root / man / gnt-cluster.sgml @ 3fc175f0
History | View | Annotate | Download (15.5 kB)
1 |
<!doctype refentry PUBLIC "-//OASIS//DTD DocBook V4.1//EN" [ |
---|---|
2 |
|
3 |
<!-- Fill in your name for FIRSTNAME and SURNAME. --> |
4 |
<!-- Please adjust the date whenever revising the manpage. --> |
5 |
<!ENTITY dhdate "<date>December 12, 2007</date>"> |
6 |
<!-- SECTION should be 1-8, maybe w/ subsection other parameters are |
7 |
allowed: see man(7), man(1). --> |
8 |
<!ENTITY dhsection "<manvolnum>8</manvolnum>"> |
9 |
<!ENTITY dhucpackage "<refentrytitle>gnt-cluster</refentrytitle>"> |
10 |
<!ENTITY dhpackage "gnt-cluster"> |
11 |
|
12 |
<!ENTITY debian "<productname>Debian</productname>"> |
13 |
<!ENTITY gnu "<acronym>GNU</acronym>"> |
14 |
<!ENTITY gpl "&gnu; <acronym>GPL</acronym>"> |
15 |
<!ENTITY footer SYSTEM "footer.sgml"> |
16 |
]> |
17 |
|
18 |
<refentry> |
19 |
<refentryinfo> |
20 |
<copyright> |
21 |
<year>2006</year> |
22 |
<year>2007</year> |
23 |
<holder>Google Inc.</holder> |
24 |
</copyright> |
25 |
&dhdate; |
26 |
</refentryinfo> |
27 |
<refmeta> |
28 |
&dhucpackage; |
29 |
|
30 |
&dhsection; |
31 |
<refmiscinfo>ganeti 1.2</refmiscinfo> |
32 |
</refmeta> |
33 |
<refnamediv> |
34 |
<refname>&dhpackage;</refname> |
35 |
|
36 |
<refpurpose>ganeti administration, cluster-wide</refpurpose> |
37 |
</refnamediv> |
38 |
<refsynopsisdiv> |
39 |
<cmdsynopsis> |
40 |
<command>&dhpackage; </command> |
41 |
|
42 |
<arg choice="req">command</arg> |
43 |
<arg>arguments...</arg> |
44 |
</cmdsynopsis> |
45 |
</refsynopsisdiv> |
46 |
<refsect1> |
47 |
<title>DESCRIPTION</title> |
48 |
|
49 |
<para> |
50 |
The <command>&dhpackage;</command> is used for cluster-wide |
51 |
administration in the ganeti system. |
52 |
</para> |
53 |
|
54 |
</refsect1> |
55 |
<refsect1> |
56 |
<title>COMMANDS</title> |
57 |
|
58 |
<refsect2> |
59 |
<title>ADD-TAGS</title> |
60 |
|
61 |
<cmdsynopsis> |
62 |
<command>add-tags</command> |
63 |
<arg choice="opt">--from <replaceable>file</replaceable></arg> |
64 |
<arg choice="req" |
65 |
rep="repeat"><replaceable>tag</replaceable></arg> |
66 |
</cmdsynopsis> |
67 |
|
68 |
<para> |
69 |
Add tags to the cluster. If any of the tags contains invalid |
70 |
characters, the entire operation will abort. |
71 |
</para> |
72 |
|
73 |
<para> |
74 |
If the <option>--from</option> option is given, the list of |
75 |
tags will be extended with the contents of that file (each |
76 |
line becomes a tag). In this case, there is not need to pass |
77 |
tags on the command line (if you do, both sources will be |
78 |
used). A file name of - will be interpreted as stdin. |
79 |
</para> |
80 |
</refsect2> |
81 |
|
82 |
<refsect2> |
83 |
<title>COMMAND</title> |
84 |
|
85 |
<cmdsynopsis> |
86 |
<command>command</command> |
87 |
<arg>-n <replaceable>node</replaceable></arg> |
88 |
<arg choice="req"><replaceable>command</replaceable></arg> |
89 |
</cmdsynopsis> |
90 |
|
91 |
<para> |
92 |
Executes a command on all nodes. If the option |
93 |
<option>-n</option> is not given, the command will be executed |
94 |
on all nodes, otherwise it will be executed only on the |
95 |
node(s) specified. Use the option multiple times for running |
96 |
it on multiple nodes, like: |
97 |
|
98 |
<screen> |
99 |
# gnt-cluster command -n node1.example.com -n node2.example.com date |
100 |
</screen> |
101 |
|
102 |
</para> |
103 |
|
104 |
<para> |
105 |
The command is executed serially on the selected nodes. If the |
