1CLVMD(8) System Manager's Manual CLVMD(8)
2
6 clvmd - cluster LVM daemon
7
9 clvmd [-d[<value>]] [-C] [-h] [-R] [-S] [-t <timeout>] [-T <start time‐
10 out>] [-V]
11
13 clvmd is the daemon that distributes LVM metadata updates around a
14 cluster. It must be running on all nodes in the cluster and will give
15 an error if a node in the cluster does not have this daemon running.
16
18 -d[<value>]
19 Enable debug logging. Value can be 0, 1 or 2.
20 0 disables debug logging
21 1 sends debug logs to stderr (implies -f option)
22 2 sends debug logs to syslog
23 If -d is specified without a value then 1 is assumed.
24 -C Only valid if -d is also specified. Tells all clvmds in a clus‐
26 ter to enable/disable debug logging. Without this switch, only
27 the local clvmd will change its debug level to that given with
28 -d.
29 This does not work correctly if specified on the command-line
30 that starts clvmd. If you want to start clvmd and enable clus‐
31 ter-wide logging then the command needs to be issued twice, eg:
32 clvmd
33 clvmd -d2
34 -t <timeout>
36 Specifies the timeout for commands to run around the cluster.
37 This should not be so small that commands with many disk updates
38 to do will fail, so you may need to increase this on systems
39 with very large disk farms. The default is 30 seconds.
40 -T <start timeout>
42 Specifies the timeout for clvmd daemon startup. If the daemon
43 does not report that it has started up within this time then the
44 parent command will exit with status of 5. This does NOT mean
45 that clvmd has not started! What it means is that the startup of
46 clvmd has been delayed for some reason; the most likely cause of
47 this is an inquorate cluster though it could be due to locking
48 latencies on a cluster with large numbers of logical volumes. If
49 you get the return code of 5 it is usually not necessary to
50 restart clvmd - it will start as soon as that blockage has
51 cleared. This flag is to allow startup scripts to exit in a
52 timely fashion even if the cluster is stalled for some reason.
53 The default is 0 (no timeout) and the value is in seconds. Don't
54 set this too small or you will experience spurious errors. 10 or
55 20 seconds might be sensible.
56 This timeout will be ignored if you start clvmd with the -d
57 switch.
58 -R Tells all the running clvmds in the cluster to reload their
60 device cache and re-read the lvm configuration file. This com‐
61 mand should be run whenever the devices on a cluster system are
62 changed.
63 -S Tells the running clvmd to exit and reexecute itself, for exam‐
65 ple at the end of a package upgrade. The new instance is
66 instructed to reacquire any locks in the same state as they were
67 previously held. (Alternative methods of restarting the daemon
68 have the side effect of changing exclusive LV locks into shared
69 locks.)
70 -I Selects the cluster manager to use for locking and internal com‐
72 munications, the available managers will be listed as part of
73 the 'clvmd -h' output. clvmd will use the first cluster manager
74 that succeeds, and it checks them in the order
75 cman,gulm,corosync,openais. As it is quite possible to have (eg)
76 corosync and cman available on the same system you might have to
77 manually specify this option to override the search.
78 -V Display the version of the cluster LVM daemon.
80
82 lvm(8)
83Red Hat Inc LVM TOOLS 2.02.84(2) (2011-02-09) CLVMD(8)