1COROSYNC-CFGTOOL(8) COROSYNC-CFGTOOL(8)
2
3
4
6 corosync-cfgtool - An administrative tool for corosync.
7
9 corosync-cfgtool [[-i IP_address] [-b] [-s] [-n] [-R] [-L] [-k nodeid]
10 [-a nodeid] [-h] [-H] [--force]
11
13 corosync-cfgtool A tool for displaying and configuring active parame‐
14 ters within corosync.
15
17 -i Finds only information about the specified interface IP address
18 or link id with -s.
19
20 -s Displays the status of the current links on this node for
21 UDP/UDPU, with extended status for KNET. After each link, the
22 nodes on that link are displayed in order with their status, for
23 example there are 3 nodes with KNET transport:
24
25 LINK ID 0
26 addr = 192.168.100.80
27 status:
28 nodeid 1: localhost
29 nodeid 2: connected
30 nodeid 3: connected
31
32 Please note that only one link is returned for a single node
33 cluster configuration, no matter how many links are configured.
34
35 -b Displays the brief status of the current links on this node when
36 used with "-s". If any interfaces are faulty, 1 is returned by
37 the binary. If all interfaces are active 0 is returned to the
38 shell. After each link, the nodes on that link are displayed in
39 order with their status encoded into a single digit, or charac‐
40 ters 'n', 'd' and '?' with special meaning. 1=link enabled,
41 2=link connected, So a 3 in a node position indicates that the
42 link is both enabled and connected. Status represented by char‐
43 acter 'n' is used for localhost link. Character '?' means that
44 Corosync was unable to get status of link from knet (log should
45 contain more information). Character 'd' shouldn't appear and it
46 means that Corosync was unable to configure a link and it is re‐
47 sult of some error which should have been logged.
48
49 The output will be:
50
51 LINK ID 0
52 addr = 192.168.100.80
53 status = n33
54
55 -n Displays the status of the current nodes in the system with
56 their link status(es).
57
58 Local node ID 1, transport knet
59 nodeid: 2 reachable onwire (min/max/cur): 0, 1, 1
60 LINK: 0 (192.168.1.101->192.168.1.102) enabled connected mtu: 1397
61 LINK: 1 (192.168.4.1->192.168.4.2) enabled mtu: 469
62 LINK: 2 (192.168.9.1->192.168.9.2) enabled mtu: 469
63
64 Only reachable nodes are displayed so "reachable" should always be
65 there.
66 ‘onwire’ versions are the knet on-wire versions that are supported/in
67 use (where appropriate).
68 IP addresses are the local and remote IP addresses (for UDP[U] only the
69 local IP address is shown)
70 enabled - means the link has been brought up
71 connected - means that the link is connected to the remote node
72 dynconnected - is not currently implemented
73 mtu - shows the size of data packets. Should be the link packet size
74 less a small amount for protocol overheads and encryption
75
76 -R Tell all instances of corosync in this cluster to reload
77 corosync.conf.
78
79 Running corosync-cfgtool -R where nodes are running different
80 versions of corosync (including minor versions) is unsupported
81 and may result in undefined behaviour.
82
83 -L Tell corosync to reopen all logging files. In contrast to other
84 subcommands, nothing is displayed on terminal if call is suc‐
85 cessful.
86
87 -k Kill a node identified by node id.
88
89 -a Display the IP address(es) of a node.
90
91 -h Print basic usage.
92
93 -H Shutdown corosync cleanly on this node. corosync-cfgtool -H
94 will request a shutdown from corosync, which means it will con‐
95 sult any interested daemons before shutting down and the shut‐
96 down maybe vetoed if a daemon regards the shutdown as inappro‐
97 priate. If --force is added to the command line then corosync
98 will shutdown regardless of the daemons' opinions on the matter.
99
100
102 corosync_overview(7),
103
105 Angus Salkeld
106
107 2020-06-02 COROSYNC-CFGTOOL(8)