source: gpfs_3.1_ker2.6.20/share/man/man8/mmcrvsd.8 @ 16

Last change on this file since 16 was 16, checked in by rock, 16 years ago
File size: 11.5 KB
Line 
1.TH mmcrvsd 03/16/06
2mmcrvsd Command
3.SH "Name"
4.PP
5\fBmmcrvsd\fR - Creates virtual shared disks for use by GPFS.
6.SH "Synopsis"
7.PP
8\fBmmcrvsd\fR [\fB-f\fR \fIFanoutNumber\fR]
9[\fB-y\fR] [\fB-c\fR] \fB-F\fR \fIDescFile\fR
10.SH "Description"
11.PP
12The \fBmmcrvsd\fR command can be used to create virtual shared
13disks for subsequent use by the
14\fBmmcrnsd\fR
15command. IBM Virtual shared disk is a subsystem that permits
16application programs that are running on different nodes of an RSCT peer
17domain access a raw logical volume as if it were local at each of the
18nodes. Virtual shared disks created with \fBmmcrvsd\fR follow the
19convention of one local volume group, one local logical volume, one global
20volume group, and one virtual shared disk per physical volume. After
21the virtual shared disk is created, it is configured and started on each node
22with a defined virtual shared disk adapter. See the
23\fBupdatevsdnode\fR command in the correct manual for your environment
24at: publib.boulder.ibm.com/clresctr/windows/public/rsctbooks.html.
25.PP
26The \fBmmcrvsd\fR command can be used only in the AIX
27environment.
28.PP
29Where possible, the \fBmmcrvsd\fR command creates and starts virtual
30shared disk components in parallel. For instance, when multiple
31physical disk servers are specified in the disk descriptor file, their LVM
32components are created in parallel. Starting of all virtual shared
33disks, on all nodes, always occurs in parallel.
34.PP
35The \fBmmcrvsd\fR command may also be restarted should one of the steps
36fail.
37.PP
38\fBResults\fR
39.PP
40Upon successful completion of the \fBmmcrvsd\fR command:
41.RS +3
42.HP 3
43\(bu Virtual shared disks are created.
44.sp
45If a desired name \fBvsdname\fR is specified on the disk descriptor,
46\fBmmcrvsd\fR uses that name for the name of the virtual shared
47disk. If a desired name is not specified, the virtual shared disk is
48assigned a name according to the convention:
49.sp
50.RS +3
51\fBgpfs\fINN\fRvsd
52\fR
53.RE
54.RS +9
55where \fINN\fR is a unique nonnegative integer not used in any prior
56virtual shared disk named with this convention.
57.RE
58.HP 3
59\(bu Virtual shared disks are synchronously started on all nodes.
60.HP 3
61\(bu If a desired name \fBvsdname\fR is specified on the disk
62descriptor, \fBmmcrvsd\fR uses that name as the basis for the names of the
63global volume group, local logical volume, and local volume group according to
64the convention:
65.sp
66.RS +3
67\fBvsdnamegvg
68\fR
69.RE
70.RS +9
71the global volume group
72.RE
73.sp
74.RS +3
75\fBvsdnamelv
76\fR
77.RE
78.RS +9
79the local logical volume
80.RE
81.sp
82.RS +3
83\fBvsdnamevg
84\fR
85.RE
86.RS +9
87the local volume group
88.RE
89.HP 3
90\(bu If a desired name is not specified, the global volume group, local logical
91volume, and local volume group for the virtual shared disk are named according
92to the convention:
93.sp
94.RS +3
95\fBgpfs\fINN\fRgvg
96\fR
97.RE
98.RS +9
99the global volume group
100.RE
101.sp
102.RS +3
103\fBgpfs\fINN\fRlv
104\fR
105.RE
106.RS +9
107the local logical volume
108.RE
109.sp
110.RS +3
111\fBgpfs\fINN\fRvg
112\fR
113.RE
114.RS +9
115the local volume group
116.RE
117.sp
118where gpfs\fINN\fRvsd was the name chosen for the virtual shared
119disk.
