Revision f69dab6d
b/man/ganeti-watcher.sgml | ||
---|---|---|
50 | 50 |
</para> |
51 | 51 |
|
52 | 52 |
<para> |
53 |
Its function is to try to keep running all instances which are |
|
54 |
marked as <emphasis>up</emphasis> in the configuration file, by |
|
55 |
trying to start them a limited number of times. |
|
53 |
Its primary function is to try to keep running all instances |
|
54 |
which are marked as <emphasis>up</emphasis> in the configuration |
|
55 |
file, by trying to start them a limited number of times. |
|
56 |
</para> |
|
57 |
|
|
58 |
<para> |
|
59 |
Its other function is to <quote>repair</quote> DRBD links by |
|
60 |
reactivating the block devices of instances which have |
|
61 |
secondaries on nodes that have been rebooted. |
|
56 | 62 |
</para> |
57 | 63 |
|
58 | 64 |
<para>In order to prevent piling up commands, all the |
... | ... | |
72 | 78 |
|
73 | 79 |
</refsect1> |
74 | 80 |
|
75 |
<refsect1> |
|
76 |
<title>KNOWN BUGS</title> |
|
77 |
|
|
78 |
<para> |
|
79 |
Due to the way we initialize DRBD peers, restarting a secondary |
|
80 |
node for an instance will cause the DRBD endpoints on that node |
|
81 |
to disappear, thus all instances which have that node as a |
|
82 |
secondary will lose redundancy. The watcher does not detect this |
|
83 |
situation. The workaround is to manually run |
|
84 |
<command>gnt-instance activate-disks</command> for all the |
|
85 |
affected instances. |
|
86 |
</para> |
|
87 |
</refsect1> |
|
88 |
|
|
89 | 81 |
&footer; |
90 | 82 |
|
91 | 83 |
</refentry> |
b/man/ganeti.sgml | ||
---|---|---|
51 | 51 |
<para> |
52 | 52 |
The ganeti software manages physical nodes and virtual instances |
53 | 53 |
of a cluster based on a virtualization software. The current |
54 |
version (1.2) supports Xen 3.0. |
|
54 |
version (1.2) supports Xen 3.0 (also tested with 3.1).
|
|
55 | 55 |
</para> |
56 | 56 |
|
57 | 57 |
</refsect1> |
b/man/gnt-backup.sgml | ||
---|---|---|
66 | 66 |
</cmdsynopsis> |
67 | 67 |
|
68 | 68 |
<para> |
69 |
Exports an instance to the target node. All the instance |
|
70 |
data and its configuration will be exported under the |
|
71 |
/srv/ganeti/exports/instance directory on the target node. |
|
69 |
Exports an instance to the target node. All the instance data |
|
70 |
and its configuration will be exported under the |
|
71 |
/srv/ganeti/exports/<replaceable>instance</replaceable> |
|
72 |
directory on the target node. |
|
72 | 73 |
</para> |
73 | 74 |
|
74 | 75 |
<para> |
... | ... | |
102 | 103 |
<arg>plain</arg> |
103 | 104 |
<arg>local_raid1</arg> |
104 | 105 |
<arg>remote_raid1</arg> |
106 |
<arg>drbd</arg> |
|
105 | 107 |
</group> |
106 | 108 |
</arg> |
109 |
<sbr> |
|
107 | 110 |
<arg choice="req">--src-node=<replaceable>source-node</replaceable></arg> |
108 | 111 |
<arg choice="req">--src-dir=<replaceable>source-dir</replaceable></arg> |
109 | 112 |
<arg choice="req"><replaceable>instance</replaceable></arg> |
... | ... | |
119 | 122 |
|
120 | 123 |
<para> |
121 | 124 |
The <option>-s</option> option specifies the disk size for |
122 |
the instance, in gigibytes (defaults to 20 GiB).
|
|
125 |
the instance, in gibibytes (defaults to 20 GiB).
|
|
123 | 126 |
</para> |
124 | 127 |
|
125 | 128 |
<para> |
... | ... | |
130 | 133 |
|
131 | 134 |
<para> |
132 | 135 |
The <option>-m</option> option specifies the memory size for |
133 |
the instance, in megibytes (defaults to 128 MiB).
