Revision 11a0e343
b/doc/glossary.rst | ||
---|---|---|
8 | 8 |
:sorted: |
9 | 9 |
|
10 | 10 |
ballooning |
11 |
A term describing runtime, dynamic changes to an instance's memory,
|
|
12 |
without having to reboot the instance. Depending on the hypervisor
|
|
13 |
and configuration, the changes need to be initiated manually, or
|
|
14 |
they can be automatically initiated by the hypervisor based on the
|
|
15 |
node and instances memory usage.
|
|
11 |
A term describing dynamic changes to an instance's memory while the instance
|
|
12 |
is running that don't require an instance reboot. Depending on the
|
|
13 |
hypervisor and configuration, changes may be automatically initiated by the
|
|
14 |
hypervisor (based on the memory usage of the node and instance), or may need
|
|
15 |
to be initiated manually.
|
|
16 | 16 |
|
17 | 17 |
BE parameter |
18 |
BE stands for *backend*. BE parameters are hypervisor-independent |
|
19 |
instance parameters such as the amount of RAM/virtual CPUs it has |
|
20 |
been allocated. |
|
18 |
BE stands for *backend*. BE parameters are hypervisor-independent instance |
|
19 |
parameters, such as the amount of RAM/virtual CPUs allocated to an instance. |
|
21 | 20 |
|
22 | 21 |
DRBD |
23 |
A block device driver that can be used to build RAID1 across the |
|
24 |
network or even shared storage, while using only locally-attached |
|
25 |
storage. |
|
22 |
A block device driver that can be used to build RAID1 across the network or |
|
23 |
across shared storage, while using only locally-attached storage. |
|
26 | 24 |
|
27 | 25 |
HV parameter |
28 |
HV stands for *hypervisor*. HV parameters are the ones that describe
|
|
29 |
the virtualization-specific aspects of the instance; for example,
|
|
30 |
what kernel to use to boot the instance (if any), or what emulation
|
|
31 |
model to use for the emulated hard drives.
|
|
26 |
HV stands for *hypervisor*. HV parameters describe the virtualization-
|
|
27 |
specific aspects of the instance. For example, a HV parameter might describe
|
|
28 |
what kernel (if any) to use to boot the instance or what emulation model to
|
|
29 |
use for the emulated hard drives. |
|
32 | 30 |
|
33 | 31 |
HVM |
34 |
Hardware virtualization mode, where the virtual machine is oblivious
|
|
35 |
to the fact that's being virtualized and all the hardware is
|
|
32 |
*Hardware Virtualization Mode*. In this mode, the virtual machine is
|
|
33 |
oblivious to the fact that it is virtualized and all its hardware is
|
|
36 | 34 |
emulated. |
37 | 35 |
|
38 | 36 |
LogicalUnit |
39 |
The code associated with an :term:`OpCode`, e.g. the code that
|
|
37 |
The code associated with an :term:`OpCode`; for example, the code that
|
|
40 | 38 |
implements the startup of an instance. |
41 | 39 |
|
42 | 40 |
LUXI |
43 |
Local UniX Interface. The IPC method over :manpage:`unix(7)` |
|
44 |
sockets used between the CLI tools/RAPI daemon and the master |
|
45 |
daemon. |
|
41 |
Local UniX Interface. The IPC method over :manpage:`unix(7)` sockets used |
|
42 |
between the CLI tools/RAPI daemon and the master daemon. |
|
46 | 43 |
|
47 | 44 |
OOB |
48 |
*Out of Band*. This term describes methods of accessing a machine |
|
49 |
(or parts of a machine) not via the usual network connection. For
|
|
50 |
example, accessing a remote server via a physical serial console or
|
|
51 |
via a virtual one IPMI counts as out of band access.
|
|
45 |
*Out of Band*. This term describes methods of accessing a machine (or parts
|
|
46 |
of a machine) by means other than the usual network connection. Examples
|
|
47 |
include accessing a remote server via a physical serial console or via a
|
|
48 |
virtual console. IPMI is also considered OOB access.
|
|
52 | 49 |
|
53 | 50 |
OpCode |
54 |
A data structure encapsulating a basic cluster operation; for |
|
55 |
example, start instance, add instance, etc.