106 |
master node is present in the list, the command will be |
107 |
executed last on the master. Regarding the other nodes, the |
108 |
execution order is somewhat alphabetic (it's smarter so that |
109 |
node2.example.com will be earlier than node10.example.com but |
110 |
after node1.example.com). |
111 |
</para> |
112 |
|
113 |
<para> |
114 |
So given the node names node1, node2, node3, node10, node11, |
115 |
with node3 being the master, the order will be: node1, node2, |
116 |
node10, node11, node3. |
117 |
</para> |
118 |
|
119 |
<para> |
120 |
The command is constructed by concatenating all other command |
121 |
line arguments. For example, to list the contents of the |
122 |
<filename class="directory">/etc</filename> directory on all |
123 |
nodes, run: |
124 |
|
125 |
<screen> |
126 |
# gnt-cluster command ls -l /etc |
127 |
</screen> |
128 |
|
129 |
and the command which will be executed will be |
130 |
<computeroutput>"ls -l /etc"</computeroutput> |
131 |
</para> |
132 |
</refsect2> |
133 |
|
134 |
<refsect2> |
135 |
<title>COPYFILE</title> |
136 |
|
137 |
<cmdsynopsis> |
138 |
<command>copyfile</command> |
139 |
<arg>-n <replaceable>node</replaceable></arg> |
140 |
<arg choice="req"><replaceable>file</replaceable></arg> |
141 |
</cmdsynopsis> |
142 |
|
143 |
<para> |
144 |
Copies a file to all or to some nodes. The argument specifies |
145 |
the source file (on the current system), the |
146 |
<option>-n</option> argument specifies the target node, or |
147 |
nodes if the option is given multiple times. If |
148 |
<option>-n</option> is not given at all, the file will be |
149 |
copied to all nodes. |
150 |
|
151 |
Example: |
152 |
<screen> |
153 |
# gnt-cluster -n node1.example.com -n node2.example.com copyfile /tmp/test |
154 |
</screen> |
155 |
|
156 |
This will copy the file <filename>/tmp/test</filename> from |
157 |
the current node to the two named nodes. |
158 |
</para> |
159 |
</refsect2> |
160 |
|
161 |
<refsect2> |
162 |
<title>DESTROY</title> |
163 |
|
164 |
<cmdsynopsis> |
165 |
<command>destroy</command> |
166 |
<arg choice="req">--yes-do-it</arg> |
167 |
</cmdsynopsis> |
168 |
|
169 |
<para> |
170 |
Remove all configuration files related to the cluster, so that |
171 |
a <command>gnt-cluster init</command> can be done again |
172 |
afterwards. |
173 |
</para> |
174 |
|
175 |
<para> |
176 |
Since this is a dangerous command, you are required to pass |
177 |
the argument <replaceable>--yes-do-it.</replaceable> |
178 |
</para> |
179 |
</refsect2> |
180 |
|
181 |
<refsect2> |
182 |
<title>GETMASTER</title> |
183 |
|
184 |
<cmdsynopsis> |
185 |
<command>getmaster</command> |
186 |
</cmdsynopsis> |
187 |
|
188 |
<para> |
189 |
Displays the current master node. |
190 |
</para> |
191 |
</refsect2> |
192 |
|
193 |
<refsect2> |
194 |
<title>INFO</title> |
195 |
|
196 |
<cmdsynopsis> |
197 |
<command>info</command> |
198 |
</cmdsynopsis> |
199 |
|
200 |
<para> |
201 |
Shows runtime cluster information: cluster name, architecture |
202 |
(32 or 64 bit), master node, node list and instance list. |
203 |
</para> |
204 |
</refsect2> |
205 |
|
206 |
<refsect2> |
207 |
<title>INIT</title> |
208 |
|
209 |
<cmdsynopsis> |
210 |
<command>init</command> |
211 |
<arg>-s <replaceable>secondary_ip</replaceable></arg> |
212 |
<arg>-b <replaceable>bridge</replaceable></arg> |
213 |
<arg>-t <replaceable>hypervisor-type</replaceable></arg> |
