raboof/pekko
1
0
Fork 0
Code Issues Pull requests Projects Releases 6 Packages Wiki Activity Actions 10

=doc #17752 Compatibility with latest version of Sphinx

Browse source
This commit is contained in:
kukido 2015-07-08 18:41:38 +03:00 • committed by Martynas Mickevičius
parent 1ebab01ebb
commit f52ee58cc8
2 changed files with 19 additions and 32 deletions
Download patch file Download diff file Expand all files Collapse all files

32
akka-docs/rst/common/cluster.rst
View file

@ -1,8 +1,8 @@
.. _cluster: .. _cluster:
###################### #####################
Cluster Specification Cluster Specification
###################### #####################
.. note:: This document describes the design concepts of the clustering. .. note:: This document describes the design concepts of the clustering.
@ -49,7 +49,7 @@ UID.
The cluster membership state is a specialized `CRDT`_, which means that it has a monotonic The cluster membership state is a specialized `CRDT`_, which means that it has a monotonic
merge function. When concurrent changes occur on different nodes the updates can always be merge function. When concurrent changes occur on different nodes the updates can always be
merged and converge to the same end result. merged and converge to the same end result.
.. _CRDT: http://hal.upmc.fr/docs/00/55/55/88/PDF/techreport.pdf .. _CRDT: http://hal.upmc.fr/docs/00/55/55/88/PDF/techreport.pdf
@ -60,7 +60,7 @@ The cluster membership used in Akka is based on Amazon's `Dynamo`_ system and
particularly the approach taken in Basho's' `Riak`_ distributed database. particularly the approach taken in Basho's' `Riak`_ distributed database.
Cluster membership is communicated using a `Gossip Protocol`_, where the current Cluster membership is communicated using a `Gossip Protocol`_, where the current
state of the cluster is gossiped randomly through the cluster, with preference to state of the cluster is gossiped randomly through the cluster, with preference to
members that have not seen the latest version. members that have not seen the latest version.
.. _Gossip Protocol: http://en.wikipedia.org/wiki/Gossip_protocol .. _Gossip Protocol: http://en.wikipedia.org/wiki/Gossip_protocol
.. _Dynamo: http://www.allthingsdistributed.com/files/amazon-dynamo-sosp2007.pdf .. _Dynamo: http://www.allthingsdistributed.com/files/amazon-dynamo-sosp2007.pdf
@ -157,9 +157,9 @@ Leader
After gossip convergence a ``leader`` for the cluster can be determined. There is no After gossip convergence a ``leader`` for the cluster can be determined. There is no
``leader`` election process, the ``leader`` can always be recognised deterministically ``leader`` election process, the ``leader`` can always be recognised deterministically
by any node whenever there is gossip convergence. The leader is just a role, any node by any node whenever there is gossip convergence. The leader is just a role, any node
can be the leader and it can change between convergence rounds. can be the leader and it can change between convergence rounds.
The ``leader`` is simply the first node in sorted order that is able to take the leadership role, The ``leader`` is simply the first node in sorted order that is able to take the leadership role,
where the preferred member states for a ``leader`` are ``up`` and ``leaving`` where the preferred member states for a ``leader`` are ``up`` and ``leaving``
(see the `Membership Lifecycle`_ section below for more information about member states). (see the `Membership Lifecycle`_ section below for more information about member states).
The role of the ``leader`` is to shift members in and out of the cluster, changing The role of the ``leader`` is to shift members in and out of the cluster, changing
@ -180,9 +180,9 @@ The seed nodes are configured contact points for new nodes joining the cluster.
When a new node is started it sends a message to all seed nodes and then sends When a new node is started it sends a message to all seed nodes and then sends
a join command to the seed node that answers first. a join command to the seed node that answers first.
The seed nodes configuration value does not have any influence on the running The seed nodes configuration value does not have any influence on the running
cluster itself, it is only relevant for new nodes joining the cluster as it cluster itself, it is only relevant for new nodes joining the cluster as it
helps them to find contact points to send the join command to; a new member helps them to find contact points to send the join command to; a new member
can send this command to any current member of the cluster, not only to the seed nodes. can send this command to any current member of the cluster, not only to the seed nodes.
@ -199,14 +199,14 @@ makes use of this version to only push the actual state as needed.
Periodically, the default is every 1 second, each node chooses another random Periodically, the default is every 1 second, each node chooses another random
node to initiate a round of gossip with. If less than ½ of the nodes resides in the node to initiate a round of gossip with. If less than ½ of the nodes resides in the
seen set (have seen the new state) then the cluster gossips 3 times instead of once seen set (have seen the new state) then the cluster gossips 3 times instead of once
every second. This adjusted gossip interval is a way to speed up the convergence process every second. This adjusted gossip interval is a way to speed up the convergence process
in the early dissemination phase after a state change. in the early dissemination phase after a state change.
The choice of node to gossip with is random but it is biased to towards nodes that The choice of node to gossip with is random but it is biased to towards nodes that
might not have seen the current state version. During each round of gossip exchange when might not have seen the current state version. During each round of gossip exchange when
no convergence it uses a probability of 0.8 (configurable) to gossip to a node not no convergence it uses a probability of 0.8 (configurable) to gossip to a node not
part of the seen set, i.e. that probably has an older version of the state. Otherwise part of the seen set, i.e. that probably has an older version of the state. Otherwise
gossip to any random live node. gossip to any random live node.
This biased selection is a way to speed up the convergence process in the late dissemination This biased selection is a way to speed up the convergence process in the late dissemination
@ -215,12 +215,12 @@ phase after a state change.
For clusters larger than 400 nodes (configurable, and suggested by empirical evidence) For clusters larger than 400 nodes (configurable, and suggested by empirical evidence)
the 0.8 probability is gradually reduced to avoid overwhelming single stragglers with the 0.8 probability is gradually reduced to avoid overwhelming single stragglers with
too many concurrent gossip requests. The gossip receiver also has a mechanism to too many concurrent gossip requests. The gossip receiver also has a mechanism to
protect itself from too many simultaneous gossip messages by dropping messages that protect itself from too many simultaneous gossip messages by dropping messages that
have been enqueued in the mailbox for too long time. have been enqueued in the mailbox for too long time.
While the cluster is in a converged state the gossiper only sends a small gossip status message containing the gossip While the cluster is in a converged state the gossiper only sends a small gossip status message containing the gossip
version to the chosen node. As soon as there is a change to the cluster (meaning non-convergence) version to the chosen node. As soon as there is a change to the cluster (meaning non-convergence)
then it goes back to biased gossip again. then it goes back to biased gossip again.
The recipient of the gossip state or the gossip status can use the gossip version The recipient of the gossip state or the gossip status can use the gossip version
(vector clock) to determine whether: (vector clock) to determine whether:
@ -337,5 +337,3 @@ Failure Detection and Unreachability
to the state signaling that the cluster is unable to talk to this node, to the state signaling that the cluster is unable to talk to this node,
after being unreachable the failure detector may detect it as reachable after being unreachable the failure detector may detect it as reachable
again and thereby remove the flag again and thereby remove the flag