|
|
51 |
A data structure encapsulating a basic cluster operation; for example: start
|
|
52 |
instance, add instance, etc. |
|
56 | 53 |
|
57 | 54 |
PVM |
58 |
(Xen) Para-virtualization mode, where the virtual machine knows it's
|
|
59 |
being virtualized and as such there is no need for hardware
|
|
60 |
emulation or virtualization.
|
|
55 |
(Xen) *Para-virtualization mode*. In this mode, the virtual machine is aware
|
|
56 |
that it is virtualized; therefore, there is no need for hardware emulation
|
|
57 |
or virtualization. |
|
61 | 58 |
|
62 | 59 |
SoR |
63 | 60 |
*State of Record*. Refers to values/properties that come from an |
64 |
authoritative configuration source. For example, the maximum VCPU |
|
65 |
over-subscription ratio is a *SoR* value, but the current |
|
66 |
over-subscription ration (based on how many instances live on the |
|
67 |
node) is a :term:`SoW` value. |
|
61 |
authoritative configuration source. For example, the maximum VCPU over- |
|
62 |
subscription ratio is a SoR value, but the current over-subscription ratio |
|
63 |
(based upon how many instances live on the node) is a :term:`SoW` value. |
|
68 | 64 |
|
69 | 65 |
SoW |
70 |
*State of the World*. Refers to values that describe directly the
|
|
71 |
world, as opposed to values that come from the
|
|
72 |
configuration. Contrast with :term:`SoR`.
|
|
66 |
*State of the World*. Refers to values that directly describe the world, as
|
|
67 |
opposed to values that come from the configuration (which are considered
|
|
68 |
:term:`SoR`).
|
|
73 | 69 |
|
74 | 70 |
tmem |
75 |
Xen Transcendent Memory |
|
76 |
(http://en.wikipedia.org/wiki/Transcendent_memory). It is a |
|
77 |
mechanism used by Xen to provide memory over-subscription. |
|
71 |
Xen Transcendent Memory (http://en.wikipedia.org/wiki/Transcendent_memory). |
|
72 |
tmem is a mechanism used by Xen to provide memory over-subscription. |
|
78 | 73 |
|
79 | 74 |
watcher |
80 |
:command:`ganeti-watcher` is a tool that should be run regularly |
|
81 |
from cron and takes care of restarting failed instances, restarting
|
|
82 |
secondary DRBD devices, etc. For more details, see the man page
|
|
75 |
:command:`ganeti-watcher` is a tool that should be run regularly from
|
|
76 |
cron. The tool executes tasks such as restarting failed instances and
|
|
77 |
restarting secondary DRBD devices. For more details, see the man page
|
|
83 | 78 |
:manpage:`ganeti-watcher(8)`. |
84 | 79 |
|
85 | 80 |
|
b/doc/index.rst | ||
---|---|---|
4 | 4 |
Welcome to Ganeti's documentation! |
5 | 5 |
================================== |
6 | 6 |
|
7 |
This page is the starting point for browsing the Ganeti documentation. It |
|
8 |
contains links to all the sections of Ganeti documentation, grouped by topic. |
|
7 |
This page is the starting point for browsing the Ganeti |
|
8 |
documentation. Below, the corpus of Ganeti documentation is grouped by |
|
9 |
topic. |
|
9 | 10 |
|
10 | 11 |
A few quick references: |
11 | 12 |
|
... | ... | |
16 | 17 |
Installing Ganeti |
17 | 18 |
+++++++++++++++++ |
18 | 19 |
|
19 |
There are a few resources you can use to install and/or upgrade Ganeti:
|
|
20 |
Use the following resources to install and/or upgrade Ganeti:
|
|
20 | 21 |
|
21 | 22 |
- :doc:`install`: Comprehensive instructions for installing Ganeti. |
22 | 23 |
- :doc:`install-quick`: A shortened installation guide for the experienced Ganeti user. |
23 |
- :doc:`upgrade`: Instructions for upgrading an existing installation to the latest version of Ganeti.
|
|
24 |
- :doc:`upgrade`: Instructions for upgrading an existing Ganeti installation to the latest version.