214 |
<arg>-g <replaceable>vg-name</replaceable></arg> |
215 |
<arg>--master-netdev <replaceable>vg-name</replaceable></arg> |
216 |
<arg>-m <replaceable>mac-prefix</replaceable></arg> |
217 |
<arg>--no-lvm-storage</arg> |
218 |
<arg choice="req"><replaceable>clustername</replaceable></arg> |
219 |
</cmdsynopsis> |
220 |
|
221 |
<para> |
222 |
This commands is only run once initially on the first node of |
223 |
the cluster. It will initialize the cluster configuration and |
224 |
setup ssh-keys and more. |
225 |
</para> |
226 |
|
227 |
<para> |
228 |
Note that the <replaceable>clustername</replaceable> is not |
229 |
any random name. It has to be resolvable to an IP address |
230 |
using DNS, and it is best if you give the fully-qualified |
231 |
domain name. Also it is advisable to reserve a per-cluster |
232 |
unique name, and not try to reuse one of the nodes' names. |
233 |
</para> |
234 |
|
235 |
<para> |
236 |
The cluster can run in two modes: single-home or |
237 |
dual-homed. In the first case, all traffic (both public |
238 |
traffic, inter-node traffic and data replication traffic) goes |
239 |
over the same interface. In the dual-homed case, the data |
240 |
replication traffic goes over the second network. The |
241 |
<option>-s</option> option here marks the cluster as |
242 |
dual-homed and its parameter represents this node's address on |
243 |
the second network. If you initialise the cluster with |
244 |
<option>-s</option>, all nodes added must have a secondary IP |
245 |
as well. |
246 |
</para> |
247 |
|
248 |
<para> |
249 |
Note that for Ganeti it doesn't matter if the secondary |
250 |
network is actually a separate physical network, or is done |
251 |
using tunneling, etc. For performance reasons, it's |
252 |
recommended to use a separate network, of course. |
253 |
</para> |
254 |
|
255 |
<para> |
256 |
The <option>-b</option> option specifies the default bridge |
257 |
for instances. |
258 |
</para> |
259 |
|
260 |
<para> |
261 |
The <option>-t</option> allows to set the hypervisor type of |
262 |
the cluster. Available hypervisor types are: xen-3.0, fake and |
263 |
xen-hvm3.1. The default is the xen-3.0 hypervisor. |
264 |
Note that if you init the cluster with hypervisor-type |
265 |
xen-hvm3.1 you also need to provide the cluster VNC password |
266 |
file <filename>/etc/ganeti/vnc-cluster-password</filename> and |
267 |
the HVM boot ISO image |
268 |
<filename>/srv/ganeti/iso/hvm-install.iso</filename> because |
269 |
instances created by the experimental HVM support require them. |
270 |
</para> |
271 |
|
272 |
<para> |
273 |
The <option>-g</option> option will let you specify a volume group |
274 |
different than 'xenvg' for ganeti to use when creating instance disks. |
275 |
This volume group must have the same name on all nodes. Once the |
276 |
cluster is initialized this can be altered by using the |
277 |
<command>modify</command> command. If you don't want to use lvm |
278 |
storage at all use the <option>--no-lvm-storage</option> option. |
279 |
Once the cluster is initialized you can change this setup with the |
280 |
<command>modify</command> command. |
281 |
</para> |
282 |
|
283 |
<para> |
284 |
The <option>--master-netdev</option> option is useful for specifying a |
285 |
different interface on which the master will activate its IP address. |
286 |
It's important that all nodes have this interface because you'll need |
287 |