120.HP 3
121\(bu The primary server is configured and the volume group is varied
122online there.
123.HP 3
124\(bu The backup server is configured and the volume group is imported there,
125but varied off.
126.HP 3
127\(bu The \fIDescFile\fR file is rewritten to contain the created virtual
128shared disk names in place of any disk descriptors containing physical disk or
129vpath names. Primary and backup servers are omitted from the rewritten
130disk descriptor and all other fields, if specified, are copied without
131modification. The rewritten disk descriptor file can then be used as
132input to the
133\fBmmcrnsd\fR command.
134.RE
135.PP
136\fBError recovery\fR
137.PP
138Each step of the \fBmmcrvsd\fR process is enumerated during
139command execution. For example at step 0, the \fBmmcrvsd\fR command
140prints:
141.sp
142.nf
143Step \fI0\fR: Setting up environment
144.fi
145.sp
146.PP
147As each step is started, its corresponding number is recorded in the
148\fIDescFile\fR file as a comment at the end. This comment serves as
149restart information to subsequent invocations of the \fBmmcrvsd\fR
150command. For example at step one, the recorded comment would be:
151.sp
152.nf
153#MMCRVSD_STEP=\fI0\fR
154.fi
155.sp
156.PP
157Upon failure, appropriate error messages from the failing system component
158are displayed along with \fBmmcrvsd\fR error messages.
159.PP
160After correcting the failing condition and restarting the \fBmmcrvsd\fR
161command with the same descriptor file, the command prompts you to restart at
162the last failing step. For example, if a prior invocation of
163\fBmmcrvsd\fR failed at step one, the prompt would be:
164.sp
165.nf
166A prior invocation of this command has recorded a partial
167completion in the file (/tmp/\fIDescFile\fR).
168Should we restart at prior failing step(\fI1\fR)?[y]/n=>
169.fi
170.sp
171.PP
172The default response is \fBy\fR; yes restart at the prior failing
173step.
174.SH "Parameters"
175.PP
176.RS +3
177\fB-F \fIDescFile\fR
178\fR
179.RE
180.RS +9
181The file containing the list of disk descriptors, one per line, in the
182form:
183.sp
184.nf
185DiskName:PrimaryServer:BackupServer:DiskUsage:FailureGroup:DesiredName:StoragePool
186.fi
187.sp
188.PP
189.RS +3
190\fB\fIDiskName\fR
191\fR
192.RE
193.RS +9
194The device name of the disk you want to use to create a virtual shared
195disk. This can be either an hdisk name or a vpath name for an SDD
196device. GPFS performance and recovery processes function best with one
197disk per virtual shared disk. If you want to create virtual shared
198disks with more than one disk, refer to the correct manual for your
199environment at: publib.boulder.ibm.com/clresctr/windows/public/rsctbooks.html
200.RE
201.PP
202.RS +3
203\fBPrimaryServer
204\fR
205.RE
206.RS +9
207The name of the virtual shared disk server node. This can be in any
208recognizable form.
209.RE
210.PP
211.RS +3
212\fBBackupServer
213\fR
214.RE
215.RS +9
216The backup server name. This can be specified in any recognizable
217form or allowed to default to none.
218.RE
219.PP
220.RS +3
221\fB\fIDisk Usage\fR
222\fR
223.RE
224.RS +9
225Specify a disk usage or accept the default (see
226\fIGeneral Parallel File System: Concepts,
227Planning, and Installation Guide\fR and search on \fIrecoverability
228considerations\fR). This field is ignored by the \fBmmcrvsd\fR
229command and is passed unchanged to the output descriptor file produced by the
230\fBmmcrvsd\fR command.
231.PP
232.RS +3
233\fBdataAndMetadata
234\fR
235.RE
236.RS +9
237Indicates that the disk contains both data and metadata. This is
238the default.