|
|
24 | 25 |
|
25 | 26 |
Using Ganeti |
26 | 27 |
++++++++++++ |
27 | 28 |
|
28 | 29 |
The following resources provide guidance on how to use Ganeti: |
29 | 30 |
|
30 |
- :doc:`admin`: Information about how to manage a Ganeti cluster after it is installed (including management of nodes, instances, and information about the tools and the monitoring agent).
|
|
31 |
- :doc:`walkthrough`: A more example-oriented guide to Ganeti.
|
|
31 |
- :doc:`admin`: Information about how to manage a Ganeti cluster after it is installed (including management of nodes and instances, and information about Ganeti's tools and monitoring agent).
|
|
32 |
- :doc:`walkthrough`: An example-oriented guide to Ganeti.
|
|
32 | 33 |
- :doc:`manpages`: Descriptions of the various tools that are part of Ganeti. |
33 | 34 |
- :doc:`security`: A description of the security model underlying a Ganeti cluster. |
34 |
- :doc:`hooks`: Information on hooking scripts, which extend Ganeti functionalities by automatically activating when certain events happen.
|
|
35 |
- :doc:`iallocator`: Description of the API for external tools which can allocate instances either manually or automatically. |
|
36 |
- :doc:`rapi`: Details the Ganeti remote API, which allows programmatic access to most of the functionalities of Ganeti.
|
|
37 |
- :doc:`ovfconverter`: Provides compatibility with the standard OVF virtual machine interchange format.
|
|
38 |
- :doc:`virtual-cluster`: Describes how to use virtual cluster support, which is utilized mainly for testing reasons.
|
|
35 |
- :doc:`hooks`: Information on hooking scripts, which extend Ganeti functionalities by automatically activating when certain events occur.
|
|
36 |
- :doc:`iallocator`: Description of the API for external tools, which can allocate instances either manually or automatically.
|
|
37 |
- :doc:`rapi`: Description of the Ganeti remote API, which allows programmatic access to most of the functionalities of Ganeti.
|
|
38 |
- :doc:`ovfconverter`: Description of a tool that provides compatibility with the standard OVF virtual machine interchange format.
|
|
39 |
- :doc:`virtual-cluster`: Explanation of how to use virtual cluster support, which is utilized mainly for testing reasons.
|
|
39 | 40 |
|
40 |
A few functionalities are explicitly targeted for big installations, in which
|
|
41 |
multiple clusters are present.
|
|
41 |
Some features are explicitly targeted for large Ganeti installations,
|
|
42 |
in which multiple clusters are present:
|
|
42 | 43 |
|
43 | 44 |
- :doc:`cluster-merge`: Describes a tool for merging two existing clusters. |
44 | 45 |
- :doc:`move-instance`: Describes how to move instances between clusters. |
... | ... | |
46 | 47 |
Developing Ganeti |
47 | 48 |
+++++++++++++++++ |
48 | 49 |
|
49 |
There are a few useful documents for developers who want to modify Ganeti: |
|
50 |
There are a few documents particularly useful for developers who want |
|
51 |
to modify Ganeti: |
|
50 | 52 |
|
51 |
- :doc:`locking`: Describes Ganeti's locking strategy and, in particular, lock order dependencies.
|
|
53 |
- :doc:`locking`: Describes Ganeti's locking strategy and lock order dependencies. |
|
52 | 54 |
- :doc:`devnotes`: Details build dependencies and other useful development-related information. |
53 | 55 |
|
54 | 56 |
|
55 | 57 |
Implemented designs |
56 | 58 |
------------------- |
57 | 59 |
|
58 |
Before actual implementation, all Ganeti features are described in a design
|
|
59 |
document. Designs fall into two categories: released versions and draft versions
|
|
60 |
(which are either incomplete or not implemented). |
|
60 |
Before actual implementation, all Ganeti features are described in a |
|
61 |
design document. Designs fall into two categories: released versions
|
|
62 |
and draft versions (which are either incomplete or not implemented).
|
|
61 | 63 |
|
62 | 64 |
.. toctree:: |
63 | 65 |
:maxdepth: 1 |
... | ... | |
124 | 126 |
upgrade.rst |
125 | 127 |
virtual-cluster.rst |
126 | 128 |
walkthrough |
129 |
|
|
130 |
.. vim: set textwidth=72 : |
Also available in: Unified diff