]#

7 s3 a% c, _+ o2 D8 G
: G' s6 t& g3 s* I[root@ceph-2 mds]# chown -R ceph:ceph ceph-mds.ceph-2/
4 |/ c6 c; \# D% [
) r% [4 v' [& B7 f! h8 @; G0 G
, o. R3 N. ~) \7 V/ T1 w# o( E2 KJul 26 14:19:36 ceph-2.novalocal ceph-mds[40265]: NOTE: a copy of the executable, or `objdump -rdS <executable>` is needed to interpret this.
) U: V9 X0 P, g2 S: UJul 26 14:19:36 ceph-2.novalocal systemd[1]: ceph-mds@ceph-2.service: main process exited, code=killed, status=6/ABRT
Jul 26 14:19:36 ceph-2.novalocal systemd[1]: Unit ceph-mds@ceph-2.service entered failed state.
Jul 26 14:19:36 ceph-2.novalocal systemd[1]: ceph-mds@ceph-2.service failed.
Jul 26 14:19:36 ceph-2.novalocal systemd[1]: ceph-mds@ceph-2.service holdoff time over, scheduling restart.
Jul 26 14:19:36 ceph-2.novalocal systemd[1]: Stopped Ceph metadata server daemon.
-- Subject: Unit ceph-mds@ceph-2.service has finished shutting down
! N: B; v, a4 h% F; t3 n-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
; B3 E  c6 Z7 j% T* Y--
2 d& B% d+ [6 o% b4 B6 n9 f-- Unit ceph-mds@ceph-2.service has finished shutting down.
' L$ C: p! ]& B( t- i2 ~4 eJul 26 14:19:36 ceph-2.novalocal systemd[1]: start request repeated too quickly for ceph-mds@ceph-2.service
& t+ T" {: E) h5 dJul 26 14:19:36 ceph-2.novalocal systemd[1]: Failed to start Ceph metadata server daemon.
-- Subject: Unit ceph-mds@ceph-2.service has failed
+ C5 m& b! t' S-- Defined-By: systemd
-- Support: http://lists.freedesktop.org/mailman/listinfo/systemd-devel
--
-- Unit ceph-mds@ceph-2.service has failed.
: U7 S3 w: p/ v2 k) W--
-- The result is failed.
Jul 26 14:19:36 ceph-2.novalocal systemd[1]: Unit ceph-mds@ceph-2.service entered failed state.
" S1 p+ ]- x5 ~$ a8 D8 dJul 26 14:19:36 ceph-2.novalocal systemd[1]: ceph-mds@ceph-2.service failed.