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 values are given in the appropriate device or multipath
47 This section defines which devices should be excluded from the
48 multipath topology discovery.
50 .B blacklist_exceptions
51 This section defines which devices should be included in the
52 multipath topology discovery, despite being listed in the
57 This section defines the multipath topologies. They are indexed by a
58 \fIWorld Wide Identifier\fR(wwid), which is taken to be the value of
59 the udev attribute given by the
60 \fIuid_attribute\fR keyword.
63 This section defines the device-specific settings.
66 .SH "defaults section"
69 section recognizes the following keywords:
72 interval between two path checks in seconds. For properly functioning paths,
73 the interval between checks will gradually increase to
74 .B max_polling_interval;
78 .B max_polling_interval
79 maximal interval between two path checks in seconds; default is
80 .I 4 * polling_interval
83 directory where the dynamic shared objects are stored; default is system
88 default verbosity. Higher values increase the verbosity level. Valid
89 levels are between 0 and 6; default is
93 enable reassigning of device-mapper maps. With this option multipathd
94 will remap existing device-mapper maps to always point to multipath
95 device, not the underlying block devices. Possible values are
96 \fIyes\fR and \fIno\fR. Default is
100 The default path selector algorithm to use; they are offered by the
101 kernel multipath target. There are three selector algorithms.
105 Loop through every path in the path group, sending the same amount of IO to
109 Send the next bunch of IO down the path with the least amount of outstanding IO.
112 Choose the path for the next bunch of IO based on the amount of outstanding IO
113 to the path and its relative throughput.
116 .B path_grouping_policy
117 The default path grouping policy to apply to unspecified
118 multipaths. Possible values are
122 1 path per priority group
125 all paths in 1 priority group
128 1 priority group per serial number
131 1 priority group per priority value. Priorities are determined by
132 callout programs specified as a global, per-controller or
133 per-multipath option in the configuration file.
135 .B group_by_node_name
136 1 priority group per target node name. Target node names are fetched
137 in /sys/class/fc_transport/target*/node_name.
139 Default value is \fImultibus\fR.
143 The udev attribute providing a unique path
144 identifier. Default value is
148 The name of the path priority routine. The specified routine
149 should return a numeric value specifying the relative priority
150 of this path. Higher number have a higher priority.
152 is a valid value. Currently the following path priority routines
157 Return a constant priority of \fI1\fR.
160 Generate the path priority for EMC arrays.
163 Generate the path priority based on the SCSI-3 ALUA settings.
166 Generate the path priority for NetApp arrays.
169 Generate the path priority for LSI/Engenio/NetApp E-Series RDAC controller.
172 Generate the path priority for Compaq/HP controller in
176 Generate the path priority for Hitachi HDS Modular storage arrays.
179 Generate a random priority between 1 and 10.
182 Generate the path priority based on the regular expression and the
183 priority provided as argument. requires prio_args keyword.
185 Default value is \fBnone\fR.
189 Arguments to pass to to the prio function. Currently only used with
190 .I weighted, which needs a value of the form
191 .I "<hbtl|devname> <regex1> <prio1> <regex2> <prio2> ..."
193 regex can be of SCSI H:B:T:L format Ex: 1:0:.:. , *:0:0:.
195 regex can be of device name format Ex: sda , sd.e
198 Specify any device-mapper features to be used. Syntax is
202 is the number of features in
204 Possible values for the feature list are
208 Queue IO if no path is active; identical to the
213 Disable automatic partitions generation via kpartx.
217 The default method used to determine the paths state. Possible values
222 (Deprecated) Read the first sector of the device. This checker is being
223 deprecated, please use \fIdirectio\fR instead
228 command to the device.
231 Query the EMC Clariion specific EVPD page 0xC0 to determine the path
235 Check the path state for HP storage arrays with Active/Standby firmware.
238 Check the path state for LSI/Engenio/NetApp E-Series RDAC storage controller.
241 Read the first sector with direct I/O.
243 Default value is \fIdirectio\fR.
247 Tell multipathd how to manage path group failback.
251 Immediately failback to the highest priority pathgroup that contains
255 Do not perform automatic failback.
258 Only perform automatic failback when the first path of a pathgroup
259 becomes active. This keeps a node from automatically failing back when
260 another node requested the failover.
263 deferred failback (time to defer in seconds)
265 Default value is \fImanual\fR.
269 The number of IO to route to a path before switching to the next in
270 the same path group. This is only for BIO based multipath. Default is
274 The number of IO requests to route to a path before switching to the
275 next in the same path group. This is only for request based multipath.
280 If set to \fIpriorities\fR the multipath configurator will assign
281 path weights as "path prio * rr_min_io". Possible values are
289 Specify the number of retries until disable queueing, or
291 for immediate failure (no queueing),
293 for never stop queueing. Default is 0.
