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 occur.
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.
75 This value will be overridden by the
77 setting in the multipathd.service definition if systemd is used.
81 .B max_polling_interval
82 maximal interval between two path checks in seconds; default is
83 .I 4 * polling_interval
86 directory where the dynamic shared objects are stored; default is system
91 default verbosity. Higher values increase the verbosity level. Valid
92 levels are between 0 and 6; default is
96 enable reassigning of device-mapper maps. With this option multipathd
97 will remap existing device-mapper maps to always point to multipath
98 device, not the underlying block devices. Possible values are
99 \fIyes\fR and \fIno\fR. Default is
103 The default path selector algorithm to use; they are offered by the
104 kernel multipath target. There are three selector algorithms.
108 Loop through every path in the path group, sending the same amount of IO to
112 Send the next bunch of IO down the path with the least amount of outstanding IO.
115 Choose the path for the next bunch of IO based on the amount of outstanding IO
116 to the path and its relative throughput.
119 .B path_grouping_policy
120 The default path grouping policy to apply to unspecified
121 multipaths. Possible values are
125 1 path per priority group
128 all paths in 1 priority group
131 1 priority group per serial number
134 1 priority group per priority value. Priorities are determined by
135 callout programs specified as a global, per-controller or
136 per-multipath option in the configuration file.
138 .B group_by_node_name
139 1 priority group per target node name. Target node names are fetched
140 in /sys/class/fc_transport/target*/node_name.
142 Default value is \fIfailover\fR.
146 The udev attribute providing a unique path
147 identifier. Default value is
151 The default program and args to callout to obtain a unique path
152 identifier. Should be specified with an absolute path.
153 This parameter is deprecated; \fIuid_attribute\fR should be used instead.
156 The name of the path priority routine. The specified routine
157 should return a numeric value specifying the relative priority
158 of this path. Higher number have a higher priority.
160 is a valid value. Currently the following path priority routines
165 Return a constant priority of \fI1\fR.
168 Generate the path priority for EMC arrays.
171 Generate the path priority based on the SCSI-3 ALUA settings.
174 Generate the path priority for NetApp arrays.
177 Generate the path priority for LSI/Engenio/NetApp E-Series RDAC controller.
180 Generate the path priority for Compaq/HP controller in
184 Generate the path priority for Hitachi HDS Modular storage arrays.
187 Generate a random priority between 1 and 10.
190 Generate the path priority based on the regular expression and the
191 priority provided as argument. requires prio_args keyword.
193 Default value is \fBnone\fR.
197 Arguments to pass to to the prio function. Currently only used with
198 .I weighted, which needs a value of the form
199 .I "<hbtl|devname> <regex1> <prio1> <regex2> <prio2> ..."
201 regex can be of SCSI H:B:T:L format Ex: 1:0:.:. , *:0:0:.
203 regex can be of device name format Ex: sda , sd.e
206 Specify any device-mapper features to be used. Syntax is
210 is the number of features in
212 Possible values for the feature list are
216 Queue IO if no path is active; identical to the
221 Disable automatic partitions generation via kpartx.
225 The default method used to determine the paths state. Possible values
230 (Deprecated) Read the first sector of the device. This checker is being
231 deprecated, please use \fIdirectio\fR instead
236 command to the device.
239 Query the EMC Clariion specific EVPD page 0xC0 to determine the path
243 Check the path state for HP storage arrays with Active/Standby firmware.
246 Check the path state for LSI/Engenio/NetApp E-Series RDAC storage controller.
249 Read the first sector with direct I/O.
251 Default value is \fIdirectio\fR.
255 Tell multipathd how to manage path group failback.
259 Immediately failback to the highest priority pathgroup that contains
263 Do not perform automatic failback.
266 Only perform automatic failback when the first path of a pathgroup
267 becomes active. This keeps a node from automatically failing back when
268 another node requested the failover.
271 deferred failback (time to defer in seconds)
273 Default value is \fImanual\fR.
277 The number of IO to route to a path before switching to the next in
278 the same path group. This is only for BIO based multipath. Default is
282 The number of IO requests to route to a path before switching to the
283 next in the same path group. This is only for request based multipath.
288 If set to \fIpriorities\fR the multipath configurator will assign
289 path weights as "path prio * rr_min_io". Possible values are
297 Specify the number of retries until disable queueing, or
299 for immediate failure (no queueing),
301 for never stop queueing. If unset no queueing is attempted.
304 .B user_friendly_names
307 , using the bindings file
308 .I /etc/multipath/bindings
309 to assign a persistent and unique alias to the multipath, in the form of mpath<n>.
312 use the WWID as the alias. In either case this be will
313 be overridden by any specific aliases in the \fImultipaths\fR section.
320 , multipathd will disable queueing when the last path to a device has been
325 Specify the maximum number of file descriptors that can be opened by multipath
326 and multipathd. This is equivalent to ulimit \-n. A value of \fImax\fR will set
327 this to the system limit from /proc/sys/fs/nr_open. If this is not set, the
328 maximum number of open fds is taken from the calling process. It is usually
329 1024. To be safe, this should be set to the maximum number of paths plus 32,
330 if that number is greated than 1024.
333 Specify the timeout to user for path checkers that issue scsi commands with an
334 explicit timeout, in seconds; default taken from
335 .I /sys/block/sd<x>/device/timeout
338 Specify the number of seconds the scsi layer will wait after a problem has been
339 detected on a FC remote port before failing IO to devices on that remote port.