|
|
136 |
the instance, in mebibytes (defaults to 128 MiB).
|
|
134 | 137 |
</para> |
135 | 138 |
|
136 | 139 |
<para> |
... | ... | |
171 | 174 |
<term>remote_raid1</term> |
172 | 175 |
<listitem> |
173 | 176 |
<para> |
174 |
Disk devices will be md raid1 arrays with one component (so |
|
175 |
it's not actually raid1): a drbd device between the instance's |
|
176 |
primary node and the node given by the second value of the |
|
177 |
Disk devices will be md raid1 arrays with one |
|
178 |
component (so it's not actually raid1): a drbd (0.7.x) |
|
179 |
device between the instance's primary node and the |
|
180 |
node given by the second value of the |
|
177 | 181 |
<option>--node</option> option. |
178 | 182 |
</para> |
179 | 183 |
</listitem> |
180 | 184 |
</varlistentry> |
185 |
<varlistentry> |
|
186 |
<term>drbd</term> |
|
187 |
<listitem> |
|
188 |
<para> |
|
189 |
Disk devices will be drbd (version 8.x) on top of lvm |
|
190 |
volumes. They are equivalent in functionality to |
|
191 |
<replaceable>remote_raid1</replaceable>, but are |
|
192 |
recommended for new instances (if you have drbd 8.x |
|
193 |
installed). |
|
194 |
</para> |
|
195 |
</listitem> |
|
196 |
</varlistentry> |
|
181 | 197 |
</variablelist> |
182 | 198 |
</para> |
183 | 199 |
|
... | ... | |
208 | 224 |
</cmdsynopsis> |
209 | 225 |
|
210 | 226 |
<para> |
211 |
Lists the exports currently available in the default directory in all |
|
212 |
the nodes of the current cluster, or optionally only a subset of them |
|
213 |
specified by the <option>--nodes</option> option. |
|
227 |
Lists the exports currently available in the default directory |
|
228 |
in all the nodes of the current cluster, or optionally only a |
|
229 |
subset of them specified by the <option>--nodes</option> |
|
230 |
option (use this option multiple times to select multiple |
|
231 |
nodes). |
|
214 | 232 |
</para> |
215 | 233 |
|
234 |
<para> |
|
235 |
Example: |
|
236 |
<screen> |
|
237 |
# gnt-backup list --nodes node1 --nodes node2 |
|
238 |
</screen> |
|
216 | 239 |
</refsect2> |
217 | 240 |
|
218 | 241 |
</refsect1> |
b/man/gnt-cluster.sgml | ||
---|---|---|
148 | 148 |
|
149 | 149 |
<cmdsynopsis> |
150 | 150 |
<command>destroy</command> |
151 |
<arg choice="req">--yes-do-it</arg> |
|
151 | 152 |
</cmdsynopsis> |
152 | 153 |
|
153 | 154 |
<para> |
... | ... | |
155 | 156 |
a <command>gnt-cluster init</command> can be done again |
156 | 157 |
afterwards. |
157 | 158 |
</para> |
159 |
|
|
160 |
<para> |
|
161 |
Since this is a dangerous command, you are required to pass |
|
162 |
the argument <replaceable>--yes-do-it.</replaceable> |
|
163 |
</para> |
|
158 | 164 |
</refsect2> |
159 | 165 |
|
160 | 166 |
<refsect2> |
b/man/gnt-instance.sgml | ||
---|---|---|
74 | 74 |
<arg>plain</arg> |
75 | 75 |
<arg>local_raid1</arg> |
76 | 76 |
<arg>remote_raid1</arg> |
77 |
<arg>drbd</arg> |
|
77 | 78 |
</group> |
78 | 79 |
</arg> |
79 | 80 |
<sbr> |
... | ... | |
155 | 156 |
<term>remote_raid1</term> |
156 | 157 |
<listitem> |
157 | 158 |
<para> |
158 |
Disk devices will be md raid1 arrays with one component (so |
|
159 |
it's not actually raid1): a drbd device between the |
|
160 |
instance's primary node and the node given by the second |
|
161 |
