Gitweb:
http://git.fedorahosted.org/git/cluster.git?p=cluster.git;a=commitdiff;h=...
Commit: 878c9fa5cd0d9a8be0f6e65361f50f43a8ec4843
Parent: ff1f5c78203613a8e28be6d7de1d558d598f8ca9
Author: David Teigland <teigland(a)redhat.com>
AuthorDate: Thu Apr 5 14:39:59 2012 -0500
Committer: David Teigland <teigland(a)redhat.com>
CommitterDate: Thu Apr 5 14:39:59 2012 -0500
dlm_controld: update man page
Signed-off-by: David Teigland <teigland(a)redhat.com>
---
dlm_controld/dlm_controld.8 | 345 +++++++++----------------------------------
dlm_controld/main.c | 5 +-
2 files changed, 71 insertions(+), 279 deletions(-)
diff --git a/dlm_controld/dlm_controld.8 b/dlm_controld/dlm_controld.8
index 7100f0e..99190da 100644
--- a/dlm_controld/dlm_controld.8
+++ b/dlm_controld/dlm_controld.8
@@ -1,313 +1,104 @@
-.TH DLM_CONTROLD 8 2009-01-18 cluster cluster
+.TH DLM_CONTROLD 8 2012-04-05 dlm dlm
.SH NAME
-dlm_controld \- daemon that configures dlm according to cluster events
+dlm_controld \- dlm cluster control daemon
.SH SYNOPSIS
.B dlm_controld
[OPTIONS]
.SH DESCRIPTION
-The dlm lives in the kernel, and the cluster infrastructure (corosync
-membership and group management) lives in user space. The dlm in the
-kernel needs to adjust/recover for certain cluster events. It's the job
-of dlm_controld to receive these events and reconfigure the kernel dlm as
-needed. dlm_controld controls and configures the dlm through sysfs and
-configfs files that are considered dlm-internal interfaces.
-
-The cman init script usually starts the dlm_controld daemon.
+The kernel dlm requires a user daemon to manage lockspace membership.
+dlm_controld manages lockspace membership using corosync cpg groups,
+and translates membership changes into dlm kernel recovery events.
+dlm_controld also manages posix locks for cluster file systems using
+the dlm.
.SH OPTIONS
-Command line options override a corresponding setting in cluster.conf.
-
-.TP
-.B \-D
-Enable debugging to stderr and don't fork.
-.br
-See also
-.B dlm_tool dump
-in
-.BR dlm_tool (8).
-
-.TP
-.B \-L
-Enable debugging to log file.
-.br
-See also
-.B logging
-in
-.BR cluster.conf (5).
-
-.TP
-.B \-K
-Enable kernel dlm debugging messages.
-.br
-See also
-.B log_debug
-below.
-
-.TP
-.BI \-r " num"
-dlm kernel lowcomms protocol, 0 tcp, 1 sctp, 2 detect.
-2 selects tcp if corosync rrp_mode is "none", otherwise sctp.
-.br
-Default 2.
-
-.TP
-.BI \-g " num"
-groupd compatibility mode, 0 off, 1 on.
-.br
-Default 0.
-
-.TP
-.BI \-f " num"
-Enable (1) or disable (0) fencing recovery dependency.
-.br
-Default 1.
-
-.TP
-.BI \-q " num"
-Enable (1) or disable (0) quorum recovery dependency.
-.br
-Default 0.
-
-.TP
-.BI \-d " num"
-Enable (1) or disable (0) deadlock detection code.
-.br
-Default 0.
-
-.TP
-.BI \-p " num"
-Enable (1) or disable (0) plock code for cluster fs.
-.br
-Default 1.
-
-.TP
-.BI \-l " num"
-Limit the rate of plock operations, 0 for no limit.
-.br
-Default 0.
-
-.TP
-.BI \-o " num"
-Enable (1) or disable (0) plock ownership.
-.br
-Default 1.
-
-.TP
-.BI \-t " ms"
-Plock ownership drop resources time (milliseconds).
-.br
-Default 10000.
-
-.TP
-.BI \-c " num"
-Plock ownership drop resources count.
-.br
-Default 10.
-
-.TP
-.BI \-a " ms"
-Plock ownership drop resources age (milliseconds).
+Command line options override a corresponding setting in
+.BR dlm.conf (5).
.br
-Default 10000.
-
-.TP
-.B \-P
-Enable plock debugging messages (can produce excessive output).
-
-.TP
-.B \-h
-Print a help message describing available options, then exit.
-
-.TP
-.B \-V
-Print program version information, then exit.
-
-
-.SH FILES
-.BR cluster.conf (5)
-is usually located at /etc/cluster/cluster.conf. It is not read directly.
-Other cluster components load the contents into memory, and the values are
-accessed through the libccs library.
-
-Configuration options for dlm (kernel) and dlm_controld are added to the
-<dlm /> section of cluster.conf, within the top level <cluster> section.
-
-.SS Kernel options
-
-.TP
-.B protocol
-The network
-.B protocol
-can be set to tcp, sctp or detect which selects tcp or sctp based on
-the corosync rrp_mode configuration (redundant ring protocol).
-The rrp_mode "none" results in tcp. Default detect.
-
-<dlm protocol="detect"/>
-
-.TP
-.B timewarn
-After waiting
-.B timewarn
-centiseconds, the dlm will emit a warning via netlink. This only applies
-to lockspaces created with the DLM_LSFL_TIMEWARN flag, and is used for
-deadlock detection. Default 500 (5 seconds).
-
-<dlm timewarn="500"/>
-
-.TP
-.B log_debug
-DLM kernel debug messages can be enabled by setting
-.B log_debug
-to 1. Default 0.
-
-<dlm log_debug="0"/>
-
-.TP
-.B clusternode/weight
-The lock directory
-.B weight
-can be specified one the clusternode lines. Weights would usually be
-used in the lock server configurations shown below instead.
-
-<clusternode name="node01" nodeid="1" weight="1"/>
-
-.SS Daemon options
-
-.TP
-.B enable_fencing
-See command line description.
-
-<dlm enable_fencing="1"/>
-
-.TP
-.B enable_quorum
-See command line description.
-
-<dlm enable_quorum="0"/>
-
-.TP
-.B enable_deadlk
-See command line description.
-
-<dlm enable_deadlk="0"/>
-
-.TP
-.B enable_plock
-See command line description.
-
-<dlm enable_plock="1"/>
-
-.TP
-.B plock_rate_limit
-See command line description.
-
-<dlm plock_rate_limit="0"/>
-
-.TP
-.B plock_ownership
-See command line description.
-
-<dlm plock_ownership="1"/>
+For default settings, see dlm_controld -h.
-.TP
-.B drop_resources_time
-See command line description.
+.B --daemon_debug | -D
+ enable debugging to stderr and don't fork
-<dlm drop_resources_time="10000"/>
+.B --log_debug | -K
+ enable kernel dlm debugging messages
-.TP
-.B drop_resources_count
-See command line description.
+.B --protocol | -r
+.I str
+ dlm kernel lowcomms protocol: tcp, sctp, detect
-<dlm drop_resources_count="10"/>
+.B --debug_logfile | -L
+ write debugging to log file
-.TP
-.B drop_resources_age
-See command line description.
+.B --enable_plock | -p
+0|1
+ enable/disable posix lock support for cluster fs
-<dlm drop_resources_age="10000"/>
+.B --plock_debug | -P
+ enable plock debugging
-.TP
-.B plock_debug
-Enable (1) or disable (0) plock debugging messages (can produce excessive
-output). Default 0.
+.B --plock_rate_limit | -l
+.I int
+ limit rate of plock operations (0 for none)
-<dlm plock_debug="0"/>
+.B --plock_ownership | -o
+0|1
+ enable/disable plock ownership
+.B --drop_resources_time | -t
+.I int
+ plock ownership drop resources time (milliseconds)
-.SS Disabling resource directory
+.B --drop_resources_count | -c
+.I int
+ plock ownership drop resources count
-Lockspaces usually use a resource directory to keep track of which node is
-the master of each resource. The dlm can operate without the resource
-directory, though, by statically assigning the master of a resource using
-a hash of the resource name. To enable, set the per-lockspace
-.B nodir
-option to 1.
+.B --drop_resources_age | -a
+.I int
+ plock ownership drop resources age (milliseconds)
-.nf
-<dlm>
- <lockspace name="foo" nodir="1">
-</dlm>
-.fi
+.B --post_join_delay | -j
+.I int
+ seconds to delay fencing after cluster join
-.SS Lock-server configuration
+.B --enable_fencing | -f
+0|1
+ enable/disable fencing
-The nodir setting can be combined with node weights to create a
-configuration where select node(s) are the master of all resources/locks.
-These
-.B master
-nodes can be viewed as "lock servers" for the other nodes.
+.B --enable_concurrent_fencing
+0|1
+ enable/disable concurrent fencing
-.nf
-<dlm>
- <lockspace name="foo" nodir="1">
- <master name="node01"/>
- </lockspace>
-</dlm>
+.B --enable_startup_fencing | -s
+0|1
+ enable/disable startup fencing
-or,
+.B --enable_quorum_fencing | -q
+0|1
+ enable/disable quorum requirement for fencing
-<dlm>
- <lockspace name="foo" nodir="1">
- <master name="node01"/>
- <master name="node02"/>
- </lockspace>
-</dlm>
-.fi
+.B --enable_quorum_lockspace
+0|1
+ enable/disable quorum requirement for lockspace operations
-Lock management will be partitioned among the available masters. There
-can be any number of masters defined. The designated master nodes will
-master all resources/locks (according to the resource name hash). When no
-masters are members of the lockspace, then the nodes revert to the common
-fully-distributed configuration. Recovery is faster, with little
-disruption, when a non-master node joins/leaves.
+.B --fence_all
+.I str
+ fence all nodes with this agent
-There is no special mode in the dlm for this lock server configuration,
-it's just a natural consequence of combining the "nodir" option with node
-weights. When a lockspace has master nodes defined, the master has a
-default weight of 1 and all non-master nodes have weight of 0. An explicit
-non-zero
-.B weight
-can also be assigned to master nodes, e.g.
+.B --unfence_all
+.I str
+ unfence all nodes with this agent
-.nf
-<dlm>
- <lockspace name="foo" nodir="1">
- <master name="node01" weight="2"/>
- <master name="node02" weight="1"/>
- </lockspace>
-</dlm>
-.fi
+.B --help | -h
+ print this help, then exit
-In which case node01 will master 2/3 of the total resources and node2 will
-master the other 1/3.
+.B --version | -V
+ Print program version information, then exit
.SH SEE ALSO
.BR dlm_tool (8),
-.BR fenced (8),
-.BR cman (5),
-.BR cluster.conf (5)
+.BR dlm.conf (5)
diff --git a/dlm_controld/main.c b/dlm_controld/main.c
index 872cdfd..cc1f0b5 100644
--- a/dlm_controld/main.c
+++ b/dlm_controld/main.c
@@ -1183,7 +1183,8 @@ static void print_usage(void)
printf("\n");
printf("dlm_controld [options]\n");
printf("\n");
- printf("Options:\n");
+ printf("Option [arg]\n");
+ printf("Description [default]\n");
printf("\n");
for (i = 0; i < dlm_options_max; i++) {
@@ -1264,7 +1265,7 @@ static void set_opt_defaults(void)
set_opt_default(enable_fscontrol_ind,
"enable_fscontrol", '\0', req_arg_bool,
0, NULL,
- "enable/disable recovery coordination with fs controld");
+ ""); /* do not advertise */
set_opt_default(enable_plock_ind,
"enable_plock", 'p', req_arg_bool,