295 .B user_friendly_names
298 , using the bindings file
299 .I /etc/multipath/bindings
300 to assign a persistent and unique alias to the multipath, in the form of mpath<n>.
303 use the WWID as the alias. In either case this be will
304 be overriden by any specific aliases in the \fImultipaths\fR section.
311 , multipathd will disable queueing when the last path to a device has been
316 Specify the maximum number of file descriptors that can be opened by multipath
317 and multipathd. This is equivalent to ulimit -n. A value of \fImax\fR will set
318 this to the system limit from /proc/sys/fs/nr_open. If this is not set, the
319 maximum number of open fds is taken from the calling process. It is usually
320 1024. To be safe, this should be set to the maximum number of paths plus 32,
321 if that number is greated than 1024.
324 Specify the timeout to user for path checkers that issue scsi commands with an
325 explict timeout, in seconds; default taken from
326 .I /sys/block/sd<x>/device/timeout
329 Specify the number of seconds the scsi layer will wait after a problem has been
330 detected on a FC remote port before failing IO to devices on that remote port.
331 This should be smaller than dev_loss_tmo. Setting this to
333 will disable the timeout.
336 Specify the number of seconds the scsi layer will wait after a problem has
337 been detected on a FC remote port before removing it from the system. This
338 can be set to "infinity" which sets it to the max value of 2147483647
339 seconds, or 68 years.
341 .B queue_without_daemon
344 , when multipathd stops, queueing will be turned off for all devices.
345 This is useful for devices that set no_path_retry. If a machine is
346 shut down while all paths to a device are down, it is possible to hang waiting
347 for IO to return from the device after multipathd has been stopped. Without
348 multipathd running, access to the paths cannot be restored, and the kernel
349 cannot be told to stop queueing IO. Setting queue_without_daemon to
351 , avoids this problem. Default is
355 The full pathname of the binding file to be used when the user_friendly_names option is set. Defaults to
356 .I /etc/multipath/bindings
359 The full pathname of the wwids file, which is used by multipath to keep track
360 of the wwids for LUNs it has created multipath devices on in the past.
362 .I /etc/multipath/wwids
367 , multipathd logs the first path checker error at logging level 2. Any later
368 errors are logged at level 3 until the device is restored. If set to
370 , multipathd always logs the path checker error at logging level 2. Default is
373 .SH "blacklist section"
376 section is used to exclude specific device from inclusion in the
377 multipath topology. It is most commonly used to exclude local disks or
378 LUNs for the array controller.
380 The following keywords are recognized:
383 The \fIWorld Wide Identification\fR of a device.
386 Regular expression of the device nodes to be excluded.
389 Subsection for the device description. This subsection recognizes the
393 keywords. For a full description of these keywords please see the
396 .SH "blacklist_exceptions section"
398 .I blacklist_exceptions
399 section is used to revert the actions of the
401 section, ie to include specific device in the
402 multipath topology. This allows to selectively include devices which
403 would normally be excluded via the
407 The following keywords are recognized:
410 The \fIWorld Wide Identification\fR of a device.
413 Regular expression of the device nodes to be excluded.
416 Subsection for the device description. This subsection recognizes the
420 keywords. For a full description of these keywords please see the
423 .SH "multipaths section"
424 The only recognized attribute for the
432 subsection recognizes the following attributes:
435 Index of the container. Mandatory for this subsection.
438 (Optional) symbolic name for the multipath map.
440 The following attributes are optional; if not set the default values
450 .B path_grouping_policy
474 .SH "devices section"
475 The only recognized attribute for the
483 subsection recognizes the following attributes:
486 (Mandatory) Vendor identifier
489 (Mandatory) Product identifier
492 (Optional) Revision identfier
495 (Optional) Product strings to blacklist for this vendor
498 (Optional) The user_friendly_names prefix to use for this
499 device type, instead of the default "mpath"
502 (Optional) The hardware handler to use for this device type.
503 The following hardware handler are implemented:
507 Hardware handler for EMC storage arrays.
510 Hardware handler for LSI/Engenio/NetApp E-Series RDAC storage controller.
513 Hardware handler for Compaq/HP storage arrays in active/standby
517 Hardware handler for SCSI-3 ALUA compatible arrays.
520 The following attributes are optional; if not set the default values
528 .B path_grouping_policy
565 processes being hung and not killable in situations where all the paths to the LUN go offline.
566 It is advisable to use the
574 might lead to a deadlock if the
576 setting results in a device being removed while I/O is still queued.
577 The multipath daemon will update the
579 setting accordingly to avoid this deadlock. Hence if both values are
580 specified the order of precedence is
581 .I no_path_retry, queue_if_no_path, dev_loss_tmo
590 was developed by Christophe Varoqui, <christophe.varoqui@opensvc.com> and others.