value of the <option>--node</option> option. |
|
159 |
Disk devices will be md raid1 arrays with one |
|
160 |
component (so it's not actually raid1): a drbd |
|
161 |
(0.7.x) device between the instance's primary node |
|
162 |
and the node given by the second value of the |
|
163 |
<option>--node</option> option. |
|
164 |
</para> |
|
165 |
</listitem> |
|
166 |
</varlistentry> |
|
167 |
<varlistentry> |
|
168 |
<term>drbd</term> |
|
169 |
<listitem> |
|
170 |
<para> |
|
171 |
Disk devices will be drbd (version 8.x) on top of |
|
172 |
lvm volumes. They are equivalent in functionality to |
|
173 |
<replaceable>remote_raid1</replaceable>, but are |
|
174 |
recommended for new instances (if you have drbd 8.x |
|
175 |
installed). |
|
162 | 176 |
</para> |
163 | 177 |
</listitem> |
164 | 178 |
</varlistentry> |
b/man/gnt-node.sgml | ||
---|---|---|
382 | 382 |
|
383 | 383 |
<cmdsynopsis> |
384 | 384 |
<command>volumes</command> |
385 |
<arg>--no-headers</arg> |
|
386 |
<arg>--human-readable</arg> |
|
387 |
<arg>--separator=<replaceable>SEPARATOR</replaceable></arg> |
|
388 |
<arg>--output=<replaceable>FIELDS</replaceable></arg> |
|
389 |
<sbr> |
|
385 | 390 |
<arg rep="repeat"><replaceable>node</replaceable></arg> |
386 | 391 |
</cmdsynopsis> |
387 | 392 |
|
... | ... | |
391 | 396 |
</para> |
392 | 397 |
|
393 | 398 |
<para> |
399 |
The <option>--no-headers</option> option will skip the initial |
|
400 |
header line. The <option>--separator</option> option takes an |
|
401 |
argument which denotes what will be used between the output |
|
402 |
fields. Both these options are to help scripting. |
|
403 |
</para> |
|
404 |
|
|
405 |
<para> |
|
406 |
The <option>-o</option> option takes a comma-separated list of |
|
407 |
output fields. The available fields and their meaning are: |
|
408 |
<variablelist> |
|
409 |
<varlistentry> |
|
410 |
<term>node</term> |
|
411 |
<listitem> |
|
412 |
<simpara>the node name on which the volume exists</simpara> |
|
413 |
</listitem> |
|
414 |
</varlistentry> |
|
415 |
<varlistentry> |
|
416 |
<term>phys</term> |
|
417 |
<listitem> |
|
418 |
<simpara>the physical drive (on which the LVM physical |
|
419 |
volume lives)</simpara> |
|
420 |
</listitem> |
|
421 |
</varlistentry> |
|
422 |
<varlistentry> |
|
423 |
<term>vg</term> |
|
424 |
<listitem> |
|
425 |
<simpara>the volume group name</simpara> |
|
426 |
</listitem> |
|
427 |
</varlistentry> |
|
428 |
<varlistentry> |
|
429 |
<term>name</term> |
|
430 |
<listitem> |
|
431 |
<simpara>the logical volume name</simpara> |
|
432 |
</listitem> |
|
433 |
</varlistentry> |
|
434 |
<varlistentry> |
|
435 |
<term>size</term> |
|
436 |
<listitem> |
|
437 |
<simpara>the logical volume size</simpara> |
|
438 |
</listitem> |
|
439 |
</varlistentry> |
|
440 |
<varlistentry> |
|
441 |
<term>instance</term> |
|
442 |
<listitem> |
|
443 |
<simpara>The name of the instance to which this volume |
|
444 |
belongs, or (in case it's an orphan volume) the |
|
445 |
character <quote>-</quote></simpara> |
|
446 |
</listitem> |
|
447 |
</varlistentry> |
|
448 |
</variablelist> |
|
449 |
</para> |
|
450 |
|
|
451 |
<para> |
|
394 | 452 |
Example: |
395 | 453 |
<screen> |
396 | 454 |
# gnt-node volumes node5.example.com |
Also available in: Unified diff