340 This should be smaller than dev_loss_tmo. Setting this to
342 will disable the timeout.
345 Specify the number of seconds the scsi layer will wait after a problem has
346 been detected on a FC remote port before removing it from the system. This
347 can be set to "infinity" which sets it to the max value of 2147483647
348 seconds, or 68 years. It will be automatically adjusted to the overall
350 \fIno_path_retry\fR * \fIpolling_interval\fR
351 if a number of retries is given with \fIno_path_retry\fR and the
352 overall retry interval is longer than the specified \fIdev_loss_tmo\fR value.
353 The linux kernel will cap this value to \fI300\fR if \fBfast_io_fail_tmo\fR
356 .B queue_without_daemon
359 , when multipathd stops, queueing will be turned off for all devices.
360 This is useful for devices that set no_path_retry. If a machine is
361 shut down while all paths to a device are down, it is possible to hang waiting
362 for IO to return from the device after multipathd has been stopped. Without
363 multipathd running, access to the paths cannot be restored, and the kernel
364 cannot be told to stop queueing IO. Setting queue_without_daemon to
366 , avoids this problem. Default is
370 The full pathname of the binding file to be used when the user_friendly_names option is set. Defaults to
371 .I /etc/multipath/bindings
374 The full pathname of the wwids file, which is used by multipath to keep track
375 of the wwids for LUNs it has created multipath devices on in the past.
377 .I /etc/multipath/wwids
382 , multipathd logs the first path checker error at logging level 2. Any later
383 errors are logged at level 3 until the device is restored. If set to
385 , multipathd always logs the path checker error at logging level 2. Default is
389 This is the service action reservation key used by mpathpersist. It must be
390 set for all multipath devices using persistent reservations, and it must be
391 the same as the RESERVATION KEY field of the PERSISTENT RESERVE OUT parameter
392 list which contains an 8-byte value provided by the application client to the
393 device server to identify the I_T nexus. It is unset by default.
395 .B retain_attached_hw_handler
398 and the scsi layer has already attached a hardware_handler to the device,
399 multipath will not force the device to use the hardware_handler specified by
400 mutipath.conf. If the scsi layer has not attached a hardware handler,
401 multipath will continue to use its configured hardware handler. Default is
407 , multipath will try to detect if the device supports ALUA. If so, the device
408 will automatically use the
410 prioritizer. If not, the prioritizer will be selected as usual. Default is
413 .SH "blacklist section"
416 section is used to exclude specific device from inclusion in the
417 multipath topology. It is most commonly used to exclude local disks or
418 LUNs for the array controller.
420 The following keywords are recognized:
423 The \fIWorld Wide Identification\fR of a device.
426 Regular expression of the device nodes to be excluded.
429 Regular expresion of the udev property to be excluded.
432 Subsection for the device description. This subsection recognizes the
436 keywords. For a full description of these keywords please see the
439 .SH "blacklist_exceptions section"
441 .I blacklist_exceptions
442 section is used to revert the actions of the
444 section, ie to include specific device in the
445 multipath topology. This allows one to selectively include devices which
446 would normally be excluded via the
450 The following keywords are recognized:
453 The \fIWorld Wide Identification\fR of a device.
456 Regular expresion of the udev property to be whitelisted. Defaults to
457 .I (ID_WWN|ID_SCSI_VPD)
460 Regular expression of the device nodes to be whitelisted.
463 Subsection for the device description. This subsection recognizes the
467 keywords. For a full description of these keywords please see the
473 blacklist and whitelist handling is different from the usual handling
474 in the sense that the whitelist
476 to be set, otherwise the device will be blacklisted.
477 In these cases the message
478 .I blacklisted, udev property missing
480 .SH "multipaths section"
481 The only recognized attribute for the
489 subsection recognizes the following attributes:
492 Index of the container. Mandatory for this subsection.
495 (Optional) symbolic name for the multipath map.
497 The following attributes are optional; if not set the default values
507 .B path_grouping_policy
533 .SH "devices section"
534 The only recognized attribute for the
542 subsection recognizes the following attributes:
545 (Mandatory) Vendor identifier
548 (Mandatory) Product identifier
551 (Optional) Revision identfier
554 (Optional) Product strings to blacklist for this vendor
557 (Optional) The user_friendly_names prefix to use for this
558 device type, instead of the default "mpath"
561 (Optional) The hardware handler to use for this device type.
562 The following hardware handler are implemented:
566 Hardware handler for EMC storage arrays.
569 Hardware handler for LSI/Engenio/NetApp E-Series RDAC storage controller.
572 Hardware handler for Compaq/HP storage arrays in active/standby
576 Hardware handler for SCSI-3 ALUA compatible arrays.
579 The following attributes are optional; if not set the default values
587 .B path_grouping_policy
617 .B retain_attached_hw_handler
628 processes being hung and not killable in situations where all the paths to the LUN go offline.
629 It is advisable to use the
637 might lead to a deadlock if the
639 setting results in a device being removed while I/O is still queued.
640 The multipath daemon will update the
642 setting accordingly to avoid this deadlock. Hence if both values are
643 specified the order of precedence is
644 .I no_path_retry, queue_if_no_path, dev_loss_tmo
653 was developed by Christophe Varoqui, <christophe.varoqui@opensvc.com> and others.