19
akka-docs/rst/dev/documentation.rst
View file

@ -92,19 +92,13 @@ Installing Sphinx on OS X
Install `Homebrew <https://github.com/mxcl/homebrew>`_ Install `Homebrew <https://github.com/mxcl/homebrew>`_
Install Python and pip: Install Python with Homebrew:
:: ::
brew install python brew install python
/usr/local/share/python/easy_install pip
Add the Homebrew Python path to your $PATH:
::
/usr/local/Cellar/python/2.7.5/bin
Homebrew will automatically add Python executable to your $PATH and pip is a part of the default Python installation with Homebrew.
More information in case of trouble: More information in case of trouble:
https://github.com/mxcl/homebrew/wiki/Homebrew-and-Python https://github.com/mxcl/homebrew/wiki/Homebrew-and-Python
@ -115,12 +109,6 @@ Install sphinx:
pip install sphinx pip install sphinx
Add sphinx_build to your $PATH:
::
/usr/local/share/python
Install BasicTeX package from: Install BasicTeX package from:
http://www.tug.org/mactex/morepackages.html http://www.tug.org/mactex/morepackages.html
@ -128,7 +116,8 @@ Add texlive bin to $PATH:
:: ::
/usr/local/texlive/2013basic/bin/universal-darwin export TEXLIVE_PATH=/usr/local/texlive/2015basic/bin/universal-darwin
export PATH=$TEXLIVE_PATH:$PATH
Add missing tex packages: Add missing tex packages: