1PACEMAKER-CONTROLD(7) Pacemaker Configuration PACEMAKER-CONTROLD(7)
2
6 pacemaker-controld - Pacemaker controller options
7
9 [dc-version=string] [cluster-infrastructure=string]
10 [cluster-name=string] [dc-deadtime=time]
11 [cluster-recheck-interval=time] [load-threshold=percentage]
12 [node-action-limit=integer] [fence-reaction=string]
13 [election-timeout=time] [shutdown-escalation=time]
14 [join-integration-timeout=time] [join-finalization-timeout=time]
15 [transition-delay=time] [stonith-watchdog-timeout=time]
16 [stonith-max-attempts=integer]
17
19 Cluster options used by Pacemaker's controller
20
22 dc-version = string [none]
23 Pacemaker version on cluster node elected Designated Controller
24 (DC)
25 Includes a hash which identifies the exact changeset the code was
27 built from. Used for diagnostic purposes.
28 cluster-infrastructure = string [corosync]
30 The messaging stack on which Pacemaker is currently running
31 Used for informational and diagnostic purposes.
33 cluster-name = string [(null)]
35 An arbitrary name for the cluster
36 This optional value is mostly for users' convenience as desired in
38 administration, but may also be used in Pacemaker configuration
39 rules via the #cluster-name node attribute, and by higher-level
40 tools and resource agents.
41 dc-deadtime = time [20s]
43 How long to wait for a response from other nodes during start-up
44 The optimal value will depend on the speed and load of your network
46 and the type of switches used.
47 cluster-recheck-interval = time [15min]
49 Polling interval to recheck cluster state and evaluate rules with
50 date specifications
51 Pacemaker is primarily event-driven, and looks ahead to know when
53 to recheck cluster state for failure timeouts and most time-based
54 rules. However, it will also recheck the cluster after this amount
55 of inactivity, to evaluate rules with date specifications and serve
56 as a fail-safe for certain types of scheduler bugs. Allowed values:
57 Zero disables polling, while positive values are an interval in
58 seconds(unless other units are specified, for example "5min")
59 load-threshold = percentage [80%]
61 Maximum amount of system load that should be used by cluster nodes
62 The cluster will slow down its recovery process when the amount of
64 system resources used (currently CPU) approaches this limit
65 node-action-limit = integer [0]
67 Maximum number of jobs that can be scheduled per node (defaults to
68 2x cores)
69 fence-reaction = string [stop]
71 How a cluster node should react if notified of its own fencing
72 A cluster node may receive notification of its own fencing if
74 fencing is misconfigured, or if fabric fencing is in use that
75 doesn't cut cluster communication. Allowed values are "stop" to
76 attempt to immediately stop Pacemaker and stay stopped, or "panic"
77 to attempt to immediately reboot the local node, falling back to
78 stop on failure.
79 election-timeout = time [2min]
81 *** Advanced Use Only ***
82 Declare an election failed if it is not decided within this much
84 time. If you need to adjust this value, it probably indicates the
85 presence of a bug.
86 shutdown-escalation = time [20min]
88 *** Advanced Use Only ***
89 Exit immediately if shutdown does not complete within this much
91 time. If you need to adjust this value, it probably indicates the
92 presence of a bug.
93 join-integration-timeout = time [3min]
95 *** Advanced Use Only ***
96 If you need to adjust this value, it probably indicates the
98 presence of a bug.
99 join-finalization-timeout = time [30min]
101 *** Advanced Use Only ***
102 If you need to adjust this value, it probably indicates the
104 presence of a bug.
105 transition-delay = time [0s]
107 *** Advanced Use Only *** Enabling this option will slow down
108 cluster recovery under all conditions
109 Delay cluster recovery for this much time to allow for additional
111 events to occur. Useful if your configuration is sensitive to the
112 order in which ping updates arrive.
113 stonith-watchdog-timeout = time [0]
115 How long to wait before we can assume nodes are safely down when
116 watchdog-based self-fencing via SBD is in use
117 If nonzero, along with `have-watchdog=true` automatically set by
119 the cluster, when fencing is required, watchdog-based self-fencing
120 will be performed via SBD without requiring a fencing resource
121 explicitly configured. If `stonith-watchdog-timeout` is set to a
122 positive value, unseen nodes are assumed to self-fence within this
123 much time. +WARNING:+ It must be ensured that this value is larger
124 than the `SBD_WATCHDOG_TIMEOUT` environment variable on all nodes.
125 Pacemaker verifies the settings individually on all nodes and
126 prevents startup or shuts down if configured wrongly on the fly.
127 It's strongly recommended that `SBD_WATCHDOG_TIMEOUT` is set to the
128 same value on all nodes. If `stonith-watchdog-timeout` is set to a
129 negative value, and `SBD_WATCHDOG_TIMEOUT` is set, twice that value
130 will be used. +WARNING:+ In this case, it's essential (currently
131 not verified by Pacemaker) that `SBD_WATCHDOG_TIMEOUT` is set to
132 the same value on all nodes.
133 stonith-max-attempts = integer [10]
135 How many times fencing can fail before it will no longer be
136 immediately re-attempted on a target
137
139 Andrew Beekhof <andrew@beekhof.net>
140 Author.
141Pacemaker Configuration 06/30/2022 PACEMAKER-CONTROLD(7)