239.RE
240.PP
241.RS +3
242\fBdataOnly
243\fR
244.RE
245.RS +9
246Indicates that the disk contains data and does not contain
247metadata.
248.RE
249.PP
250.RS +3
251\fBmetadataOnly
252\fR
253.RE
254.RS +9
255Indicates that the disk contains metadata and does not contain
256data.
257.RE
258.PP
259.RS +3
260\fBdescOnly
261\fR
262.RE
263.RS +9
264Indicates that the disk contains no data and no file metadata. Such
265a disk is used solely to keep a copy of the file system descriptor, and can be
266used as a third failure group in certain disaster recovery
267configurations. For more information, see
268\fIGeneral Parallel File System: Advanced
269Administration\fR and search on \fISynchronous mirroring utilizing
270GPFS replication\fR.
271.RE
272.PP
273.PP
274\fBDisk usage considerations: \fR
275.RS +3
276.sp
277.HP 3
2781. The \fIDiskUsage\fR parameter is not utilized by the \fBmmcrvsd\fR
279command but is copied intact to the output file that the command
280produces. The output file may then be used as input to the
281\fBmmcrnsd\fR command.
282.sp
283.HP 3
2842. RAID devices are not well-suited for performing small block writes.
285Since GPFS metadata writes are often smaller than a full block, you may find
286using non-RAID devices for GPFS metadata better for performance.
287.RE
288.RE
289.PP
290.RS +3
291\fB\fIFailure Group\fR
292\fR
293.RE
294.RS +9
295A number identifying the failure group to which this disk belongs.
296You can specify any value from -1 (where -1 indicates that the disk has no
297point of failure in common with any other disk) to 4000. All disks that
298have a common point of failure, such as all disks that are attached to the
299same virtual shared disk server node, should be placed in the same failure
300group. The value is passed unchanged to the output descriptor file
301produced by the \fBmmcrvsd\fR command. If you do not specify a
302failure group, a failure group will be assigned later by the
303\fBmmcrnsd\fR command.
304.RE
305.PP
306.RS +3
307\fB\fIDesiredName\fR
308\fR
309.RE
310.RS +9
311Specify the name you desire for the virtual shared disk to be
312created. This name must not already be used as another GPFS or AIX disk
313name, and it must not begin with the reserved string 'gpfs'.
314.RS +3
315\fBNote:\fR
316.RE
317.RS +9
318This name can contain only the following characters: 'A'
319through 'Z', 'a' through 'z', '0'
320through '9', or '_' (the underscore). All other
321characters are not valid. The maximum size of this name is 13
322characters.
323.RE
324.RE
325.PP
326.RS +3
327\fB\fIStoragePool\fR
328\fR
329.RE
330.RS +9
331Specifies the name of the storage pool that the NSD is assigned to.
332This field is ignored by the \fBmmcrvsd\fR command, and is passed unchanged
333to the output descriptor file produced by the \fBmmcrvsd\fR
334command.
335.RE
336.RE
337.SH "Options"
338.PP
339.RS +3
340\fB-f \fIFanoutNumber\fR
341\fR
342.RE
343.RS +9
344The maximum number of concurrent nodes to communicate with during parallel
345operations. The default value is 10.
346.RE
347.PP
348.RS +3
349\fB-y
350\fR
351.RE
352.RS +9
353Specifies no prompting for any queries the command may produce. All
354default values are accepted.
355.RE
356.PP
357.RS +3
358\fB-c
359\fR
360.RE
361.RS +9
362Specifies to create Concurrent Virtual Shared Disks. This option is
363valid only for disk descriptors that specify both a primary and a backup
364virtual shared disk server.
365.RE
366.SH "Exit status"
367.PP
368.PP
369.RS +3
370\fB0
371\fR
372.RE
373.RS +9
374Successful completion.
375.RE
376.PP
377.RS +3
378\fBnonzero
379\fR
380.RE
381.RS +9
382A failure has occurred.