it for a master failover. |
288 |
</para> |
289 |
|
290 |
<para> |
291 |
The <option>-m</option> option will let you specify a three byte prefix |
292 |
under which the virtual MAC addresses of your instances will be |
293 |
generated. The prefix must be specified in the format XX:XX:XX and the |
294 |
default is aa:00:00. |
295 |
</para> |
296 |
|
297 |
<para> |
298 |
The <option>--no-lvm-storage</option> allows you to initialize the |
299 |
cluster without lvm support. This means that only instances using |
300 |
files as storage backend will be possible to create. Once the cluster |
301 |
is initialized you can change this setup with the |
302 |
<command>modify</command> command. |
303 |
</para> |
304 |
</refsect2> |
305 |
|
306 |
<refsect2> |
307 |
<title>LIST-TAGS</title> |
308 |
|
309 |
<cmdsynopsis> |
310 |
<command>list-tags</command> |
311 |
</cmdsynopsis> |
312 |
|
313 |
<para>List the tags of the cluster.</para> |
314 |
</refsect2> |
315 |
|
316 |
<refsect2> |
317 |
<title>MASTERFAILOVER</title> |
318 |
|
319 |
<cmdsynopsis> |
320 |
<command>masterfailover</command> |
321 |
</cmdsynopsis> |
322 |
|
323 |
<para> |
324 |
Failover the master role to the current node. |
325 |
</para> |
326 |
</refsect2> |
327 |
|
328 |
<refsect2> |
329 |
<title>MODIFY</title> |
330 |
|
331 |
<cmdsynopsis> |
332 |
<command>modify</command> |
333 |
<arg choice="opt">-g <replaceable>vg-name</replaceable></arg> |
334 |
<arg choice="opt">--no-lvm-storage</arg> |
335 |
</cmdsynopsis> |
336 |
|
337 |
<para> |
338 |
Modify the options for the cluster. |
339 |
</para> |
340 |
|
341 |
<para> |
342 |
The <option>-g</option> and <option>--no-lvm-stoarge</option> are |
343 |
described in the <command>init</command> command. |
344 |
</para> |
345 |
</refsect2> |
346 |
|
347 |
<refsect2> |
348 |
<title>REMOVE-TAGS</title> |
349 |
|
350 |
<cmdsynopsis> |
351 |
<command>remove-tags</command> |
352 |
<arg choice="opt">--from <replaceable>file</replaceable></arg> |
353 |
<arg choice="req" |
354 |
rep="repeat"><replaceable>tag</replaceable></arg> |
355 |
</cmdsynopsis> |
356 |
|
357 |
<para> |
358 |
Remove tags from the cluster. If any of the tags are not |
359 |
existing on the cluster, the entire operation will abort. |
360 |
</para> |
361 |
|
362 |
<para> |
363 |
If the <option>--from</option> option is given, the list of |
364 |
tags will be extended with the contents of that file (each |
365 |
line becomes a tag). In this case, there is not need to pass |
366 |
tags on the command line (if you do, both sources will be |
367 |
used). A file name of - will be interpreted as stdin. |
368 |
</para> |
369 |
</refsect2> |
370 |
|
371 |
<refsect2> |
372 |
<title>RENAME</title> |
373 |
|
374 |
<cmdsynopsis> |
375 |
<command>rename</command> |
376 |
<arg>-f</arg> |
377 |
<arg choice="req"><replaceable>name</replaceable></arg> |
378 |
</cmdsynopsis> |
379 |
|
380 |
<para> |
381 |
Renames the cluster and in the process updates the master IP |
382 |
address to the one the new name resolves to. At least one of |
383 |
either the name or the IP address must be different, otherwise |
384 |
the operation will be aborted. |
385 |
</para> |
386 |
|
387 |
<para> |
388 |
Note that since this command can be dangerous (especially when |
389 |
run over SSH), the command will require confirmation unless |
390 |
run with the <option>-f</option> option. |
391 |
</para> |
392 |
</refsect2> |
393 |
|
394 |
<refsect2> |
395 |
