1 .TH MULTIPATH.CONF 5 "30 November 2006"
3 multipath.conf \- multipath daemon configuration file
6 is the configuration file for the multipath daemon. It is used to
7 overwrite the built-in configuration table of \fBmultipathd\fP.
8 Any line whose first non-white-space character is a '#' is considered
9 a comment line. Empty lines are ignored.
11 The configuration file contains entries of the form:
35 Each \fIsection\fP contains one or more attributes or subsections. The
36 recognized keywords for attributes or subsections depend on the
37 section in which they occor.
39 The following \fIsection\fP keywords are recognized:
42 This section defines default values for attributes which are used
43 whenever no specific setting is given.
46 This section defines which devices should be excluded from the
47 multipath topology discovery.
49 .B blacklist_exceptions
50 This section defines which devices should be included in the
51 multipath topology discovery, despite being listed in the
56 This section defines the multipath topologies. They are indexed by a
57 \fIWorld Wide Identifier\fR(wwid), which is the result of the
58 \fIgetuid_callout\fR program.
61 This section defines the device-specific settings.
64 .SH "defaults section"
67 section recognizes the following keywords:
70 interval between two path checks in seconds For properly functioning paths,
71 the interval between checks will gradually increase to (4 * polling_interval);
76 directory where udev creates its device nodes; default is
80 default verbosity. Higher values increase the verbosity level. Valid
81 levels are between 0 and 6; default is
85 The default path selector algorithm to use; they are offered by the
86 kernel multipath target. The only currently implemented is
89 .B path_grouping_policy
90 The default path grouping policy to apply to unspecified
91 multipaths. Possible values are
95 1 path per priority group
98 all paths in 1 priority group
101 1 priority group per serial number
104 1 priority group per priority value. Priorities are determined by
105 callout programs specified as a global, per-controller or
106 per-multipath option in the configuration file.
108 .B group_by_node_name
109 1 priority group per target node name. Target node names are fetched
110 in /sys/class/fc_transport/target*/node_name.
112 Default value is \fImultibus\fR.
116 The default program and args to callout to obtain a unique path
117 identifier. Should be specified with an absolute path. Default value
119 .I /lib/udev/scsi_id --whitelisted --device=/dev/%n
122 The default program and args to callout to obtain a path priority
123 value. The specified program will be executed and should return a
124 numeric value specifying the relative priority of this path. Higher
125 number have a higher priority. A '%n' in the command line will be expanded
126 to the device name, a '%b' will be expanded to the device number in
130 is a valid value. Currently the following path priority programs are
134 .B mpath_prio_emc /dev/%n
135 Generate the path priority for EMC arrays
137 .B mpath_prio_alua /dev/%n
138 Generate the path priority based on the SCSI-3 ALUA settings.
140 .B mpath_prio_netapp /dev/%n
141 Generate the path priority for NetApp arrays.
143 .B mpath_prio_rdac /dev/%n
144 Generate the path priority for LSI/Engenio RDAC controller.
146 .B mpath_prio_hp_sw /dev/%n
147 Generate the path priority for Compaq/HP controller in
150 .B mpath_prio_hds_modular %b
151 Generate the path priority for Hitachi HDS Modular storage arrays.
153 Default value is \fBnone\fR.
157 Specify any device-mapper features to be used. The most common of
159 .I "1 queue_if_no_path"
160 Note that this can also be set via the
165 The default method used to determine the paths' state. Possible values
170 Read the first sector of the device
175 command to the device.
178 Query the EMC Clariion specific EVPD page 0xC0 to determine the path
182 Check the path state for HP storage arrays with Active/Standby firmware.
185 Check the path state for LSI/Engenio RDAC storage controller.
188 Read the first sector with direct I/O.
190 Default value is \fIreadsector0\fR.
194 Tell the daemon to manage path group failback, or not to. 0 or
196 means immediate failback, values >0 means deferred failback (in
199 means no failback. Default value is
203 The number of IO to route to a path before switching to the next in
204 the same path group. Default is
208 If set to \fIpriorities\fR the multipath configurator will assign
209 path weights as "path prio * rr_min_io". Possible values are
217 Specify the number of retries until disable queueing, or
219 for immediate failure (no queueing),
221 for never stop queueing. Default is 0.
223 .B user_friendly_names
226 , using the bindings file
227 .I /etc/multipath/bindings
228 to assign a persistent and unique alias to the multipath, in the form of mpath<n>.
231 use the WWID as the alias. In either case this be will
232 be overriden by any specific aliases in the \fImultipaths\fR section.
237 Specify the maximum number of file descriptors that can be opened by multipath
238 and multipathd. This is equivalent to ulimit -n. A value of \fImax\fR will set
239 this to the system limit from /proc/sys/fs/nr_open. If this is not set, the
240 maximum number of open fds is taken from the calling process. It is usually
241 1024. To be safe, this should be set to the maximum number of paths plus 32,
242 if that number is greated than 1024.
245 Specify the number of seconds the scsi layer will wait after a problem has been
246 detected on a FC remote port before failing IO to devices on that remote port.
247 This should be smaller than dev_loss_tmo. Setting this to
249 will disable the timeout.
252 Specify the number of seconds the scsi layer will wait after a problem has
253 been detected on a FC remote port before removing it from the system.
255 .B queue_without_daemon
258 , when multipathd stops, queueing will be turned off for all devices.
259 This is useful for devices that set no_path_retry. If a machine is
260 shut down while all paths to a device are down, it is possible to hang waiting
261 for IO to return from the device after multipathd has been stopped. Without
262 multipathd running, access to the paths cannot be restored, and the kernel
263 cannot be told to stop queueing IO. Setting queue_without_daemon to
265 , avoids this problem. Default is
268 .SH "blacklist section"
271 section is used to exclude specific device from inclusion in the
272 multipath topology. It is most commonly used to exclude local disks or
273 LUNs for the array controller.
275 The following keywords are recognized:
278 The \fIWorld Wide Identification\fR of a device.
281 Regular expression of the device nodes to be excluded.
284 Subsection for the device description. This subsection recognizes the
288 keywords. For a full description of these keywords please see the
291 .SH "blacklist_exceptions section"
293 .I blacklist_exceptions
294 section is used to revert the actions of the
296 section, ie to include specific device in the
297 multipath topology. This allows to selectively include devices which
298 would normally be excluded via the
302 The following keywords are recognized:
305 The \fIWorld Wide Identification\fR of a device.
308 Regular expression of the device nodes to be excluded.
311 Subsection for the device description. This subsection recognizes the
315 keywords. For a full description of these keywords please see the
318 .SH "multipaths section"
319 The only recognized attribute for the
327 subsection recognizes the following attributes:
330 Index of the container. Mandatory for this subsection.
333 (Optional) symbolic name for the multipath map.
335 The following attributes are optional; if not set the default values
343 .B path_grouping_policy
355 .SH "devices section"
356 The only recognized attribute for the
364 subsection recognizes the following attributes:
367 (Mandatory) Vendor identifier
370 (Mandatory) Product identifier
373 Product strings to blacklist for this vendor
376 (Optional) The hardware handler to use for this device type.
377 The following hardware handler are implemented:
381 Hardware handler for EMC storage arrays.
384 The following attributes are optional; if not set the default values
392 .B path_grouping_policy
423 processes being hung and not killable in situations where all the paths to the LUN go offline.
424 It is advisable to use the
434 was developed by Christophe Varoqui, <christophe.varoqui@opensvc.com> and others.