383.RE
384.SH "Security"
385.PP
386You must have root authority to run the \fBmmcrvsd\fR command.
387.PP
388You may issue the \fBmmcrvsd\fR command from any node in the GPFS
389cluster.
390.PP
391When using the \fBrcp\fR and \fBrsh\fR commands for remote
392communication, a properly configured \fB.rhosts\fR file must exist
393in the root user's home directory, on each node in the GPFS
394cluster. If you have designated the use of a different remote
395communication program on either the
396\fBmmcrcluster\fR
397or the
398\fBmmchcluster\fR command, you must
399ensure:
400.RS +3
401.HP 3
4021. Proper authorization is granted to all nodes in the GPFS cluster.
403.HP 3
4042. The nodes in the GPFS cluster can communicate without the use of a
405password, and without any extraneous messages.
406.RE
407.SH "Examples"
408.PP
409To create a virtual shared disk with the descriptor file \fBvsdesc\fR
410containing:
411.sp
412.nf
413 hdisk2:k145n01:k145n02:dataOnly:4
414 hdisk3:k145n06::dataAndMetadata:5:ABC
415.fi
416.sp
417.PP
418These descriptors translate as:
419.PP
420.RS +3
421\fBDisk Name
422\fR
423.RE
424.RS +9
425hdisk2
426.RE
427.PP
428.RS +3
429\fBServer Name
430\fR
431.RE
432.RS +9
433k145n01
434.RE
435.PP
436.RS +3
437\fBBackup Server Name
438\fR
439.RE
440.RS +9
441k145n02
442.RE
443.PP
444.RS +3
445\fBDisk Usage
446\fR
447.RE
448.RS +9
449\fBdataOnly\fR
450.RE
451.PP
452.RS +3
453\fBFailure Group
454\fR
455.RE
456.RS +9
4574
458.RE
459.PP
460.RS +3
461\fBDesired Name
462\fR
463.RE
464.RS +9
465Name defaults to gpfs20vsd
466.RE
467.PP
468and
469.PP
470.RS +3
471\fBDisk Name
472\fR
473.RE
474.RS +9
475hdisk3
476.RE
477.PP
478.RS +3
479\fBServer Name
480\fR
481.RE
482.RS +9
483k145n06
484.RE
485.PP
486.RS +3
487\fBBackup Server Name
488\fR
489.RE
490.RS +9
491none
492.RE
493.PP
494.RS +3
495\fBDisk Usage
496\fR
497.RE
498.RS +9
499\fBdataAndMetadata\fR
500.RE
501.PP
502.RS +3
503\fBFailure Group
504\fR
505.RE
506.RS +9
5075
508.RE
509.PP
510.RS +3
511\fBDesired Name
512\fR
513.RE
514.RS +9
515ABC
516.RE
517.PP
518The low level components of the virtual shared disk \fBgpfs20vsd\fR are
519created:
520.PP
521.RS +3
522\fBgpfs20gvg
523\fR
524.RE
525.RS +9
526global volume group
527.RE
528.PP
529.RS +3
530\fBgpfs20lv
531\fR
532.RE
533.RS +9
534local logical volume
535.RE
536.PP
537.RS +3
538\fBgpfs20vg
539\fR
540.RE
541.RS +9
542local volume group
543.RE
544.PP
545The low level components of the virtual shared disk \fBABC\fR are
546created:
547.PP
548.RS +3
549\fBABCgvg
550\fR
551.RE
552.RS +9
553global volume group
554.RE
555.PP
556.RS +3
557\fBABClv
558\fR
559.RE
560.RS +9
561local logical volume
562.RE
563.PP
564.RS +3
565\fBABCvg
566\fR
567.RE
568.RS +9
569local volume group
570.RE
571.SH "See also"
572.PP
573mmcrnsd Command
574.SH "Location"
575.PP
576\fB/usr/lpp/mmfs/bin\fR
577.PP
Note: See TracBrowser for help on using the repository browser.