<title>SEARCH-TAGS</title> |
396 |
|
397 |
<cmdsynopsis> |
398 |
<command>search-tags</command> |
399 |
<arg choice="req"><replaceable>pattern</replaceable></arg> |
400 |
</cmdsynopsis> |
401 |
|
402 |
<para> |
403 |
Searches the tags on all objects in the cluster (the cluster |
404 |
itself, the nodes and the instances) for a given pattern. The |
405 |
pattern is interpreted as a regular expression and a search |
406 |
will be done on it (i.e. the given pattern is not anchored to |
407 |
the beggining of the string; if you want that, prefix the |
408 |
pattern with <literal>^</literal>). |
409 |
</para> |
410 |
|
411 |
<para> |
412 |
If no tags are matching the pattern, the exit code of the |
413 |
command will be one. If there is at least one match, the exit |
414 |
code will be zero. Each match is listed on one line, the |
415 |
object and the tag separated by a space. The cluster will be |
416 |
listed as <filename>/cluster</filename>, a node will be listed |
417 |
as |
418 |
<filename>/nodes/<replaceable>name</replaceable></filename>, |
419 |
and an instance as |
420 |
<filename>/instances/<replaceable>name</replaceable></filename>. |
421 |
Example: |
422 |
</para> |
423 |
<screen> |
424 |
# gnt-cluster search time |
425 |
/cluster ctime:2007-09-01 |
426 |
/nodes/node1.example.com mtime:2007-10-04 |
427 |
</screen> |
428 |
</refsect2> |
429 |
|
430 |
<refsect2> |
431 |
<title>VERIFY</title> |
432 |
|
433 |
<cmdsynopsis> |
434 |
<command>verify</command> |
435 |
<arg choice="opt">--no-nplus1-mem</arg> |
436 |
</cmdsynopsis> |
437 |
|
438 |
<para> |
439 |
Verify correctness of cluster configuration. This is safe with |
440 |
respect to running instances, and incurs no downtime of the |
441 |
instances. |
442 |
</para> |
443 |
|
444 |
<para> |
445 |
If the <option>--no-nplus1-mem</option> option is given, ganeti won't |
446 |
check whether if it loses a node it can restart all the instances on |
447 |
their secondaries (and report an error otherwise). |
448 |
</para> |
449 |
</refsect2> |
450 |
|
451 |
<refsect2> |
452 |
<title>VERIFY-DISKS</title> |
453 |
|
454 |
<cmdsynopsis> |
455 |
<command>verify-disks</command> |
456 |
</cmdsynopsis> |
457 |
|
458 |
<para> |
459 |
The command checks which instances have degraded DRBD disks |
460 |
and activates the disks of those instances. |
461 |
</para> |
462 |
|
463 |
<para> |
464 |
This command is run from the <command>ganeti-watcher</command> |
465 |
tool, which also has a different, complementary algorithm for |
466 |
doing this check. Together, these two should ensure that DRBD |
467 |
disks are kept consistent. |
468 |
</para> |
469 |
</refsect2> |
470 |
|
471 |
<refsect2> |
472 |
<title>VERSION</title> |
473 |
|
474 |
<cmdsynopsis> |
475 |
<command>version</command> |
476 |
</cmdsynopsis> |
477 |
|
478 |
<para> |
479 |
Show the cluster version. |
480 |
</para> |
481 |
</refsect2> |
482 |
|
483 |
</refsect1> |
484 |
|
485 |
&footer; |
486 |
|
487 |
</refentry> |
488 |
|
489 |
<!-- Keep this comment at the end of the file |
490 |
Local variables: |
491 |
mode: sgml |
492 |
sgml-omittag:t |
493 |
sgml-shorttag:t |
494 |
sgml-minimize-attributes:nil |
495 |
sgml-always-quote-attributes:t |
496 |
sgml-indent-step:2 |
497 |
sgml-indent-data:t |
498 |
sgml-parent-document:nil |
499 |
sgml-default-dtd-file:nil |
500 |
sgml-exposed-tags:nil |
501 |
sgml-local-catalogs:nil |
502 |
sgml-local-ecat-files:nil |
503 |
End: |
504 |
--> |